- Email: [email protected]
Elevated urinary d -lactate levels in patients with diabetes and microalbuminuria
G Model
ARTICLE IN PRESS
PBA-10132; No. of Pages 6
Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria Chu-Kuang Chou a,e,1 , Ya-Ting Lee b,1 , Shih-Ming Chen c , Chi-Wen Hsieh c,d , Tzu-Chuan Huang c , Yi-Chieh Li c , Jen-Ai Lee c,∗ a
Department of Internal Medicine, Chia-Yi Christian Hospital, No. 539, Jhongsiao Rd., Chia-Yi 60002, Taiwan Division of Nephrology, Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung 80708, Taiwan c School of Pharmacy, College of Pharmacy, Taipei Medical University, No. 250, Wuxing St., Taipei 11031, Taiwan d Food and Drug Administration, Ministry of Health and Welfare, No. 161-2, Kunyang St, Nangang District, Taipei 11561, Taiwan e Department of Internal Medicine, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei 10002, Taiwan b
a r t i c l e
i n f o
Article history: Received 13 January 2015 Received in revised form 27 May 2015 Accepted 10 June 2015 Available online xxx Keywords: d-lactate Microalbumin Fluorescence derivatization NBD-PZ Diabetes
a b s t r a c t Diabetic nephropathy (DN) has become the major cause of end-stage renal disease. Early detection of disease risk, to enable intervention before advanced renal damage occurs, is an important goal. Microalbuminuria has been used to monitor renal damage in clinical settings for years. In this study, we divided patients with diabetes into different groups based on their microalbumin values to elucidate the relationship between urinary d-lactate and corresponding microalbumin values. Group DM1 comprised of patients with an albumin-to-creatinine ratio (ACR) of less than 30 g albumin/mg creatinine (normal range); Group DM2 comprised of patients with an ACR of 30–299 g albumin/mg creatinine (microalbuminuria); and Group DM3 comprised of patients with an ACR of ≥300 g albumin/mg creatinine (macroalbuminuria). The urinary d-lactate concentration of patients with diabetes was determined by pre-column fluorescence derivatization with 4-nitro-7-piperazino-2,1,3-benzoxadiazole (NBD-PZ), and the accuracy (recovery) and precision (relative standard deviation; RSD) were validated. The measured values showed an accuracy that was in the acceptable range (91.59–112.96%), with an RSD in the range of 3.13–13.21%. The urinary d-lactate levels of the 3 diabetic groups (groups DM1, DM2, and DM3) were significantly higher than those of healthy subjects (78.31 ± 22.13, 92.47 ± 21.98, and 47.29 ± 17.51 vs. 6.28 ± 2.39 nmol/mg creatinine, respectively; p < 0.05), with urinary d-lactate levels in the DM2 group being the highest. This modified fluorescence-based, high-performance liquid chromatography method to quantify d-lactate concentrations in the urine of patients with diabetes was established. Also, measuring the new risk marker identified in this study (d-lactate) in combination with microalbumin may facilitate the prevention of DN. © 2015 Elsevier B.V. All rights reserved.
1. Introduction d-Lactate is a lactic acid enantiomer. d-Lactate and l-lactate, both of which are naturally present in the human body, share similar physical and chemical properties, but exhibit different optical properties [1]. However, the levels of d-lactate in human tissues are only ∼1% of that of l-lactate [2]. l-Lactate is produced during carbohydrate metabolism under aerobic conditions, or during the reduction of pyruvic acid under anaerobic conditions [3]. How-
∗ Corresponding author at: School of Pharmacy, College of Pharmacy, Taipei Medical University No. 250, Wuxing St. Taipei 11031, Taiwan, ROC. Fax: +886 2 2736 1661. E-mail address: [email protected] (J.-A. Lee). 1 These authors contributed equally to the work.
ever, the origin of d-lactate is different from that of l-lactate. For example, d-lactate can be obtained from endogenous (e.g., methylglyoxal metabolism) as well as exogenous sources (e.g., bacterial fermentation in the gastrointestinal tract, foods, and drugs) [4,5]. The in vivo concentration of d-lactate in mammals tends to increase under certain disease conditions. For instance, the plasma concentration of d-lactate is significantly higher in patients with diabetes than in healthy subjects [4,6,7,8]. Diseases including brain lesions [9], appendicitis [10], short bowel syndrome [11,12], infections [13], ischemia [14], and sepsis [15] and kidney disease [16] also trigger an increase in d-lactate concentrations. These observations suggest that an abnormal elevation in the plasma d-lactate levels is closely associated with pathological states. For instance, patients with diabetes also show elevated serum and urinary concentrations of d-lactate [17]. However, the relationship between the urinary d-lactate concentration and renal function in patients
http://dx.doi.org/10.1016/j.jpba.2015.06.014 0731-7085/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model
ARTICLE IN PRESS
PBA-10132; No. of Pages 6
C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
2
with diabetes has not been fully elucidated. Globally, diabetes is one of the most common causes of end-stage renal disease (ESRD), accounting for approximately one-third of all incidences [18,19]. Patients with ESRD need to receive late hemodialysis or kidney transplantation, which require a large commitment of medical resources. Therefore, it is important to further explore the relationship between renal function and urinary d-lactate concentrations in patients with diabetes. The methods developed for analyzing d-lactate concentrations range from the less sensitive method of early-stage nonenantiomer chromatography, which dates back to 1927 [20,21], to the more recent enzyme assay designed in 1980s [22,23]. These methods require the use of d-lactate dehydrogenase (EC 1.1.1.28.) to transform d-lactate to pyruvate in the presence of the coenzyme nicotinamide adenine dinucleotide (NAD+ ). The amount of NADH generated from NAD+ can be determined by ultraviolet spectrophotometry at a wavelength of 340 nm. The biggest drawback of this method is that d-lactate dehydrogenase may cross-react with many endogenous substances, including fructose 1,6-bisphosphate, 3-phosphoglyceric acid, pyruvic acid, l-lactate, and S-lactoyl-glutathione [23,24], leading to poor accuracy and low reproducibility of experimental results. Therefore, in this study, we adapted a previously established column-switching fluorescence, high-performance liquid chromatography (HPLC) method to analyze d-lactate [25]. To date, this method has only been used in rats and mice. In this study, it was used to detect human urinary dlactate levels for the first time; thus, corresponding validation tests were also performed. In this study, a fluorescence-based HPLC method was employed to quantify the urinary concentration of d-lactate in patients with diabetes. Patients were divided into groups based on renal function indicator and microalbumin values to clarify the relationship between urinary d-lactate and the corresponding microalbumin values. 2. Materials and methods 2.1. Study population Human urine samples were collected from outpatients visiting Kaohsiung Medical University (Kaohsiung, Taiwan). A total of 12 healthy individuals and 32 patients with diabetes were enrolled in this study. The healthy individuals included 7 males and 5 females, with an average age of 36 ± 17 years. Patients with diabetes included 16 males and 16 females, with an average age of 60 ± 14 years. The study protocol (KMUH-IRB-940177) was approved by the institutional review board. As per the recommendations of the American Diabetes Association (2003), patients with diabetes were divided into 3 groups (as shown in Table 1) based on the albumin-to-creatinine ratio (ACR) of their urine samples. Group DM1 comprised 15 patients with an ACR of less than 30 g albumin/mg creatinine (normal Table 1 The albumin- to- creatinine ratio (ACR) was tested in both healthy and patients with diabetes and then the subjects were divided into groups according to their ACR values. Group Healthy subjects DM1 DM2 DM3
Number 12 15 13 4
Urinary microalbumin 10.37 11.02 118.96 349.38
± ± ± ±
0.56 2.65 22.60 15.72
All data shown as mean ± SE. DM1 group: Albumin-creatinine-ratio < 30 g albumin/mg creatinine. DM2 group: Albumin-creatinine-ratio 30–299 g albumin/mg creatinine. DM3 group: Albumin-creatinine-ratio ≥300 g albumin/mg creatinine.
