Accepted Manuscript Title: Plasma metabolic profiling analysis of nephrotoxicity induced by acyclovir using metabonomics coupled with multivariate data analysis Author: Xiuxiu Zhang Yubo Li Huifang Zhou Simiao Fan Zhenzhu Zhang Lei Wang Yanjun Zhang PII: DOI: Reference:
S0731-7085(14)00229-5 http://dx.doi.org/doi:10.1016/j.jpba.2014.04.036 PBA 9562
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
18-2-2014 16-4-2014 28-4-2014
Please cite this article as: X. Zhang, Y. Li, H. Zhou, S. Fan, Z. Zhang, L. Wang, Y. Zhang, Plasma metabolic profiling analysis of nephrotoxicity induced by acyclovir using metabonomics coupled with multivariate data analysis, Journal of Pharmaceutical and Biomedical Analysis (2014), http://dx.doi.org/10.1016/j.jpba.2014.04.036 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Plasma
metabolic
profiling
2
nephrotoxicity
3
metabonomics coupled with multivariate data
4
analysis
by
acyclovir
of using
ip t
induced
analysis
cr
5
Xiuxiu Zhanga, Yubo Lia, Huifang Zhoub, Simiao Fana, Zhenzhu Zhanga, Lei Wanga,
7
Yanjun Zhanga, *
8
a
9
Chinese Materia Medica, Tianjin University of Traditional Chinese Medicine, 312
us
6
Anshan west Road, Tianjin 300193, China.
11
b
12
Medicine, 88 Yuquan Road, Tianjin 300193, China.
13
*
14
18 19
Acknowledgments
20
This project was supported by the National Basic Research Program of China (973
21
Program) (2011CB505300, 2011CB505302), and the National Natural Science
22
foundation of China (No. 81273998).
15 16 17
M
10
Ac ce p
an
Tianjin State Key Laboratory of Modern Chinese Medicine, School of Traditional
te
d
Department of experimental teaching, Tianjin University of Traditional Chinese
Author for correspondence:
e-mail: [email protected]. Tel and Fax number: +86-22-59596223. Xiuxiu Zhang and Yubo Li contributed equally to the work as co-first authors. Conflict of interest statement
The authors declare no competing financial interest.
1 Page 1 of 21
Abstract
2
Acyclovir (ACV) is an antiviral agent. However, its use is limited by adverse side
3
effects, particularly by its nephrotoxicity. Metabonomics technology can provide
4
essential information on the metabolic profiles of biofluids and organs upon drug
5
administration. Therefore, in this study, mass spectrometry–based metabonomics
6
coupled with multivariate data analysis was used to identify the plasma metabolites
7
and metabolic pathways related to nephrotoxicity caused by intraperitoneal injection
8
of low (50 mg/kg) and high (100 mg/kg) doses of acyclovir. Sixteen biomarkers were
9
identified by metabonomics and nephrotoxicity results revealed the dose-dependent
10
effect of acyclovir on kidney tissues. The present study showed that the top four
11
metabolic pathways interrupted by acyclovir included the metabolisms of arachidonic
12
acid, tryptophan, arginine and proline, and glycerophospholipid. This research proves
13
the established metabonomic approach can provide information on changes in
14
metabolites and metabolic pathways, which can be applied to in-depth research on the
16
17
cr
us
an
M
d
te
Ac ce p
15
ip t
1
mechanism of acyclovir–induced kidney injury.
Keyword: Acyclovir, Nephrotoxicity, Metabonomics, Multivariate data analysis. 1. Introduction
18
Acyclovir (ACV) is a synthetic nucleoside analog known for its therapeutic effects
19
on viral infections, especially those caused by the herpes simplex and herpes zoster
20
viruses [1,2]. Unfortunately, the clinical use of ACV may result in adverse side effects,
21
of which nephrotoxicity is one of the most severe [3]. ACV-induced nephrotoxicity
2 Page 2 of 21
1
largely affects the quality of patients’ lives. Thus, finding approaches to reveal
2
biomarkers characteristic of ACV-induced renal injury is meaningful objective. Metabonomics, a branch of systems biology belonging to the “omics” field,
4
provides global metabolic profile information on biological samples, including cell-,
5
tissue-, and organism-level samples, under a specific condition, such as by
6
pathophysiological stimulation, genetic modification, or environmental conditions
7
[4-7]. Metabonomics focuses on small and low-weight molecules that are the final
8
product of biological metabolite pathways and, as such, play an important role in
9
metabolism [8]. Metabonomics has become a powerful method in the study of many
10
fields, such as toxicology, disease diagnosis, and therapeutic efficacy [7,9]. It is
11
largely used in finding new biomarkers related to drug-induced renal injury [10,11]
M
an
us
cr
ip t
3
In the present study, we used mass spectrometry–based metabonomics to determine
13
biomarkers of nephrotoxicity induced by acyclovir, trying to find disturbed pathways
14
related to ACV-induced kidney injury. The results found in this study may provide
16 17 18
te
Ac ce p
15
d
12
information for clinical use of acyclovir, and lay the foundation for further research on mechanism of ACV–induced nephrotoxicity.
2. Material and methods 2.1 Reagents and materials
19
Acetonitrile (HPLC-grade) was purchased from Oceanpak (Goteborg, Sweden).
20
Formic acid was purchased from ROE (USA). Distilled water was obtained from
21
Wahaha Company (Hangzhou, China). ACV was obtained from Jiangsu Hansoh
22
Pharmaceutical Co., Ltd. (Lianyungang, China). Assay kits for BUN (blood urea
3 Page 3 of 21
1
nitrogen) and Scr (serum creatinine) were bought from the Biosino Bio-technology
2
and Science Inc. (Beijing, China).