range) and a mean ACR of 11.02 ± 2.65 g albumin/mg creatinine. Group DM2 comprised 13 patients with an ACR of 30–299 g albumin/mg creatinine and a mean ACR of 118.96 ± 22.60 g albumin/mg creatinine. The last group, group DM3, included 4 patients with an ACR of >300 g albumin/mg creatinine and a mean ACR of 349.38 ± 15.72 g albumin/mg creatinine. 2.2. Chemicals Lithium d- and l-lactate, creatinine, cimetidine, sodium lauryl sulfate, glycine, methylumbelliferone, and bovine serum albumin were purchased from Sigma Chemical (St Louis, MO, USA). NBD-PZ, triphenylphosphine, and 2,2 -dipyridyl disulfide were purchased from Tokyo Kasei Chemicals (Tokyo, Japan). Trifluoroacetic acid, sodium dihydrogen phosphate, and o-phosphoric acid were purchased from Riedel-de Haën (Seelze, Germany). Propionic acid, sodium hydroxide, and hydrochloric acid were acquired from Nacalai Tesque Co., Ltd. (Kyoto, Japan). HPLC-grade acetonitrile (ACN) and methanol (MeOH) were purchased from Merck (Darmstadt, Germany). 2.3. Analysis of d-lactate in human urine The method for determining d-lactate levels in human urine samples was modified from a previously published, columnswitching HPLC method [26,27]. Briefly, 20 L of urine was added to 10 L of 1 mM propionic acid (used as the internal standard) and mixed with 170 L of ACN. For derivatization, the solution was centrifuged at 700 × g at 4 ◦ C for 10 min. Then, 100 L of the supernatant was added to 100 L of 8 mM NBD-PZ in ACN in the presence of 25 L of 280 mM triphenylphosphine and 25 L of 280 mM 2,2 dipyridyl disulfide. After allowing the solution to stand at 30 ◦ C for 3 h, 250 L of aqueous 0.1% trifluoroacetic acid solution was added to terminate the reaction. To remove the excess NBD-PZ, 100 L of the resulting solution was loaded twice onto a mobile phase-preconditioned, solid-phase extraction cartridge (EmporeTM SBD-RPS; 4 mm/mL; 3 M, St. Paul, MN, USA), and the eluates were subsequently combined. Twenty microliters of this combined eluate was injected into the HPLC column. A TSKgel® octadecylsilyl (ODS)-80Ts column (250 mm × 4.6 mm i.d., 5 m; Tosoh Co.; Tokyo, Japan) was used to isolate and quantify (d + l)-lactate derivatives in the samples. The mobile phase was ACN-MeOH-H2 O (12:20:68, v/v/v), and the flow rate was 0.7 mL/min. A Chiralpak® AD-RH column (150 × 4.6 mm i.d., 5 m; Daicel Co., Osaka, Japan) was used to separate the d- and l-lactates. The mobile phase was ACN-H2 O (60:40, v/v) and the isocratic flow rate was 0.3 mL/min. Fluorescence was detected at 547 nm following excitation at 491 nm (L-2385 FL detector; Hitachi; Tokyo, Japan). Lactate enantiomers were quantitatively determined from the peak areas of the chromatograms (D-7500 integrator; Hitachi; Tokyo, Japan). The urinary d-lactate concentration was expressed as nanomoles per milligram of creatinine. 2.4. Calibration curves and validation study To generate calibration curves, 20 L of lactate (D: L = 1: 1) solutions at concentrations of 0.05, 0.10, 0.20, 0.30, 0.40, or 0.50 mM were added to 10 L of 1.00 mM propionic acid (internal standard) and diluted to 170 L with ACN. Derivatization procedures were performed in a manner similar to that described above (n = 5). The calibration curve was plotted from the peak area ratio of the lactate to the internal standard, as a function of the lactate concentration. To estimate the precision and accuracy of the present analytical method, lactate determinations in human urine samples were performed by adding different concentrations of d-lactate (0, 3.91, 15.63, and 31.25 M) to 20 L of human urine in the presence of
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model
ARTICLE IN PRESS
PBA-10132; No. of Pages 6
C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
3
Fig. 1. Representative chromatograms obtained using the proposed column-switching HPLC system with NBD-PZ derivatization. (A) Separation of the D, l-lactate standard. (B) Separation of the D, l-lactate standard on a chiral column. The propionate derivative was used as an internal standard (I.S.).
10 L of 1.0 mM propionic acid (internal standard). Each sample was analyzed using the method described above (n = 5). The precision was represented by the relative standard deviation (RSD), and the accuracy was expressed in terms of the calculated recovery. 2.5. Determination of urinary creatinine levels Urinary creatinine levels were measured as described previously [28]. Briefly, cimetidine was added to urine samples as the internal standard, and the samples were separated on a Shiseido CAPCELL PAK C18 column (250 mm × 4.6 mm i.d.; particle size, 5 m; Shiseido; Tokyo, Japan). The mobile phase comprised ACNsodium lauryl sulfate (30 mM)- potassium dihydrogen phosphate (pH 3.0; 100 mM)(36:30:30, v/v/v). The absorbance was monitored at 234 nm. 2.6. Urinary microalbumin assays Urinary microalbumin was analyzed using a kit (Good Biotech Corp.; Taichung, Taiwan) according to the manufacturer’s instructions. Briefly, urine samples were conditioned with Tris buffer prior to their incubation with an anti-human albumin antibody for 10 min. These samples were then analyzed spectrophotometrically at 405 nm. To compensate for variations in the overall concentration of the spot-checked urine samples, it was helpful to normalize the amount of albumin (g/dL) in the sample with the corresponding creatinine concentration (mg/dL). 2.7. Statistical analysis Differences were determined using 1-way analysis of variance (ANOVA), and the levels of significant differences between treatment groups were determined using Scheffe’s multiple range test. Differences with p values <0.05 were considered statistically significant. 3. Results 3.1. d-Lactate standard determinations The chromatograms of the derivatives of the (D + L)-lactate standard (D: L = 1: 1) and the separated d-lactate are shown in Fig. 1. The (D + L)-lactate derivative was eluted from the ODS column at a retention time of 28.4 min (Fig. 1A). With regard to the determination of d-lactate and l-lactate, the 6-port valve in the HPLC system was switched after the appearance of the peak fraction derived from the (D + L)-lactate derivative to trap it in the loop. Then, the trapped (D + L)-lactate derivative was injected directly into the chi-
Table 2 Precision and accuracy of urinary d-lactate measurements (n = 5). d-Lactate added (M)
Measured (M) RSD (%) Recovery (%)
0
3.91
15.63
31.25
4.99 13.21 –
9.40 5.16 112.96
19.30 3.13 91.59
38.43 8.01 107.01
ral column by switching the valve back to the former position. The d-lactate derivative was eluted at 13.3 min and l-lactate derivative was eluted at 16.2 min (Fig. 1B). As per the results from the ODS column, a calibration curve was plotted from the peak area ratios of lactate to the internal standard, as a function of the lactate concentration. The calibration curve of (D+L)-lactate was generated using standards ranging from 31.25 to 2000 M, resulting in a line equation (after linear regression analysis) of Y = 0.0057X + 0.0285 (Y: the peak ratio of (D + L)-lactate to propionic acid (internal standard), X: the concentration of (D + L)-lactate). We observed good linearity with a correlation coefficient (r2) 0.9987. 3.2. Analysis of urinary d-lactate in healthy subjects and patients with diabetes As depicted in Fig. 2, the d- and l-lactate derivatives in human urine were separated as described above and successfully analyzed. The urinary lactate derived from healthy subjects and patients with diabetes was eluted from the ODS column (Fig. 2A and C, respectively). The urinary d-lactate derivative was separated with a chiral column, and data for healthy subjects and patients with diabetes are shown in Fig. 2B and D, respectively. Using the peak area proportion (P) of the d-lactate obtained from chiral column and the total (T) lactate (D + L) concentration obtained from the calibration curve of the ODS column, the d-lactate concentration (C) was quantitated using the equation C = P × T [29]. The precision (RSD) and accuracy (recovery) of the d-lactate determinations in human urine are shown in Table 2. The measured values showed an accuracy of 91.59–112.96%, and a precision of 3.13–13.21% (RSD). These results helped us to validate the present method. 3.3. Comparison of urinary d-lactate values between subjects with different albuminuria levels The albumin-to-creatinine ratio (ACR) was tested in both healthy subjects and patients with diabetes. The subjects were then divided into groups according to their ACR values (Table 1). The urinary d-lactate levels in healthy subjects and patients with diabetes are shown in Fig. 3. Patients with diabetes patients were
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model PBA-10132; No. of Pages 6 4
ARTICLE IN PRESS C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
Fig. 3. Patients with diabetes urinary d-lactate levels changed according to the albumin-to-creatinine ratio (ACR) in their urine. Group DM1 comprised patients with an ACR of less than 30 g albumin/mg creatinine (normal range); Group DM2 comprised patients with an ACR of 30–299 g albumin/mg creatinine (microalbuminuria); and Group DM3 comprised patients with an ACR of ≥300 g albumin/mg creatinine (macroalbuminuria). *p < 0.05 compared to healthy subjects.
divided into 3 groups according to their urinary ACR values (as described in Section 2.1). The urinary d-lactate levels of groups DM1–DM3 were significantly higher than those of the healthy subjects (78.31 ± 22.13, 92.47 ± 21.98, and 47.29 ± 17.51, respectively, vs. 6.28 ± 2.39 nmol/mg creatinine; p < 0.05). Among the 3 diabetic groups, the urinary d-lactate level of the DM2 group was highest. 4. Discussion
Fig. 2. Representative chromatograms of urine samples from (A) healthy subjects and (C) patients with diabetes, derivatized with NBD-PZ and separated on an ODS column. Separation of D, l-lactate in urine samples from (B) healthy subjects and (D) patients with diabetes on a chiral column. The propionate derivative was used as an internal standard (I.S.).
This was the first study to analyze the human urinary d-lactate levels using a previously developed fluorescence-based HPLC analysis system. After adding three different known concentrations of d-lactate to human urine, followed by quantitative analysis, we obtained a recovery rate in the range 91.59–112.96%, with an RSD of 3.13–13.21%. These results met the standards of the validation protocol. The recovery rate ranged between 85 and 115%, and the RSD was less than 15%. Thus, the method was validated for use in the analysis of human urinary d-lactate levels. Because urinary microalbumin can be readily measured with high sensitivity, it is widely used as a clinical indicator of renal function. Based on the standards established by the American Diabetes Association in 2003, we divided the patients with diabetes into 3 groups (DM1–DM3), according to their urinary ACRs, which can be used to estimate normal microalbumin levels and diagnose microalbuminuria and other proteinuria conditions. Diabetic nephropathy (DN) is one of the most common complications of diabetes and poses a high risk of ESRD development, if poorly controlled. Patients diagnosed with diabetes need to monitor their renal functions regularly, in consultation with their physician. When renal function enters the microalbuminuria stage, it is recommended that patients receive appropriate kidney treatments. Once the microalbumin levels exceed 299 g/mg creatinine, the kidneys develop irreversible damage and renal treatment is futile after such damage. Therefore, monitoring renal functions in patients with diabetes is critical and should be conducted in a continuous and intensive fashion. However, microalbuminuria is not only an indicator for predicting diabetic complications. It can also serve as a powerful independent risk factor for predicting cardiovascular diseases [30,31], a marker of uric acid metabolic disorders [32], and a prognostic indicator of ischemic stroke. Microalbuminuria is an early indicator of vascular endothelial cell dysfunction, as revealed by associated kidney dysfunctions. Furthermore, a significant proportion of patients with diabetes and stable normoalbuminuria have well-established DN lesions, ∼40% of which are at risk for progres-
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model PBA-10132; No. of Pages 6
ARTICLE IN PRESS C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
sion to proteinuria. These findings indicate that urinary albumin levels alone cannot be used as a specific biomarker for the detection of renal dysfunction. Thus, investigations into new risk markers or into the combined use of several currently available markers may lead to important advances in this field. d-Lactate is regarded as a stable end product of methylglyoxal [33], a highly reactive dicarbonyl compound that can cause carbonyl stress and protein malfunctions. Methylglyoxal levels are increased in the kidneys by diabetes, gentamicin, and aristolochic acid-induced nephropathy [34–36]. These findings imply that urinary d-lactate could be a marker of renal damage because it reflects toxic methylglyoxal levels in the kidney. Urinary d-lactate values varied significantly between different groups in this study. The urinary d-lactate concentration of patients in group DM1 was 12.5 times greater than that observed in healthy subjects; however, the microalbumin values of this group were still within the normal range. Patients in group DM2 were classified as having a microalbuminuria condition, and their urinary d-lactate levels were the highest among all groups studied. The kidney damages observed in patients belonging to group DM3 were considered clinically irreversible and their urinary d-lactate levels were lower than that in the patients from group DM2, but higher than that in the healthy subjects. The results suggested that in patients with diabetes whose microalbumin values were still normal, the urinary d-lactate levels were already elevated to a significant extent. However, the urinary d-lactate levels showed a decrease in patients with severe proteinuria conditions. The changes observed in urinary d-lactate levels can be explained as follows: Albuminuria is widely accepted as an early marker of DN and is observed before the occurrence of elevated serum creatinine levels and decreased glomerular filtration rates. Regarding the pathophysiological aspects of albuminuria, abnormal leakage of albumin into the urine may be caused due to increased permeability of the glomerular capillary walls, and subsequently impaired reabsorption by epithelial cells of the proximal tubules during the course of diabetes. With the progression of diabetes, increased levels of urinary albumin have been reported [37]. Early detectable markers are required before the loss of proteinfiltration function. The results of this study show that d-lactate levels are significant higher in subjects with diabetic normalalbuminuria, micro-albuminuria, and macro-albuminuria, than the levels in the healthy subjects. d-Lactate is the end product of methylglyoxal and elevated production of d-lactate may represent methylglyoxal accumulation and carbonyl stress, which commonly occur during diabetes [38]. Thus, in contrast to renal-dysfunction markers such as micro-albuminuria, d-lactate can potentially serve as an early marker because it reflects carbonyl stress that can lead to renal damage. When damage occurs to renal cells, the cells begin to consume energy that is required for repair processes. Methylglyoxal may come from the glycosis, also the fatty acid and amino acid decomposition. It accumulates during the process of energy production and can be metabolized into d-lactate, which was detected. Thus, we suggest that when lesions begin to form, the injured tissues are still capable of self-reparation. Self-reparation requires substantial energy consumption, leading to elevated production of d-lactate, which may explain the highly increased d-lactate levels observed in the DM1 and DM2 groups. In this two stage, d-lactates rise in parallel with urine microalbumin levels; thus, individuals in the DM2 group had the highest d-lactate levels. After the microalbumin level reaches more than 300 g/mg creatinine, the injured tissue begins to undergo fibrosis, and the ability for self-reparation decreases and energy utilization drops. As a consequence, d-lactate production in reduced, as observed in the DM3 group. The early control of associated risk factors, such as high blood pressure and high cholesterol, along with the adminis-
5