3
2.2 Animal treatment Male Wistar rats weighing 200 ± 20 g were raised in an SPF-level lab and were
5
acclimatized in metabolism cages for one week before drug administration. The
6
animal were randomly divided into three groups: the control group (NS), the low-dose
7
ACV group (LA), and the high-dose ACV group (HA). ACV was first dissolved in
8
normal saline (0.9% w/v) before administration. ACV was administered
9
intraperitoneally to the rats daily for three consecutive days, and the dosages for LA
10
and HA groups were set as 50 and 100 mg/kg. The control group was intraperitoneally
11
administered with an equivalent volume of normal (0.9% saline) for three consecutive
12
days. On the last day, 800 μL of blood was drawn from the intraorbital angular vein
13
after slight anesthetization. After blood collection, the rats were sacrificed, and kidney
14
tissues were immediately removed and stored in 10% formalin solution. Blood
16 17 18 19
cr
us
an
M
d
te
Ac ce p
15
ip t
4
samples (including serum and plasma) were centrifuged at 3500 rpm for 15 min. The obtained serum was used for biochemical assay. Plasma samples were stored at –80 °C prior to metabonomics analysis. Kidney slices were stained with hematoxylin and eosin to observe pathological features using a light microscope at 200× magnification. 2.3 Metabonomics data acquisition
20
Data acquisition was performed by rapid-resolution liquid chromatography coupled
21
with quadruple-time-of-flight mass spectrometry (RRLC-Q-TOF-MS) (Agilent, USA).
22
The preparation of plasma was as follows: 100 μL thawed plasma was mixed with 300
4 Page 4 of 21
μL of acetonitrile, ultrasonicated for 10 min in cold water, vortexed for 1 min, and
2
centrifuged at 15000 rpm for 15 min. A 10 µL aliquot of the supernatant was injected
3
onto an ACQUITY UPLC HSS C18 column (2.1 × 100 mm, 1.7 μm; Waters). The
4
column temperature was 40 °C, and the flow rate was 0.3 mL/min. The RRLC binary
5
solvent system consisted of mobile phases A (0.1% formic acid in water) and B (0.1%
6
formic acid in acetonitrile). Gradient elution was adopted to obtain the whole profile
7
of plasma metabolites. The gradient started with 1% B, then, 0-3 min, B: 1%-52%;
8
3-7 min, B: 52%-74%; 7-9 min, B: 74%-80%; 9-10 min, B: 80%-90%; 10-12 min, B:
9
90%-99%; 12-16 min, B: 99%-99%; 16-17min, B: 99%-1%; 17-20 min, B: 1%-1%.
cr
us
an
2.4 Metabonomics data processing
M
10
ip t
1
Multivariate data analysis, including principal component analysis (PCA) and
12
partial least squares-discriminant analysis (PLS-DA), was used to identify the plasma
13
metabolites responsible for the differentiation of the treatment groups (NS, LA, HA).
14
Multivariate data analysis was performed with SIMCA-P+ 11.5 software (Umetrics
16 17 18 19
te
Ac ce p
15
d
11
AB, Umea, Sweden). In our present study, metabolites that were far from the cluster in the s-plot and loading plots and with a VIP value >1.0 were chosen as biomarkers. An independent sample t-test was performed using SPSS 17.0 software to determine whether or not the chosen biomarkers were significantly changed. 2.5 Characterisation of biomarkers
20
Candidates of metabolites were searched from human metabolome database
21
(HMDB, http://www.hmdb.ca/) utilising the detected molecular weights. The
22
biomarkers were unambiguously characterised by comparison with reference
5 Page 5 of 21
1
standards or MS/MS fragment information. The perturbation was interrogated using
2
MetPA
3
important tool in metabonomics pathway analysis.
4
3. Results and discussion
5
3.1 Clinical chemistry and histopathological examination
which
was
an
cr
ip t
(http://metpa.metabolomics.ca./MetPA/faces/Home.jsp),
The blood urea nitrogen (BUN) level significantly increased in rats exposed to the
7
high dose of ACV but showed no obvious changes in rats administered the low dose
8
of ACV (Fig 1A). Serum creatinine (Scr) activity significantly increased by both high
9
and low doses of ACV (Fig 1B). The detailed data of clinical chemistry was shown in
10
Table S1. BUN and Scr are a generally conventional monitor of nephrotoxicity and
11
have been used as a standard determiner of kidney injury for many years. When BUN
12
and Scr level significantly elevate, it shows the kidney has been injured.
te
d
M
an
us
6
As shown in Fig. S1, the high dose of ACV caused apparent tubular dilation,
14
irregular arrangement of cells, and development of interstitial inflammatory cells.
15 16 17 18 19
Ac ce p
13
Compared with the HA group, the LA group showed milder kidney damage, such as apparent tubular dilation, tubular epithelial edema, and irregular arrangement of cells. Coupled with the results from the biochemical analysis, ACV-induced nephrotoxicity could be confirmed in rats in the HA and LA groups. 3.2 Metabolic profiling analysis
20
Typical total ion current (TIC) chromatograms of plasma samples were obtained
21
from rats in the control and ACV-dosed groups in positive mode. Some discrimination
22
was found in the TIC chromatograms in positive mode between NS, HA and LA
6 Page 6 of 21
groups (Fig. S2). PCA and PLS-DA was further applied to better visualize the
2
discrimination among the gained complex data. Fig 2A demonstrated distinctions
3
between the different groups by PCA. The samples in the low-dose ACV group were
4
located near those of the control group, whereas samples from the high-dose ACV
5
group were found far from those of the control group. These data indicated that the
6
nephrotoxicity induced by ACV was dose-dependent. The differentiation by PLS-DA
7
between the LA and NS and HA and NS groups are shown in Fig 2B and 2C. Based
8
on these scores plots, the rat plasma showed clear changes due to ACV-administration.
9
The S-plot and loading plot of PLS-DA model between HA and NS group were shown
10
in Fig 2D and 2E. The ions in the top or bottom of S-plot or the variables located far
11
from the middle cluster in the loading plot most likely corresponded to the
12
representative of high score values in VIP plot[12]. It was these ions and variables
13
that mostly contributed to the significant changes between the different groups, and
14
these were selected as the potential biomarkers.