tration of microalbuminuria-reducing drugs, can help prevent further damage to the target organs. d-Lactate appears earlier than microalbumin in patients with diabetes. We expect that the combined measurement of urinary d-lactate and microalbumin levels can be helpful for the early identification of patients who are required to undergo therapy for diabetic nephrology. 5. Conclusion The level of urinary d-lactate is intimately related to DN. We expect urinary d-lactate can be used as an early renal damage marker; however, further experiments should be performed to investigate this topic in greater detail. Conflict of interest statement The authors have no conflicts of interest to declare. Acknowledgements This work was financially supported by the National Science Council of the Republic of China (NSC 97-2320-B-038-007-MY3). References [1] M.R. Wright, F. Jamali, Methods for the analysis of enantiomers of racemic drugs application to pharmacological and pharmacokinetic studies, J. Pharmacol. Toxicol. Methods 29 (1993) 1–9. [2] E. Racker, The mechanism of action of glyoxalase, J. Biol. Chem. 190 (1951) 685–696. [3] M. Adeva-Andany, M. Lopez-Ojen, R. Funcasta-Calderon, E. Ameneiros-Rodriguez, C. Donapetry-Garcia, M. Vila-Altesor, J. Rodriguez-Seijas, Comprehensive review on lactate metabolism in human health, Mitochondrion 17 (2014) 76–100. [4] P.J. Beisswenger, S.K. Howell, A.D. Touchette, S. Lal, B.S. Szwergold, Metformin reduces systemic methylglyoxal levels in type 2 diabetes, Diabetes 48 (1999) 198–202. [5] P.J. Thornalley, N.I. Hooper, P.E. Jennings, C.M. Florkowski, A.F. Jones, J. Lunec, A.H. Barnett, The human red blood cell glyoxalase system in diabetes mellitus, Diabetes Res. Clin. Pract. 7 (1989) 115–120. [6] A.C. McLellan, S.A. Phillips, P.J. Thornalley, The assay of methylglyoxal in biological systems by derivatization with 1,2-diamino-4,5-dimethoxybenzene, Anal. Biochem. 206 (1992) 17–23. [7] A.C. McLellan, S.A. Phillips, P.J. Thornalley, Fluorimetric assay of d-lactate, Anal. Biochem. 206 (1992) 12–16. [8] M.M. Christopher, J.D. Broussard, C.W. Fallin, N.J. Drost, M.E. Peterson, Increased serum D-lactate associated with diabetic ketoacidosis, Metabolism 44 (1995) 287–290. [9] A.D. Hingorani, N.N. Chan, d-Lactate encephalopathy, Lancet 358 (2001) 1814. [10] F. Caglayan, M. Cakmak, O. Caglayan, T. Cavusoglu, Plasma d-lactate levels in diagnosis of appendicitis, J. Invest. Surg. 16 (2003) 233–237. [11] A.J. Mayne, D.J. Handy, M.A. Preece, R.H. George, I.W. Booth, Dietary management of D-lactic acidosis in short bowel syndrome, Arch. Dis. Child. 65 (1990) 229–231. [12] G. Bongaerts, J. Tolboom, T. Naber, J. Bakkeren, R. Severijnen, H. Willems, D-Lactic acidemia and aciduria in pediatric and adult patients with short bowel syndrome, Clin. Chem. 41 (1995) 107–110. [13] S.M. Smith, R.H. Eng, J.M. Campos, H. Chmel, D-Lactic acid measurements in the diagnosis of bacterial infections, J. Clin. Microbiol. 27 (1989) 385–388. [14] A. Assadian, O. Assadian, C. Senekowitsch, R. Rotter, S. Bahrami, W. Furst, W. Jaksch, G.W. Hagmuller, W. Hubl, Plasma d-Lactate as a potential early marker for colon ischaemia after open aortic reconstruction, Eur. J. Vasc. Endovasc. Surg. 31 (2006) 470–474. [15] Y. Sun, L.X. Wang, Y.F. Zhou, S.G. Sun, X.F. Mao, X.D. Deng, P. Zhang, Effects of glutamine combined with ulinastatin on inflammatory response of patients with severe burn injury, Zhonghua Shao Shang Za Zhi. 29 (2013) 349–354. [16] T.C. Huang, S.M. Chen, Y.C. Li, J.A. Lee, Urinary d-lactate levels reflect renal function in aristolochic acid-induced nephropathy in mice, Biomed. Chromatogr. 27 (2013) 1100–1106. [17] J.P. Talasniemi, S. Pennanen, H. Savolainen, L. Niskanen, J. Liesivuori, Analytical investigation: assay of d-lactate in diabetic plasma and urine, Clin. Biochem. 41 (2008) 1099–1103. [18] M.A. Buyukbese, Complications of diabetes: chronic kidney disease (CKD) and diabetic nephropathy, J. Periodontol. 15 (2014) 2787. [19] G. Pugliese, Updating the natural history of diabetic nephropathy, Acta. Diabetol. 51 (2014) 905–915. [20] T.E. Friedemann, A.I. Kendall, The determination of lactic acid, J. Biol. Chem. 82 (1929) 23–43.
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model PBA-10132; No. of Pages 6 6
ARTICLE IN PRESS C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
[21] S.B. Barker, W.H. Summerson, The colorimetric determination of lactic acid in biological material, J. Biol. Chem. 138 (1941) 535–554. [22] S. Ohmori, T. Iwamoto, Sensitive determination of D-lactic acid in biological samples by high-performance liquid chromatography, J. Chromatogr. 431 (1988) 239–247. [23] R.B. Brandt, S.A. Siegel, M.G. Waters, M.H. Bloch, Spectrophotometric assay for d-(−)-lactate in plasma, Anal. Biochem. 102 (1980) 39–46. [24] R. Marti, E. Varela, R.M. Segura, J. Alegre, J.M. Surinach, C. Pascual, Determination of d-lactate by enzymatic methods in biological fluids: study of interferences, Clin. Chem. 43 (1997) 1010–1015. [25] J.-A. Lee, Y.-C. Tsai, H.-Y. Chen, C.-C. Wang, S.-M. Chen, T. Fukushima, K. Imai, Fluorimetric determination of d-lactate in urine of normal and diabetic rats by column-switching high-performance liquid chromatography, Anal. Chim. Acta 534 (2005) 185–191. [26] M.H. Lin, H.Y. Chen, T.H. Liao, T.C. Huang, C.M. Chen, J.A. Lee, Determination of time-dependent accumulation of d-lactate in the streptozotocin-induced diabetic rat kidney by column-switching HPLC with fluorescence detection, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 879 (2011) 3214–3219. [27] H. Hasegawa, T. Fukushima, J.A. Lee, K. Tsukamoto, K. Moriya, Y. Ono, K. Imai, Determination of serum d-lactic and l-lactic acids in normal subjects and diabetic patients by column-switching HPLC with pre-column fluorescence derivatization, Anal. Bioanal. Chem. 377 (2003) 886–891. [28] R. Marsilio, R. Dall’Amico, G. Giordano, L. Murer, G. Montini, M. Ros, L. Bacelle, M. Plebani, N. Dussini, G. Zacchello, Rapid determination of creatinine in serum and urine by ion-pair high-performance liquid chromatography, Int. J. Clin. Lab. Res. 29 (1999) 103–109. [29] H. Ichihara, T. Fukushima, K. Imai, Enantiomeric determination of d- and l-lactate in rat serum using high-performance liquid chromatography with a cellulose-type chiral stationary phase and fluorescence detection, Anal. Biochem. 269 (1999) 379–385.
[30] C.D. Stehouwer, Y.M. Smulders, Microalbuminuria and risk for cardiovascular disease: analysis of potential mechanisms, J. Am. Soc. Nephrol. 17 (2006) 2106–2111. [31] D. de Zeeuw, H.H. Parving, R.H. Henning, Microalbuminuria as an early marker for cardiovascular disease, J. Am. Soc. Nephrol. 17 (2006) 2100–2105. [32] T. Stakhova, A.V. Shcherbak, L.V. Kozlovskaia, M.V. Taranova, I.M. Balkarov, Clinical value of the determination of markers for endothelial dysfunction (endothelin-1, microalbuminuria) and tubulointerstitial tissue lesion (beta2-microglobulin, monocyte chemotactic protein-1) in hypertensive patients with uric acid metabolic disorders, Ter. Arkh. 86 (2014) 45–51. [33] C. Masterjohn, E. Mah, Y. Park, R. Pei, J. Lee, J.E. Manautou, R.S. Bruno, Acute glutathione depletion induces hepatic methylglyoxal accumulation by impairing its detoxification to d-lactate, Exp. Biol. Med. (Maywood) 238 (2013) 360–369. [34] Y.C. Li, Y.M. Shih, J.A. Lee, Gentamicin caused renal injury deeply related to methylglyoxal and N-epsilon-(carboxyethyl) lysine (CEL), Toxicol. Lett. 219 (2013) 85–92. [35] Y.C. Li, S.H. Tsai, S.M. Chen, Y.M. Chang, T.C. Huang, Y.P. Huang, C.T. Chang, J.A. Lee, Aristolochic acid-induced accumulation of methylglyoxal and N-epsilon-(carboxymethyl) lysine: an important and novel pathway in the pathogenic mechanism for aristolochic acid nephropathy, Biochem. Biophys. Res. Commun. 423 (2012) 832–837. [36] N. Rabbani, P.J. Thornalley, The critical role of methylglyoxal and glyoxalase 1 in diabetic nephropathy, Diabetes 63 (2014) 50–52. [37] G. D’Amico, C. Bazzi, Pathophysiology of proteinuria, Kidney Int. 63 (2003) 809–825. [38] D.L. Vander Jagt, Methylglyoxal, diabetes mellitus and diabetic complications, Drug Metabol. Drug Interact. 23 (2008) 93–124.
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
ARTICLE IN PRESS
PBA-10132; No. of Pages 6
Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria Chu-Kuang Chou a,e,1 , Ya-Ting Lee b,1 , Shih-Ming Chen c , Chi-Wen Hsieh c,d , Tzu-Chuan Huang c , Yi-Chieh Li c , Jen-Ai Lee c,∗ a
Department of Internal Medicine, Chia-Yi Christian Hospital, No. 539, Jhongsiao Rd., Chia-Yi 60002, Taiwan Division of Nephrology, Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung 80708, Taiwan c School of Pharmacy, College of Pharmacy, Taipei Medical University, No. 250, Wuxing St., Taipei 11031, Taiwan d Food and Drug Administration, Ministry of Health and Welfare, No. 161-2, Kunyang St, Nangang District, Taipei 11561, Taiwan e Department of Internal Medicine, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei 10002, Taiwan b
a r t i c l e
i n f o
Article history: Received 13 January 2015 Received in revised form 27 May 2015 Accepted 10 June 2015 Available online xxx Keywords: d-lactate Microalbumin Fluorescence derivatization NBD-PZ Diabetes
a b s t r a c t Diabetic nephropathy (DN) has become the major cause of end-stage renal disease. Early detection of disease risk, to enable intervention before advanced renal damage occurs, is an important goal. Microalbuminuria has been used to monitor renal damage in clinical settings for years. In this study, we divided patients with diabetes into different groups based on their microalbumin values to elucidate the relationship between urinary d-lactate and corresponding microalbumin values. Group DM1 comprised of patients with an albumin-to-creatinine ratio (ACR) of less than 30 g albumin/mg creatinine (normal range); Group DM2 comprised of patients with an ACR of 30–299 g albumin/mg creatinine (microalbuminuria); and Group DM3 comprised of patients with an ACR of ≥300 g albumin/mg creatinine (macroalbuminuria). The urinary d-lactate concentration of patients with diabetes was determined by pre-column fluorescence derivatization with 4-nitro-7-piperazino-2,1,3-benzoxadiazole (NBD-PZ), and the accuracy (recovery) and precision (relative standard deviation; RSD) were validated. The measured values showed an accuracy that was in the acceptable range (91.59–112.96%), with an RSD in the range of 3.13–13.21%. The urinary d-lactate levels of the 3 diabetic groups (groups DM1, DM2, and DM3) were significantly higher than those of healthy subjects (78.31 ± 22.13, 92.47 ± 21.98, and 47.29 ± 17.51 vs. 6.28 ± 2.39 nmol/mg creatinine, respectively; p < 0.05), with urinary d-lactate levels in the DM2 group being the highest. This modified fluorescence-based, high-performance liquid chromatography method to quantify d-lactate concentrations in the urine of patients with diabetes was established. Also, measuring the new risk marker identified in this study (d-lactate) in combination with microalbumin may facilitate the prevention of DN. © 2015 Elsevier B.V. All rights reserved.
1. Introduction d-Lactate is a lactic acid enantiomer. d-Lactate and l-lactate, both of which are naturally present in the human body, share similar physical and chemical properties, but exhibit different optical properties [1]. However, the levels of d-lactate in human tissues are only ∼1% of that of l-lactate [2]. l-Lactate is produced during carbohydrate metabolism under aerobic conditions, or during the reduction of pyruvic acid under anaerobic conditions [3]. How-
∗ Corresponding author at: School of Pharmacy, College of Pharmacy, Taipei Medical University No. 250, Wuxing St. Taipei 11031, Taiwan, ROC. Fax: +886 2 2736 1661. E-mail address: [email protected] (J.-A. Lee). 1 These authors contributed equally to the work.
ever, the origin of d-lactate is different from that of l-lactate. For example, d-lactate can be obtained from endogenous (e.g., methylglyoxal metabolism) as well as exogenous sources (e.g., bacterial fermentation in the gastrointestinal tract, foods, and drugs) [4,5]. The in vivo concentration of d-lactate in mammals tends to increase under certain disease conditions. For instance, the plasma concentration of d-lactate is significantly higher in patients with diabetes than in healthy subjects [4,6,7,8]. Diseases including brain lesions [9], appendicitis [10], short bowel syndrome [11,12], infections [13], ischemia [14], and sepsis [15] and kidney disease [16] also trigger an increase in d-lactate concentrations. These observations suggest that an abnormal elevation in the plasma d-lactate levels is closely associated with pathological states. For instance, patients with diabetes also show elevated serum and urinary concentrations of d-lactate [17]. However, the relationship between the urinary d-lactate concentration and renal function in patients
http://dx.doi.org/10.1016/j.jpba.2015.06.014 0731-7085/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model
ARTICLE IN PRESS
PBA-10132; No. of Pages 6
C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
2
with diabetes has not been fully elucidated. Globally, diabetes is one of the most common causes of end-stage renal disease (ESRD), accounting for approximately one-third of all incidences [18,19]. Patients with ESRD need to receive late hemodialysis or kidney transplantation, which require a large commitment of medical resources. Therefore, it is important to further explore the relationship between renal function and urinary d-lactate concentrations in patients with diabetes. The methods developed for analyzing d-lactate concentrations range from the less sensitive method of early-stage nonenantiomer chromatography, which dates back to 1927 [20,21], to the more recent enzyme assay designed in 1980s [22,23]. These methods require the use of d-lactate dehydrogenase (EC 1.1.1.28.) to transform d-lactate to pyruvate in the presence of the coenzyme nicotinamide adenine dinucleotide (NAD+ ). The amount of NADH generated from NAD+ can be determined by ultraviolet spectrophotometry at a wavelength of 340 nm. The biggest drawback of this method is that d-lactate dehydrogenase may cross-react with many endogenous substances, including fructose 1,6-bisphosphate, 3-phosphoglyceric acid, pyruvic acid, l-lactate, and S-lactoyl-glutathione [23,24], leading to poor accuracy and low reproducibility of experimental results. Therefore, in this study, we adapted a previously established column-switching fluorescence, high-performance liquid chromatography (HPLC) method to analyze d-lactate [25]. To date, this method has only been used in rats and mice. In this study, it was used to detect human urinary dlactate levels for the first time; thus, corresponding validation tests were also performed. In this study, a fluorescence-based HPLC method was employed to quantify the urinary concentration of d-lactate in patients with diabetes. Patients were divided into groups based on renal function indicator and microalbumin values to clarify the relationship between urinary d-lactate and the corresponding microalbumin values. 2. Materials and methods 2.1. Study population Human urine samples were collected from outpatients visiting Kaohsiung Medical University (Kaohsiung, Taiwan). A total of 12 healthy individuals and 32 patients with diabetes were enrolled in this study. The healthy individuals included 7 males and 5 females, with an average age of 36 ± 17 years. Patients with diabetes included 16 males and 16 females, with an average age of 60 ± 14 years. The study protocol (KMUH-IRB-940177) was approved by the institutional review board. As per the recommendations of the American Diabetes Association (2003), patients with diabetes were divided into 3 groups (as shown in Table 1) based on the albumin-to-creatinine ratio (ACR) of their urine samples. Group DM1 comprised 15 patients with an ACR of less than 30 g albumin/mg creatinine (normal Table 1 The albumin- to- creatinine ratio (ACR) was tested in both healthy and patients with diabetes and then the subjects were divided into groups according to their ACR values. Group Healthy subjects DM1 DM2 DM3
Number 12 15 13 4
Urinary microalbumin 10.37 11.02 118.96 349.38
± ± ± ±
0.56 2.65 22.60 15.72
All data shown as mean ± SE. DM1 group: Albumin-creatinine-ratio < 30 g albumin/mg creatinine. DM2 group: Albumin-creatinine-ratio 30–299 g albumin/mg creatinine. DM3 group: Albumin-creatinine-ratio ≥300 g albumin/mg creatinine.