16 17 18
cr
us
an
M
d
te
Ac ce p
15
ip t
1
3.3 Identification of potential biomarkers From multivariate statistical analysis, these specific metabolites stood out from
the large amount of plasma endogenous metabolites. Meanwhile, the identification of the m/z of metabolites can be achieved. We searched candidates from biochemical
19
databases of HMDB, http://www.hmdb.ca/. We ignored the candidates from the
20
database identified as exogenous and focused on endogenous metabolites. In this
21
study, 16 metabolites were identified as potential biomarkers of nephrotoxicity in the
22
rat plasma. The metabolites /biomarkers identified are summarized in Table 1. Urea,
7 Page 7 of 21
proline, creatinine and arachidonic acid were identified by comparison with reference.
2
The residual 12 metabolites were identified by their MS/MS fragmentation. The
3
fragment information of metabolites is provided in Table 1. The heatmap helped to
4
visualise the biomarkers responsible for the differentiation of the three groups (Fig.
5
S3). An example of the characterization of the biomarker, lysophosphatidylcholine
6
(16:0) (tR=9.92min, m/z 496.3405), is shown in Fig. S4. Among the selected
7
biomarkers, some of them belonged to lysophosphatidylcholine (LPC). In blood
8
plasma significant amounts of lysophosphatidylcholine are formed by the enzyme,
9
phospholipase A2. This enzyme catalyzes the transfer of the fatty acids of position
10
sn-2 of phosphatidylcholine to the free cholesterol in plasma, with formation of
11
cholesterol esters and lysophosphatidylcholine[13].
12
3.4 Relationship between biomarkers and nephrotoxicity
te
d
M
an
us
cr
ip t
1
In this study, increased urea was detected in the HA dose group following
14
ACV-administration. Blood urea is excreted via the kidneys, hence increased levels of
15 16 17 18
Ac ce p
13
urea in the plasma can indicate impaired kidney function. The observed increased levels of urea implied that ACV administration has damaged the normal functions of the glomerulus [14-16]. The increased level of another biomarker in plasma, creatinine, also supported the damage in the normal functions of the glomerulus
19
caused by ACV, especially at the high dose [9,14]. The increased level of creatinine in
20
blood was also detected after cyclosporin A administration[17], indicating
21
drug-induced nephrotoxicity. Raised arachidonic acid, which is an essential
22
polyunsaturated fatty acid, has been shown previously to mediate inflammation
8 Page 8 of 21
[18-20]. This finding indicated that ACV has induced kidney injury through active
2
regulation of the inflammatory pathway. Proline is part of the arginine metabolism.
3
Tryptophan is an essential amino acid and the precursor of serotonin[12]. These
4
disturbed amino acid metabolism pathways indicated that ACV may affect normal
5
energy intake and consumption. Phytosphingosine is a phospholipid that is part of
6
sphingolipid metabolism[21], implying that ACV may disturb this metabolism.
7
Lysophosphatidylcholines (LPCs) were also found as possible biomarkers, which
8
were also identified in other nephrotic related work[22]. LPC control many biological
9
pathways, including cell proliferation [23], inflammation [24], and vasodilation [25].
cr
us
an
The change in LPC content may be a probable result of ACV-induced kidney injury.
M
10
ip t
1
From a metabolic level, ACV-induced nephrotoxicity of the biomarkers identified
12
by metabonomics may be related to damage to the glomerulus, inflammation,
13
abnormal energy intake and consumption, disturbed sphingolipid metabolism, and
14
changes in LPC.
16 17 18
te
Ac ce p
15
d
11
3.5 Disturbed metabolic pathways MetPA was applied for pathway analysis of selected biomarkers. The top four
metabolism pathways disturbed in the rats of the LA group were (a) arachidonic acid metabolism, (b) tryptophan metabolism, (c) arginine and proline metabolism, and (d)
19
glycerophospholipid metabolism (Fig. S5). In the HA group, pathways a–d were also
20
responsible for ACV-induced kidney injury. ACV may induce renal injury by initially
21
disrupting these four pathways.
22
4. Conclusion
9 Page 9 of 21
In the present study, two doses (50 and 100 mg/kg) of ACV were chosen, a kidney
2
injury model was built, and the dose-dependent effects of ACV on nephrotoxicity
3
were observed. An MS-based metabonomic approach was used to analyze changes in
4
the plasma. Sixteen biomarkers were identified to be related to ACV-induced
5
nephrotoxicity. The top four interrupted metabolic pathways included metabolism of
6
arachidonic acid, tryptophan, arginine and proline, and glycerophospholipid. The
7
method established in the present study provides useful information for ACV-induced
8
nephrotoxicity and may lay foundation for further toxic mechanism studies.
9
References
an
us
cr
ip t
1
[1] D. Lembo, S. Swaminathan, M. Donalisio, A. Civra, L. Pastero, D. Aquilano, P.
11
Vavia, F. Trotta, R. Cavalli, Encapsulation of Acyclovir in new carboxylated
12
cyclodextrin-based nanosponges improves the agent’s antiviral efficacy, Int J Pharm.
13
443 (2013) 262-272.
14
[2] N.N. Lam, M.A. Weir, Z. Yao, P.G. Blake, M.M. Beyea, T. Gomes, S. Gandhi, M.
16 17 18
d
te
Ac ce p
15
M
10
Mamdani, R. Wald, C.R. Parikh, D.G. Hackam, A.X. Garg, Risk of acute kidney injury from oral Acyclovir: a population-based study, Am J Kidney Dis. 61 (2013) 723-729.
[3] H. Yoon, K.Y. Rhew, Famciclovir as an antiviral agent for a patient with acute
19
renal failure, Int J Clin Pharm. 35 (2013) 173-175.
20
[4] D. Portilla, L. Schnackenberg, R.D. Beger, Metabolomics as an extension of
21
proteomic analysis: study of acute kidney injury, Semin Nephrol. 27 (2007) 609-620.
22
[5] X. Wang, B. Yang, H. Sun, A. Zhang, Pattern recognition approaches and
10 Page 10 of 21
computational systems tools for ultra performance liquid chromatography-mass
2
spectrometry-based comprehensive metabolomic profiling and pathways analysis of
3
biological data sets, Anal Chem. 84 (2012) 428-439.