range) and a mean ACR of 11.02 ± 2.65 g albumin/mg creatinine. Group DM2 comprised 13 patients with an ACR of 30–299 g albumin/mg creatinine and a mean ACR of 118.96 ± 22.60 g albumin/mg creatinine. The last group, group DM3, included 4 patients with an ACR of >300 g albumin/mg creatinine and a mean ACR of 349.38 ± 15.72 g albumin/mg creatinine. 2.2. Chemicals Lithium d- and l-lactate, creatinine, cimetidine, sodium lauryl sulfate, glycine, methylumbelliferone, and bovine serum albumin were purchased from Sigma Chemical (St Louis, MO, USA). NBD-PZ, triphenylphosphine, and 2,2 -dipyridyl disulfide were purchased from Tokyo Kasei Chemicals (Tokyo, Japan). Trifluoroacetic acid, sodium dihydrogen phosphate, and o-phosphoric acid were purchased from Riedel-de Haën (Seelze, Germany). Propionic acid, sodium hydroxide, and hydrochloric acid were acquired from Nacalai Tesque Co., Ltd. (Kyoto, Japan). HPLC-grade acetonitrile (ACN) and methanol (MeOH) were purchased from Merck (Darmstadt, Germany). 2.3. Analysis of d-lactate in human urine The method for determining d-lactate levels in human urine samples was modified from a previously published, columnswitching HPLC method [26,27]. Briefly, 20 L of urine was added to 10 L of 1 mM propionic acid (used as the internal standard) and mixed with 170 L of ACN. For derivatization, the solution was centrifuged at 700 × g at 4 ◦ C for 10 min. Then, 100 L of the supernatant was added to 100 L of 8 mM NBD-PZ in ACN in the presence of 25 L of 280 mM triphenylphosphine and 25 L of 280 mM 2,2 dipyridyl disulfide. After allowing the solution to stand at 30 ◦ C for 3 h, 250 L of aqueous 0.1% trifluoroacetic acid solution was added to terminate the reaction. To remove the excess NBD-PZ, 100 L of the resulting solution was loaded twice onto a mobile phase-preconditioned, solid-phase extraction cartridge (EmporeTM SBD-RPS; 4 mm/mL; 3 M, St. Paul, MN, USA), and the eluates were subsequently combined. Twenty microliters of this combined eluate was injected into the HPLC column. A TSKgel® octadecylsilyl (ODS)-80Ts column (250 mm × 4.6 mm i.d., 5 m; Tosoh Co.; Tokyo, Japan) was used to isolate and quantify (d + l)-lactate derivatives in the samples. The mobile phase was ACN-MeOH-H2 O (12:20:68, v/v/v), and the flow rate was 0.7 mL/min. A Chiralpak® AD-RH column (150 × 4.6 mm i.d., 5 m; Daicel Co., Osaka, Japan) was used to separate the d- and l-lactates. The mobile phase was ACN-H2 O (60:40, v/v) and the isocratic flow rate was 0.3 mL/min. Fluorescence was detected at 547 nm following excitation at 491 nm (L-2385 FL detector; Hitachi; Tokyo, Japan). Lactate enantiomers were quantitatively determined from the peak areas of the chromatograms (D-7500 integrator; Hitachi; Tokyo, Japan). The urinary d-lactate concentration was expressed as nanomoles per milligram of creatinine. 2.4. Calibration curves and validation study To generate calibration curves, 20 L of lactate (D: L = 1: 1) solutions at concentrations of 0.05, 0.10, 0.20, 0.30, 0.40, or 0.50 mM were added to 10 L of 1.00 mM propionic acid (internal standard) and diluted to 170 L with ACN. Derivatization procedures were performed in a manner similar to that described above (n = 5). The calibration curve was plotted from the peak area ratio of the lactate to the internal standard, as a function of the lactate concentration. To estimate the precision and accuracy of the present analytical method, lactate determinations in human urine samples were performed by adding different concentrations of d-lactate (0, 3.91, 15.63, and 31.25 M) to 20 L of human urine in the presence of
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model
ARTICLE IN PRESS
PBA-10132; No. of Pages 6
C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
3
Fig. 1. Representative chromatograms obtained using the proposed column-switching HPLC system with NBD-PZ derivatization. (A) Separation of the D, l-lactate standard. (B) Separation of the D, l-lactate standard on a chiral column. The propionate derivative was used as an internal standard (I.S.).
10 L of 1.0 mM propionic acid (internal standard). Each sample was analyzed using the method described above (n = 5). The precision was represented by the relative standard deviation (RSD), and the accuracy was expressed in terms of the calculated recovery. 2.5. Determination of urinary creatinine levels Urinary creatinine levels were measured as described previously [28]. Briefly, cimetidine was added to urine samples as the internal standard, and the samples were separated on a Shiseido CAPCELL PAK C18 column (250 mm × 4.6 mm i.d.; particle size, 5 m; Shiseido; Tokyo, Japan). The mobile phase comprised ACNsodium lauryl sulfate (30 mM)- potassium dihydrogen phosphate (pH 3.0; 100 mM)(36:30:30, v/v/v). The absorbance was monitored at 234 nm. 2.6. Urinary microalbumin assays Urinary microalbumin was analyzed using a kit (Good Biotech Corp.; Taichung, Taiwan) according to the manufacturer’s instructions. Briefly, urine samples were conditioned with Tris buffer prior to their incubation with an anti-human albumin antibody for 10 min. These samples were then analyzed spectrophotometrically at 405 nm. To compensate for variations in the overall concentration of the spot-checked urine samples, it was helpful to normalize the amount of albumin (g/dL) in the sample with the corresponding creatinine concentration (mg/dL). 2.7. Statistical analysis Differences were determined using 1-way analysis of variance (ANOVA), and the levels of significant differences between treatment groups were determined using Scheffe’s multiple range test. Differences with p values <0.05 were considered statistically significant. 3. Results 3.1. d-Lactate standard determinations The chromatograms of the derivatives of the (D + L)-lactate standard (D: L = 1: 1) and the separated d-lactate are shown in Fig. 1. The (D + L)-lactate derivative was eluted from the ODS column at a retention time of 28.4 min (Fig. 1A). With regard to the determination of d-lactate and l-lactate, the 6-port valve in the HPLC system was switched after the appearance of the peak fraction derived from the (D + L)-lactate derivative to trap it in the loop. Then, the trapped (D + L)-lactate derivative was injected directly into the chi-
Table 2 Precision and accuracy of urinary d-lactate measurements (n = 5). d-Lactate added (M)
Measured (M) RSD (%) Recovery (%)
0
3.91
15.63
31.25
4.99 13.21 –
9.40 5.16 112.96
19.30 3.13 91.59
38.43 8.01 107.01
ral column by switching the valve back to the former position. The d-lactate derivative was eluted at 13.3 min and l-lactate derivative was eluted at 16.2 min (Fig. 1B). As per the results from the ODS column, a calibration curve was plotted from the peak area ratios of lactate to the internal standard, as a function of the lactate concentration. The calibration curve of (D+L)-lactate was generated using standards ranging from 31.25 to 2000 M, resulting in a line equation (after linear regression analysis) of Y = 0.0057X + 0.0285 (Y: the peak ratio of (D + L)-lactate to propionic acid (internal standard), X: the concentration of (D + L)-lactate). We observed good linearity with a correlation coefficient (r2) 0.9987. 3.2. Analysis of urinary d-lactate in healthy subjects and patients with diabetes As depicted in Fig. 2, the d- and l-lactate derivatives in human urine were separated as described above and successfully analyzed. The urinary lactate derived from healthy subjects and patients with diabetes was eluted from the ODS column (Fig. 2A and C, respectively). The urinary d-lactate derivative was separated with a chiral column, and data for healthy subjects and patients with diabetes are shown in Fig. 2B and D, respectively. Using the peak area proportion (P) of the d-lactate obtained from chiral column and the total (T) lactate (D + L) concentration obtained from the calibration curve of the ODS column, the d-lactate concentration (C) was quantitated using the equation C = P × T [29]. The precision (RSD) and accuracy (recovery) of the d-lactate determinations in human urine are shown in Table 2. The measured values showed an accuracy of 91.59–112.96%, and a precision of 3.13–13.21% (RSD). These results helped us to validate the present method. 3.3. Comparison of urinary d-lactate values between subjects with different albuminuria levels The albumin-to-creatinine ratio (ACR) was tested in both healthy subjects and patients with diabetes. The subjects were then divided into groups according to their ACR values (Table 1). The urinary d-lactate levels in healthy subjects and patients with diabetes are shown in Fig. 3. Patients with diabetes patients were
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model PBA-10132; No. of Pages 6 4
ARTICLE IN PRESS C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
Fig. 3. Patients with diabetes urinary d-lactate levels changed according to the albumin-to-creatinine ratio (ACR) in their urine. Group DM1 comprised patients with an ACR of less than 30 g albumin/mg creatinine (normal range); Group DM2 comprised patients with an ACR of 30–299 g albumin/mg creatinine (microalbuminuria); and Group DM3 comprised patients with an ACR of ≥300 g albumin/mg creatinine (macroalbuminuria). *p < 0.05 compared to healthy subjects.
divided into 3 groups according to their urinary ACR values (as described in Section 2.1). The urinary d-lactate levels of groups DM1–DM3 were significantly higher than those of the healthy subjects (78.31 ± 22.13, 92.47 ± 21.98, and 47.29 ± 17.51, respectively, vs. 6.28 ± 2.39 nmol/mg creatinine; p < 0.05). Among the 3 diabetic groups, the urinary d-lactate level of the DM2 group was highest. 4. Discussion
Fig. 2. Representative chromatograms of urine samples from (A) healthy subjects and (C) patients with diabetes, derivatized with NBD-PZ and separated on an ODS column. Separation of D, l-lactate in urine samples from (B) healthy subjects and (D) patients with diabetes on a chiral column. The propionate derivative was used as an internal standard (I.S.).