4
[6] S. Rochfort, Metabolomics reviewed: a New “Omics” platform technology for
5
systems biology and implications for natural products research, J Nat Prod. 68 (2005)
6
1813-1820.
7
[7] J. Xuan, G. Pan, Y. Qiu, L. Yang, M. Su, Y. Liu, J. Chen, G. Feng, Y. Fang, W. Jia,
8
Q. Xing, L. He, Metabolomic profiling to identify potential serum biomarkers for
9
schizophrenia and risperidone action, J Proteome Res. 10 (2011) 5433-5443.
an
us
cr
ip t
1
[8] S. Rezzi, Z. Ramadan, L.B. Fay, S. Kochhar, Nutritional metabonomics:
11
applications and perspectives, J Proteome Res. 6 (2007) 513-525.
12
[9] R.D. Beger, J. Sun, L.K. Schnackenberg, Metabolomics approaches for
13
discovering biomarkers of drug-induced hepatotoxicity and nephrotoxicity, Toxicol
14
Appl Pharm. 243 (2010) 154-166.
16 17 18
d
te
Ac ce p
15
M
10
[10] C. Ma, K. Bi, M. Zhang, D. Su, X. Fan, W. Ji, C. Wang, X. Chen, Metabonomic study of biochemical changes in the urine of Morning Glory Seed treated rat, J Pharm Biomed Anal. 53 (2010) 559-566. [11] S. Peng, L. Yan, J. Zhang, Z. Wang, M. Tian, H. Shen, An integrated
19
metabonomics and transcriptomics approach to understanding metabolic pathway
20
disturbance induced by perfluorooctanoic acid, J Pharm Biomed Anal. 86 (2013)
21
56-64.
22
[12] X. Wang, H. Wang, A. Zhang, X. Lu, H. Sun, H. Dong, P. Wang, Metabolomics
11 Page 11 of 21
Study on the Toxicity of Aconite Root and Its Processed Products Using Ultrap-
2
erformance Liquid-Chromatography/ Electrospray-Ionization Synapt High-Definition
3
Mass Spectrometry Coupled with Pattern Recognition Approach and Ingenuity
4
Pathways Analysis, J Proteome Res. 11 (2012) 1284–1301.
5
[13] U.P. Steinbrecher, P.H. Pritchard, Hydrolysis of phosphatidylcholine during
6
LDLoxidation is mediated by platelet-activating factor acetylhydrolase, J Lipid Res.
7
30 (1989) 305-315.
8
[14] O. Shemesh, H. Golbetz, J.P. Kriss, B.D. Myers, Limitations of creatinine as a
9
filtration marker in glomerulopathic patients, Kidney Int. 28 (1985) 830-838.
an
us
cr
ip t
1
[15] M.W. Su, S.S. Chang, C.H. Chen, C.C. Huang, S.W. Chang, Y.C. Tsai, C.F. Lam,
11
Preconditioning renoprotective effect of isoflurane in a rat model of virtual renal
12
transplant, J Surg Res. doi: 10.1016/j.jss.2014.02.035.
13
[16] C.U. Niemann, N.J. Serkova, Biochemical mechanisms of nephrotoxicity:
14
application for metabolomics, Expert Opin Drug Metab Toxicol. 3 (2007) 527-544.
16 17 18
d
te
Ac ce p
15
M
10
[17] N.J. Serkova, U. Christians, Biomarkers for toxicodynamic monitoring of immunosuppressants: NMR-based quantitative metabonomics of the blood, Ther Drug Monit. 27 (2005) 733-737. [18] M. Shoeb, U.C. Yadav, S.K. Srivastava, K.V. Ramana, Inhibition of Aldose
19
Reductase Prevents Endotoxin-Induced Inflammation by Regulating Arachidonic
20
Acid Pathway in Murine Macrophages, Free Radic Biol Med. 51 (2011) 1686-1696.
21
[19] C. Pompeia, T. Lima, R. Curi, Arachidonic acid cytotoxicity: can arachidonic
22
acid be a physiological mediator of cell death? Cell Biochem Funct. 21(2003) 97-104.
12 Page 12 of 21
[20] M. Chen, M. Su, L. Zhao, J. Jiang, P. Liu, J. Cheng, Y. Lai, Y. Liu, W. Jia,
2
Metabonomic Study of Aristolochic Acid-Induced Nephrotoxicity in rats, J Proteome
3
Res. 5 (2006) 995-1002.
4
[21] M.T. Park, J.A. Kang, J.A. Choi, C.M. Kang, T.H. Kim, S. Bae, S. Kang, S. Kim,
5
W.I. Choi, C.K. Cho, H.Y. Chung, Y.S. Lee, S.J. Lee. Phytosphingosine induces
6
apoptotic cell death via caspase 8 activation and Bax translocation in human cancer
7
cells, Clin Cancer Res. 9 (2003) 878-885.
8
[22] C. Ma, K. Bi, D. Su, W. Ji, M. Zhang, X. Fan, C. Wang, X. Chen, Serum and
9
kidney metabolic changes of rat nephrotoxicity induced by Morning Glory Seed, Food
an
us
cr
ip t
1
Chem Toxicol. 48 (2010) 2988-2993.
11
[23] J. Yang, X. Zhao, X. Liu, C. Wang, P. Gao, J. Wang, L. Li, J. Gu, S. Yang, G. Xu,
12
High Performance Liquid Chromatography-Mass Spectrometry for Metabonomics:
13
Potential Biomarkers for Acute Deterioration of Liver Function in Chronic Hepatitis
14
B, J Proteome Res. 5 (2006) 554-561.
16 17 18
d
te
Ac ce p
15
M
10
[24] S. Koizumi, S. Yamamoto, T. Hayasaka, Y. Konishi, M. Yamaguchi-Okada, N. Goto-Inoue, Y. Sugiura, M. Setou, H. Namba, Imaging Mass Spectrometry Revealed The Production Of Lyso-Phosphatidylcholine In The Injured Ischemic Rat Brain, Neuroscience. 168 (2010) 219-225.
19
[25] T.D. Vuong, K.S. de, R.R. de, T.J. Rabelink, H.A. Koomans, J.A. Joles, Albumin
20
restores lysophosphatidylcholine-induced inhibition of vasodilation in rat aorta,
21
Kidney Int. 60 (2001) 1088-1096.
22
Figure Captions
13 Page 13 of 21
Fig 1. Effect of ACV on BUN and Scr levels. (A) Changes in BUN levels. (B)
2
Changes in Scr levels. Data represents mean ± SD (** p < 0.01, compared with the
3
control group).
4
Fig 2. Result of multivariate statistical analysis. (A) PCA score plot of the LA, HA
5
and NS groups. (B) PLS-DA score plot of the HA and NS groups. (C) PLS-DA score
6
plot of the LA and NS groups. (D) S-plot of PLS-DA model between the HA and NS
7
groups. (E) Loading plot of PLS-DA model between the HA and NS groups.
us
cr
ip t
1
Ac ce p
te
d
M
an
8
14 Page 14 of 21
1
Highlights
2
• Sixteen plasma metabolites were identified as changed following acyclovir
6 7 8 9
ip t
metabonomics.
cr
5
• Four disturbed metabolic pathways due to acyclovir were identified by
• Kidney injury model was built following intraperitoneal injection of two doses of
us
4
administration.
acyclovir.
• Metabonomics technology can provide information for toxic mechanism of
an
3
acyclovir.
M
10
Ac ce p
te
d
11
17 Page 15 of 21
Ac
ce
pt
ed
M
an
us
cr
i
*Graphical Abstract
Page 16 of 21
Identified potential biomarkers in different groups.
tR (min)
metabolite
obsd [M+H]+
calcd [M+H]+
1 2 3 4
0.89 0.92 0.93 5.17
Ureab Creatinineb Prolineb Tryptophanc
61.0398 114.0661 116.0704 205.0968
61.0396 114.0662 116.0706 205.0972
3.3 -0.9 -0.9 -2.0
CH4N2O C4H7N3O C5H9NO2 C11H12N2O2
↑ ↑ ↓ ↓
5
11.64
305.2474
305.2475
-0.3
C20H32O2
↑
LA and HA
6
7.99
Arachidonic acidb Phytosphingosine
-2.8
C18H39NO3
↓
LA and HA
7
12.66
LPC(15:0)c
8
9.46
9
9.92
formula
M an
318.3003
ce pt
c
318.2994
error (ppm)
us
no
The group identified this biomarkerd LA and HA LA and HA LA and HA LA and HA
ed
Table 1.
cr
ip t
Table(s)
Content variancea
482.3241
-1.7
C23H48NO7P ↑
LA and HA
LPC(16:1)c
494.3236
494.3241
-1.0
C24H48NO7P ↓
HA
LPC(16:0)c
496.3405
496.3398
1.4
C24H50NO7P ↓
LA and HA
Ac
482.3233
MS/MS
205.1 [M+H]+ 188.1 [M+H-NH2]+
318.3 [M+H]+ 300.2 [M+H-H2O]+ 256.2 [M+H-C2H5O2]+ 482.3 [M+H]+ 464.3 [M+H-H2O]+ 405.3 [M+H-C3H10NO]+ 299.3 [M+H-C5H13NO4P]+ 184.1 [M+H-C18H35NO2]+ 494.3 [M+H]+ 476.3 [M+H-H2O]+ 184.1 [M+H-C19H35NO2]+ 125.0 [M+H-C22H42NO3]+ 104.0 [M+H-C20H40NO4P]+ 496.3 [M+H]+ 478.3 [M+H-H2O]+
Page 17 of 21
ip t cr LPC(20:5)c
542.3223
542.3241
11
10.71
LPC(20:3)c
546.3556
546.3554
12
11.35
LPC(20:2)c
548.3706
13
12.98
LPC(20:1)c
14
10.40
15
12.98
-3.3
us
9.78
C28H48NO7P ↓
ed
M an
10
ce pt
548.3711
HA
0.4
C28H52NO7P ↓
HA
-0.9
C28H54NO7P ↓
LA and HA
550.3867
1.3
C28H56NO7P ↓
LA and HA
LPC(22:5)c
570.3556
570.3554
0.4
C30H52NO7P ↓
HA
LPC(22:4)c
572.3695
572.3711
-2.8
C30H54NO7P ↓
HA
Ac
550.3874
313.2 [M+H-C5H13NO4P]+ 184.1 [M+H-C19H37NO2]+ 104.1 [M+H-C20H42NO4P]+ 542.3 [M+H]+ 524.3 [M+H-H2O]+ 259.1 [M+H-C17H33NO2]+ 185.1 [M+H-C19H36NO3P]+ 126.0 [M+H-C26H42NO3]+ 546.4 [M+H]+ 528.4 [M+H-H2O]+ 184.1 [M+H-C18H36NO4P]+ 125.1 [M+H-C26H46NO3]+ 548.4 [M+H]+ 530.4 [M+H-H2O]+ 184.1 [M+H-C23H41NO2]+ 149.1 [M+H-C24H48NO3]+ 104.1 [M+H-C24H46NO4P]+ 550.4 [M+H]+ 532.4 [M+H-H2O]+ 185.1 [M+H-C23H43NO2]+ 105.1 [M+H-C24H48NO4P]+ 570.4 [M+H]+ 552.3 [M+H-H2O]+ 184.1 [M+H-C21H41NO3P]+ 125.0 [M+H-C28H47NO3]+ 572.4 [M+H]+ 554.4 [M+H-H2O]+ 258.2 [M+H-C13H32NO5P]+
Page 18 of 21
ip t LPC(22:2)c
576.4017
576.4024
-1.2
cr
13.42
C30H58NO7P ↓
a
M an
us
16
HA
184.1 [M+H-C21H42NO3P]+ 576.4 [M+H]+ 558.4 [M+H-H2O]+ 184.1 [M+H-C25H45NO2]+ 166.1 [M+H-C25H45O4]+ 104.1 [M+H-C26H50NO4P]+
↑, content increased; ↓, content decreased. bconfirmed by commercial standards. cconfirmed by MS/MS information. d the groups in
Ac
ce pt
ed
which biomarkers were identified.