This was the first study to analyze the human urinary d-lactate levels using a previously developed fluorescence-based HPLC analysis system. After adding three different known concentrations of d-lactate to human urine, followed by quantitative analysis, we obtained a recovery rate in the range 91.59–112.96%, with an RSD of 3.13–13.21%. These results met the standards of the validation protocol. The recovery rate ranged between 85 and 115%, and the RSD was less than 15%. Thus, the method was validated for use in the analysis of human urinary d-lactate levels. Because urinary microalbumin can be readily measured with high sensitivity, it is widely used as a clinical indicator of renal function. Based on the standards established by the American Diabetes Association in 2003, we divided the patients with diabetes into 3 groups (DM1–DM3), according to their urinary ACRs, which can be used to estimate normal microalbumin levels and diagnose microalbuminuria and other proteinuria conditions. Diabetic nephropathy (DN) is one of the most common complications of diabetes and poses a high risk of ESRD development, if poorly controlled. Patients diagnosed with diabetes need to monitor their renal functions regularly, in consultation with their physician. When renal function enters the microalbuminuria stage, it is recommended that patients receive appropriate kidney treatments. Once the microalbumin levels exceed 299 g/mg creatinine, the kidneys develop irreversible damage and renal treatment is futile after such damage. Therefore, monitoring renal functions in patients with diabetes is critical and should be conducted in a continuous and intensive fashion. However, microalbuminuria is not only an indicator for predicting diabetic complications. It can also serve as a powerful independent risk factor for predicting cardiovascular diseases [30,31], a marker of uric acid metabolic disorders [32], and a prognostic indicator of ischemic stroke. Microalbuminuria is an early indicator of vascular endothelial cell dysfunction, as revealed by associated kidney dysfunctions. Furthermore, a significant proportion of patients with diabetes and stable normoalbuminuria have well-established DN lesions, ∼40% of which are at risk for progres-
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model PBA-10132; No. of Pages 6
ARTICLE IN PRESS C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
sion to proteinuria. These findings indicate that urinary albumin levels alone cannot be used as a specific biomarker for the detection of renal dysfunction. Thus, investigations into new risk markers or into the combined use of several currently available markers may lead to important advances in this field. d-Lactate is regarded as a stable end product of methylglyoxal [33], a highly reactive dicarbonyl compound that can cause carbonyl stress and protein malfunctions. Methylglyoxal levels are increased in the kidneys by diabetes, gentamicin, and aristolochic acid-induced nephropathy [34–36]. These findings imply that urinary d-lactate could be a marker of renal damage because it reflects toxic methylglyoxal levels in the kidney. Urinary d-lactate values varied significantly between different groups in this study. The urinary d-lactate concentration of patients in group DM1 was 12.5 times greater than that observed in healthy subjects; however, the microalbumin values of this group were still within the normal range. Patients in group DM2 were classified as having a microalbuminuria condition, and their urinary d-lactate levels were the highest among all groups studied. The kidney damages observed in patients belonging to group DM3 were considered clinically irreversible and their urinary d-lactate levels were lower than that in the patients from group DM2, but higher than that in the healthy subjects. The results suggested that in patients with diabetes whose microalbumin values were still normal, the urinary d-lactate levels were already elevated to a significant extent. However, the urinary d-lactate levels showed a decrease in patients with severe proteinuria conditions. The changes observed in urinary d-lactate levels can be explained as follows: Albuminuria is widely accepted as an early marker of DN and is observed before the occurrence of elevated serum creatinine levels and decreased glomerular filtration rates. Regarding the pathophysiological aspects of albuminuria, abnormal leakage of albumin into the urine may be caused due to increased permeability of the glomerular capillary walls, and subsequently impaired reabsorption by epithelial cells of the proximal tubules during the course of diabetes. With the progression of diabetes, increased levels of urinary albumin have been reported [37]. Early detectable markers are required before the loss of proteinfiltration function. The results of this study show that d-lactate levels are significant higher in subjects with diabetic normalalbuminuria, micro-albuminuria, and macro-albuminuria, than the levels in the healthy subjects. d-Lactate is the end product of methylglyoxal and elevated production of d-lactate may represent methylglyoxal accumulation and carbonyl stress, which commonly occur during diabetes [38]. Thus, in contrast to renal-dysfunction markers such as micro-albuminuria, d-lactate can potentially serve as an early marker because it reflects carbonyl stress that can lead to renal damage. When damage occurs to renal cells, the cells begin to consume energy that is required for repair processes. Methylglyoxal may come from the glycosis, also the fatty acid and amino acid decomposition. It accumulates during the process of energy production and can be metabolized into d-lactate, which was detected. Thus, we suggest that when lesions begin to form, the injured tissues are still capable of self-reparation. Self-reparation requires substantial energy consumption, leading to elevated production of d-lactate, which may explain the highly increased d-lactate levels observed in the DM1 and DM2 groups. In this two stage, d-lactates rise in parallel with urine microalbumin levels; thus, individuals in the DM2 group had the highest d-lactate levels. After the microalbumin level reaches more than 300 g/mg creatinine, the injured tissue begins to undergo fibrosis, and the ability for self-reparation decreases and energy utilization drops. As a consequence, d-lactate production in reduced, as observed in the DM3 group. The early control of associated risk factors, such as high blood pressure and high cholesterol, along with the adminis-
5