Page 19 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Fig.1
Page 20 of 21
Ac
ce
pt
ed
M
an
us
cr
i
Fig.2
Page 21 of 21
S0731-7085(14)00229-5 http://dx.doi.org/doi:10.1016/j.jpba.2014.04.036 PBA 9562
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
18-2-2014 16-4-2014 28-4-2014
Please cite this article as: X. Zhang, Y. Li, H. Zhou, S. Fan, Z. Zhang, L. Wang, Y. Zhang, Plasma metabolic profiling analysis of nephrotoxicity induced by acyclovir using metabonomics coupled with multivariate data analysis, Journal of Pharmaceutical and Biomedical Analysis (2014), http://dx.doi.org/10.1016/j.jpba.2014.04.036 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Plasma
metabolic
profiling
2
nephrotoxicity
3
metabonomics coupled with multivariate data
4
analysis
by
acyclovir
of using
ip t
induced
analysis
cr
5
Xiuxiu Zhanga, Yubo Lia, Huifang Zhoub, Simiao Fana, Zhenzhu Zhanga, Lei Wanga,
7
Yanjun Zhanga, *
8
a
9
Chinese Materia Medica, Tianjin University of Traditional Chinese Medicine, 312
us
6
Anshan west Road, Tianjin 300193, China.
11
b
12
Medicine, 88 Yuquan Road, Tianjin 300193, China.
13
*
14
18 19
Acknowledgments
20
This project was supported by the National Basic Research Program of China (973
21
Program) (2011CB505300, 2011CB505302), and the National Natural Science
22
foundation of China (No. 81273998).
15 16 17
M
10
Ac ce p
an
Tianjin State Key Laboratory of Modern Chinese Medicine, School of Traditional
te
d
Department of experimental teaching, Tianjin University of Traditional Chinese
Author for correspondence:
e-mail: [email protected]. Tel and Fax number: +86-22-59596223. Xiuxiu Zhang and Yubo Li contributed equally to the work as co-first authors. Conflict of interest statement
The authors declare no competing financial interest.
1 Page 1 of 21
Abstract
2
Acyclovir (ACV) is an antiviral agent. However, its use is limited by adverse side
3
effects, particularly by its nephrotoxicity. Metabonomics technology can provide
4
essential information on the metabolic profiles of biofluids and organs upon drug
5
administration. Therefore, in this study, mass spectrometry–based metabonomics
6
coupled with multivariate data analysis was used to identify the plasma metabolites
7
and metabolic pathways related to nephrotoxicity caused by intraperitoneal injection
8
of low (50 mg/kg) and high (100 mg/kg) doses of acyclovir. Sixteen biomarkers were
9
identified by metabonomics and nephrotoxicity results revealed the dose-dependent
10
effect of acyclovir on kidney tissues. The present study showed that the top four
11
metabolic pathways interrupted by acyclovir included the metabolisms of arachidonic
12
acid, tryptophan, arginine and proline, and glycerophospholipid. This research proves
13
the established metabonomic approach can provide information on changes in
14
metabolites and metabolic pathways, which can be applied to in-depth research on the
16
17
cr
us
an
M
d
te
Ac ce p
15
ip t
1
mechanism of acyclovir–induced kidney injury.
Keyword: Acyclovir, Nephrotoxicity, Metabonomics, Multivariate data analysis. 1. Introduction
18
Acyclovir (ACV) is a synthetic nucleoside analog known for its therapeutic effects
19
on viral infections, especially those caused by the herpes simplex and herpes zoster
20
viruses [1,2]. Unfortunately, the clinical use of ACV may result in adverse side effects,
21
of which nephrotoxicity is one of the most severe [3]. ACV-induced nephrotoxicity
2 Page 2 of 21
1
largely affects the quality of patients’ lives. Thus, finding approaches to reveal
2
biomarkers characteristic of ACV-induced renal injury is meaningful objective. Metabonomics, a branch of systems biology belonging to the “omics” field,
4
provides global metabolic profile information on biological samples, including cell-,
5
tissue-, and organism-level samples, under a specific condition, such as by
6
pathophysiological stimulation, genetic modification, or environmental conditions
7
[4-7]. Metabonomics focuses on small and low-weight molecules that are the final
8
product of biological metabolite pathways and, as such, play an important role in
9
metabolism [8]. Metabonomics has become a powerful method in the study of many
10
fields, such as toxicology, disease diagnosis, and therapeutic efficacy [7,9]. It is
11
largely used in finding new biomarkers related to drug-induced renal injury [10,11]
M
an
us
cr
ip t
3
In the present study, we used mass spectrometry–based metabonomics to determine
13
biomarkers of nephrotoxicity induced by acyclovir, trying to find disturbed pathways
14
related to ACV-induced kidney injury. The results found in this study may provide
16 17 18
te
Ac ce p
15
d
12
information for clinical use of acyclovir, and lay the foundation for further research on mechanism of ACV–induced nephrotoxicity.
2. Material and methods 2.1 Reagents and materials
19
Acetonitrile (HPLC-grade) was purchased from Oceanpak (Goteborg, Sweden).
20
Formic acid was purchased from ROE (USA). Distilled water was obtained from
21
Wahaha Company (Hangzhou, China). ACV was obtained from Jiangsu Hansoh
22
Pharmaceutical Co., Ltd. (Lianyungang, China). Assay kits for BUN (blood urea
3 Page 3 of 21
1
nitrogen) and Scr (serum creatinine) were bought from the Biosino Bio-technology
2
and Science Inc. (Beijing, China).
3
2.2 Animal treatment Male Wistar rats weighing 200 ± 20 g were raised in an SPF-level lab and were
5
acclimatized in metabolism cages for one week before drug administration. The
6
animal were randomly divided into three groups: the control group (NS), the low-dose
7
ACV group (LA), and the high-dose ACV group (HA). ACV was first dissolved in
8
normal saline (0.9% w/v) before administration. ACV was administered
9
intraperitoneally to the rats daily for three consecutive days, and the dosages for LA
10
and HA groups were set as 50 and 100 mg/kg. The control group was intraperitoneally
11
administered with an equivalent volume of normal (0.9% saline) for three consecutive
12
days. On the last day, 800 μL of blood was drawn from the intraorbital angular vein
13
after slight anesthetization. After blood collection, the rats were sacrificed, and kidney
14
tissues were immediately removed and stored in 10% formalin solution. Blood
16 17 18 19
cr
us
an
M
d
te
Ac ce p
15
ip t
4
samples (including serum and plasma) were centrifuged at 3500 rpm for 15 min. The obtained serum was used for biochemical assay. Plasma samples were stored at –80 °C prior to metabonomics analysis. Kidney slices were stained with hematoxylin and eosin to observe pathological features using a light microscope at 200× magnification. 2.3 Metabonomics data acquisition
20
Data acquisition was performed by rapid-resolution liquid chromatography coupled
21
with quadruple-time-of-flight mass spectrometry (RRLC-Q-TOF-MS) (Agilent, USA).
22
The preparation of plasma was as follows: 100 μL thawed plasma was mixed with 300
4 Page 4 of 21
μL of acetonitrile, ultrasonicated for 10 min in cold water, vortexed for 1 min, and
2
centrifuged at 15000 rpm for 15 min. A 10 µL aliquot of the supernatant was injected
3
onto an ACQUITY UPLC HSS C18 column (2.1 × 100 mm, 1.7 μm; Waters). The
4
column temperature was 40 °C, and the flow rate was 0.3 mL/min. The RRLC binary
5
solvent system consisted of mobile phases A (0.1% formic acid in water) and B (0.1%
6
formic acid in acetonitrile). Gradient elution was adopted to obtain the whole profile
7
of plasma metabolites. The gradient started with 1% B, then, 0-3 min, B: 1%-52%;
8
3-7 min, B: 52%-74%; 7-9 min, B: 74%-80%; 9-10 min, B: 80%-90%; 10-12 min, B:
9
90%-99%; 12-16 min, B: 99%-99%; 16-17min, B: 99%-1%; 17-20 min, B: 1%-1%.
cr
us
an
2.4 Metabonomics data processing
M
10
ip t
1
Multivariate data analysis, including principal component analysis (PCA) and
12
partial least squares-discriminant analysis (PLS-DA), was used to identify the plasma
13
metabolites responsible for the differentiation of the treatment groups (NS, LA, HA).
14
Multivariate data analysis was performed with SIMCA-P+ 11.5 software (Umetrics
16 17 18 19
te
Ac ce p
15
d
11
AB, Umea, Sweden). In our present study, metabolites that were far from the cluster in the s-plot and loading plots and with a VIP value >1.0 were chosen as biomarkers. An independent sample t-test was performed using SPSS 17.0 software to determine whether or not the chosen biomarkers were significantly changed. 2.5 Characterisation of biomarkers
20
Candidates of metabolites were searched from human metabolome database
21
(HMDB, http://www.hmdb.ca/) utilising the detected molecular weights. The
22
biomarkers were unambiguously characterised by comparison with reference
5 Page 5 of 21
1
standards or MS/MS fragment information. The perturbation was interrogated using
2
MetPA
3
important tool in metabonomics pathway analysis.
4
3. Results and discussion
5
3.1 Clinical chemistry and histopathological examination
which
was
an
cr
ip t
(http://metpa.metabolomics.ca./MetPA/faces/Home.jsp),
The blood urea nitrogen (BUN) level significantly increased in rats exposed to the
7
high dose of ACV but showed no obvious changes in rats administered the low dose
8
of ACV (Fig 1A). Serum creatinine (Scr) activity significantly increased by both high
9
and low doses of ACV (Fig 1B). The detailed data of clinical chemistry was shown in
10
Table S1. BUN and Scr are a generally conventional monitor of nephrotoxicity and
11
have been used as a standard determiner of kidney injury for many years. When BUN
12
and Scr level significantly elevate, it shows the kidney has been injured.
te
d
M
an
us
6
As shown in Fig. S1, the high dose of ACV caused apparent tubular dilation,
14
irregular arrangement of cells, and development of interstitial inflammatory cells.
15 16 17 18 19
Ac ce p
13
Compared with the HA group, the LA group showed milder kidney damage, such as apparent tubular dilation, tubular epithelial edema, and irregular arrangement of cells. Coupled with the results from the biochemical analysis, ACV-induced nephrotoxicity could be confirmed in rats in the HA and LA groups. 3.2 Metabolic profiling analysis
20
Typical total ion current (TIC) chromatograms of plasma samples were obtained
21
from rats in the control and ACV-dosed groups in positive mode. Some discrimination
22
was found in the TIC chromatograms in positive mode between NS, HA and LA
6 Page 6 of 21
groups (Fig. S2). PCA and PLS-DA was further applied to better visualize the
2
discrimination among the gained complex data. Fig 2A demonstrated distinctions
3
between the different groups by PCA. The samples in the low-dose ACV group were
4
located near those of the control group, whereas samples from the high-dose ACV
5
group were found far from those of the control group. These data indicated that the
6
nephrotoxicity induced by ACV was dose-dependent. The differentiation by PLS-DA
7
between the LA and NS and HA and NS groups are shown in Fig 2B and 2C. Based
8
on these scores plots, the rat plasma showed clear changes due to ACV-administration.
9
The S-plot and loading plot of PLS-DA model between HA and NS group were shown
10
in Fig 2D and 2E. The ions in the top or bottom of S-plot or the variables located far
11
from the middle cluster in the loading plot most likely corresponded to the
12
representative of high score values in VIP plot[12]. It was these ions and variables
13
that mostly contributed to the significant changes between the different groups, and
14
these were selected as the potential biomarkers.