tration of microalbuminuria-reducing drugs, can help prevent further damage to the target organs. d-Lactate appears earlier than microalbumin in patients with diabetes. We expect that the combined measurement of urinary d-lactate and microalbumin levels can be helpful for the early identification of patients who are required to undergo therapy for diabetic nephrology. 5. Conclusion The level of urinary d-lactate is intimately related to DN. We expect urinary d-lactate can be used as an early renal damage marker; however, further experiments should be performed to investigate this topic in greater detail. Conflict of interest statement The authors have no conflicts of interest to declare. Acknowledgements This work was financially supported by the National Science Council of the Republic of China (NSC 97-2320-B-038-007-MY3). References [1] M.R. Wright, F. Jamali, Methods for the analysis of enantiomers of racemic drugs application to pharmacological and pharmacokinetic studies, J. Pharmacol. Toxicol. Methods 29 (1993) 1–9. [2] E. Racker, The mechanism of action of glyoxalase, J. Biol. Chem. 190 (1951) 685–696. [3] M. Adeva-Andany, M. Lopez-Ojen, R. Funcasta-Calderon, E. Ameneiros-Rodriguez, C. Donapetry-Garcia, M. Vila-Altesor, J. Rodriguez-Seijas, Comprehensive review on lactate metabolism in human health, Mitochondrion 17 (2014) 76–100. [4] P.J. Beisswenger, S.K. Howell, A.D. Touchette, S. Lal, B.S. Szwergold, Metformin reduces systemic methylglyoxal levels in type 2 diabetes, Diabetes 48 (1999) 198–202. [5] P.J. Thornalley, N.I. Hooper, P.E. Jennings, C.M. Florkowski, A.F. Jones, J. Lunec, A.H. Barnett, The human red blood cell glyoxalase system in diabetes mellitus, Diabetes Res. Clin. Pract. 7 (1989) 115–120. [6] A.C. McLellan, S.A. Phillips, P.J. Thornalley, The assay of methylglyoxal in biological systems by derivatization with 1,2-diamino-4,5-dimethoxybenzene, Anal. Biochem. 206 (1992) 17–23. [7] A.C. McLellan, S.A. Phillips, P.J. Thornalley, Fluorimetric assay of d-lactate, Anal. Biochem. 206 (1992) 12–16. [8] M.M. Christopher, J.D. Broussard, C.W. Fallin, N.J. Drost, M.E. Peterson, Increased serum D-lactate associated with diabetic ketoacidosis, Metabolism 44 (1995) 287–290. [9] A.D. Hingorani, N.N. Chan, d-Lactate encephalopathy, Lancet 358 (2001) 1814. [10] F. Caglayan, M. Cakmak, O. Caglayan, T. Cavusoglu, Plasma d-lactate levels in diagnosis of appendicitis, J. Invest. Surg. 16 (2003) 233–237. [11] A.J. Mayne, D.J. Handy, M.A. Preece, R.H. George, I.W. Booth, Dietary management of D-lactic acidosis in short bowel syndrome, Arch. Dis. Child. 65 (1990) 229–231. [12] G. Bongaerts, J. Tolboom, T. Naber, J. Bakkeren, R. Severijnen, H. Willems, D-Lactic acidemia and aciduria in pediatric and adult patients with short bowel syndrome, Clin. Chem. 41 (1995) 107–110. [13] S.M. Smith, R.H. Eng, J.M. Campos, H. Chmel, D-Lactic acid measurements in the diagnosis of bacterial infections, J. Clin. Microbiol. 27 (1989) 385–388. [14] A. Assadian, O. Assadian, C. Senekowitsch, R. Rotter, S. Bahrami, W. Furst, W. Jaksch, G.W. Hagmuller, W. Hubl, Plasma d-Lactate as a potential early marker for colon ischaemia after open aortic reconstruction, Eur. J. Vasc. Endovasc. Surg. 31 (2006) 470–474. [15] Y. Sun, L.X. Wang, Y.F. Zhou, S.G. Sun, X.F. Mao, X.D. Deng, P. Zhang, Effects of glutamine combined with ulinastatin on inflammatory response of patients with severe burn injury, Zhonghua Shao Shang Za Zhi. 29 (2013) 349–354. [16] T.C. Huang, S.M. Chen, Y.C. Li, J.A. Lee, Urinary d-lactate levels reflect renal function in aristolochic acid-induced nephropathy in mice, Biomed. Chromatogr. 27 (2013) 1100–1106. [17] J.P. Talasniemi, S. Pennanen, H. Savolainen, L. Niskanen, J. Liesivuori, Analytical investigation: assay of d-lactate in diabetic plasma and urine, Clin. Biochem. 41 (2008) 1099–1103. [18] M.A. Buyukbese, Complications of diabetes: chronic kidney disease (CKD) and diabetic nephropathy, J. Periodontol. 15 (2014) 2787. [19] G. Pugliese, Updating the natural history of diabetic nephropathy, Acta. Diabetol. 51 (2014) 905–915. [20] T.E. Friedemann, A.I. Kendall, The determination of lactic acid, J. Biol. Chem. 82 (1929) 23–43.
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014
G Model PBA-10132; No. of Pages 6 6
ARTICLE IN PRESS C.-K. Chou et al. / Journal of Pharmaceutical and Biomedical Analysis xxx (2015) xxx–xxx
[21] S.B. Barker, W.H. Summerson, The colorimetric determination of lactic acid in biological material, J. Biol. Chem. 138 (1941) 535–554. [22] S. Ohmori, T. Iwamoto, Sensitive determination of D-lactic acid in biological samples by high-performance liquid chromatography, J. Chromatogr. 431 (1988) 239–247. [23] R.B. Brandt, S.A. Siegel, M.G. Waters, M.H. Bloch, Spectrophotometric assay for d-(−)-lactate in plasma, Anal. Biochem. 102 (1980) 39–46. [24] R. Marti, E. Varela, R.M. Segura, J. Alegre, J.M. Surinach, C. Pascual, Determination of d-lactate by enzymatic methods in biological fluids: study of interferences, Clin. Chem. 43 (1997) 1010–1015. [25] J.-A. Lee, Y.-C. Tsai, H.-Y. Chen, C.-C. Wang, S.-M. Chen, T. Fukushima, K. Imai, Fluorimetric determination of d-lactate in urine of normal and diabetic rats by column-switching high-performance liquid chromatography, Anal. Chim. Acta 534 (2005) 185–191. [26] M.H. Lin, H.Y. Chen, T.H. Liao, T.C. Huang, C.M. Chen, J.A. Lee, Determination of time-dependent accumulation of d-lactate in the streptozotocin-induced diabetic rat kidney by column-switching HPLC with fluorescence detection, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 879 (2011) 3214–3219. [27] H. Hasegawa, T. Fukushima, J.A. Lee, K. Tsukamoto, K. Moriya, Y. Ono, K. Imai, Determination of serum d-lactic and l-lactic acids in normal subjects and diabetic patients by column-switching HPLC with pre-column fluorescence derivatization, Anal. Bioanal. Chem. 377 (2003) 886–891. [28] R. Marsilio, R. Dall’Amico, G. Giordano, L. Murer, G. Montini, M. Ros, L. Bacelle, M. Plebani, N. Dussini, G. Zacchello, Rapid determination of creatinine in serum and urine by ion-pair high-performance liquid chromatography, Int. J. Clin. Lab. Res. 29 (1999) 103–109. [29] H. Ichihara, T. Fukushima, K. Imai, Enantiomeric determination of d- and l-lactate in rat serum using high-performance liquid chromatography with a cellulose-type chiral stationary phase and fluorescence detection, Anal. Biochem. 269 (1999) 379–385.
[30] C.D. Stehouwer, Y.M. Smulders, Microalbuminuria and risk for cardiovascular disease: analysis of potential mechanisms, J. Am. Soc. Nephrol. 17 (2006) 2106–2111. [31] D. de Zeeuw, H.H. Parving, R.H. Henning, Microalbuminuria as an early marker for cardiovascular disease, J. Am. Soc. Nephrol. 17 (2006) 2100–2105. [32] T. Stakhova, A.V. Shcherbak, L.V. Kozlovskaia, M.V. Taranova, I.M. Balkarov, Clinical value of the determination of markers for endothelial dysfunction (endothelin-1, microalbuminuria) and tubulointerstitial tissue lesion (beta2-microglobulin, monocyte chemotactic protein-1) in hypertensive patients with uric acid metabolic disorders, Ter. Arkh. 86 (2014) 45–51. [33] C. Masterjohn, E. Mah, Y. Park, R. Pei, J. Lee, J.E. Manautou, R.S. Bruno, Acute glutathione depletion induces hepatic methylglyoxal accumulation by impairing its detoxification to d-lactate, Exp. Biol. Med. (Maywood) 238 (2013) 360–369. [34] Y.C. Li, Y.M. Shih, J.A. Lee, Gentamicin caused renal injury deeply related to methylglyoxal and N-epsilon-(carboxyethyl) lysine (CEL), Toxicol. Lett. 219 (2013) 85–92. [35] Y.C. Li, S.H. Tsai, S.M. Chen, Y.M. Chang, T.C. Huang, Y.P. Huang, C.T. Chang, J.A. Lee, Aristolochic acid-induced accumulation of methylglyoxal and N-epsilon-(carboxymethyl) lysine: an important and novel pathway in the pathogenic mechanism for aristolochic acid nephropathy, Biochem. Biophys. Res. Commun. 423 (2012) 832–837. [36] N. Rabbani, P.J. Thornalley, The critical role of methylglyoxal and glyoxalase 1 in diabetic nephropathy, Diabetes 63 (2014) 50–52. [37] G. D’Amico, C. Bazzi, Pathophysiology of proteinuria, Kidney Int. 63 (2003) 809–825. [38] D.L. Vander Jagt, Methylglyoxal, diabetes mellitus and diabetic complications, Drug Metabol. Drug Interact. 23 (2008) 93–124.
Please cite this article in press as: C.-K. Chou, et al., Elevated urinary d-lactate levels in patients with diabetes and microalbuminuria, J. Pharm. Biomed. Anal. (2015), http://dx.doi.org/10.1016/j.jpba.2015.06.014