16 17 18
cr
us
an
M
d
te
Ac ce p
15
ip t
1
3.3 Identification of potential biomarkers From multivariate statistical analysis, these specific metabolites stood out from
the large amount of plasma endogenous metabolites. Meanwhile, the identification of the m/z of metabolites can be achieved. We searched candidates from biochemical
19
databases of HMDB, http://www.hmdb.ca/. We ignored the candidates from the
20
database identified as exogenous and focused on endogenous metabolites. In this
21
study, 16 metabolites were identified as potential biomarkers of nephrotoxicity in the
22
rat plasma. The metabolites /biomarkers identified are summarized in Table 1. Urea,
7 Page 7 of 21
proline, creatinine and arachidonic acid were identified by comparison with reference.
2
The residual 12 metabolites were identified by their MS/MS fragmentation. The
3
fragment information of metabolites is provided in Table 1. The heatmap helped to
4
visualise the biomarkers responsible for the differentiation of the three groups (Fig.
5
S3). An example of the characterization of the biomarker, lysophosphatidylcholine
6
(16:0) (tR=9.92min, m/z 496.3405), is shown in Fig. S4. Among the selected
7
biomarkers, some of them belonged to lysophosphatidylcholine (LPC). In blood
8
plasma significant amounts of lysophosphatidylcholine are formed by the enzyme,
9
phospholipase A2. This enzyme catalyzes the transfer of the fatty acids of position
10
sn-2 of phosphatidylcholine to the free cholesterol in plasma, with formation of
11
cholesterol esters and lysophosphatidylcholine[13].
12
3.4 Relationship between biomarkers and nephrotoxicity
te
d
M
an
us
cr
ip t
1
In this study, increased urea was detected in the HA dose group following
14
ACV-administration. Blood urea is excreted via the kidneys, hence increased levels of
15 16 17 18
Ac ce p
13
urea in the plasma can indicate impaired kidney function. The observed increased levels of urea implied that ACV administration has damaged the normal functions of the glomerulus [14-16]. The increased level of another biomarker in plasma, creatinine, also supported the damage in the normal functions of the glomerulus
19
caused by ACV, especially at the high dose [9,14]. The increased level of creatinine in
20
blood was also detected after cyclosporin A administration[17], indicating
21
drug-induced nephrotoxicity. Raised arachidonic acid, which is an essential
22
polyunsaturated fatty acid, has been shown previously to mediate inflammation
8 Page 8 of 21
[18-20]. This finding indicated that ACV has induced kidney injury through active
2
regulation of the inflammatory pathway. Proline is part of the arginine metabolism.
3
Tryptophan is an essential amino acid and the precursor of serotonin[12]. These
4
disturbed amino acid metabolism pathways indicated that ACV may affect normal
5
energy intake and consumption. Phytosphingosine is a phospholipid that is part of
6
sphingolipid metabolism[21], implying that ACV may disturb this metabolism.
7
Lysophosphatidylcholines (LPCs) were also found as possible biomarkers, which
8
were also identified in other nephrotic related work[22]. LPC control many biological
9
pathways, including cell proliferation [23], inflammation [24], and vasodilation [25].
cr
us
an
The change in LPC content may be a probable result of ACV-induced kidney injury.
M
10
ip t
1
From a metabolic level, ACV-induced nephrotoxicity of the biomarkers identified
12
by metabonomics may be related to damage to the glomerulus, inflammation,
13
abnormal energy intake and consumption, disturbed sphingolipid metabolism, and
14
changes in LPC.
16 17 18
te
Ac ce p
15
d
11
3.5 Disturbed metabolic pathways MetPA was applied for pathway analysis of selected biomarkers. The top four
metabolism pathways disturbed in the rats of the LA group were (a) arachidonic acid metabolism, (b) tryptophan metabolism, (c) arginine and proline metabolism, and (d)
19
glycerophospholipid metabolism (Fig. S5). In the HA group, pathways a–d were also
20
responsible for ACV-induced kidney injury. ACV may induce renal injury by initially
21
disrupting these four pathways.
22
4. Conclusion
9 Page 9 of 21
In the present study, two doses (50 and 100 mg/kg) of ACV were chosen, a kidney
2
injury model was built, and the dose-dependent effects of ACV on nephrotoxicity
3
were observed. An MS-based metabonomic approach was used to analyze changes in
4
the plasma. Sixteen biomarkers were identified to be related to ACV-induced
5
nephrotoxicity. The top four interrupted metabolic pathways included metabolism of
6
arachidonic acid, tryptophan, arginine and proline, and glycerophospholipid. The
7
method established in the present study provides useful information for ACV-induced
8
nephrotoxicity and may lay foundation for further toxic mechanism studies.
9
References
an
us
cr
ip t
1
[1] D. Lembo, S. Swaminathan, M. Donalisio, A. Civra, L. Pastero, D. Aquilano, P.
11
Vavia, F. Trotta, R. Cavalli, Encapsulation of Acyclovir in new carboxylated
12
cyclodextrin-based nanosponges improves the agent’s antiviral efficacy, Int J Pharm.
13
443 (2013) 262-272.
14
[2] N.N. Lam, M.A. Weir, Z. Yao, P.G. Blake, M.M. Beyea, T. Gomes, S. Gandhi, M.
16 17 18
d
te
Ac ce p
15
M
10
Mamdani, R. Wald, C.R. Parikh, D.G. Hackam, A.X. Garg, Risk of acute kidney injury from oral Acyclovir: a population-based study, Am J Kidney Dis. 61 (2013) 723-729.
[3] H. Yoon, K.Y. Rhew, Famciclovir as an antiviral agent for a patient with acute
19
renal failure, Int J Clin Pharm. 35 (2013) 173-175.
20
[4] D. Portilla, L. Schnackenberg, R.D. Beger, Metabolomics as an extension of
21
proteomic analysis: study of acute kidney injury, Semin Nephrol. 27 (2007) 609-620.
22
[5] X. Wang, B. Yang, H. Sun, A. Zhang, Pattern recognition approaches and
10 Page 10 of 21
computational systems tools for ultra performance liquid chromatography-mass
2
spectrometry-based comprehensive metabolomic profiling and pathways analysis of
3
biological data sets, Anal Chem. 84 (2012) 428-439.
4
[6] S. Rochfort, Metabolomics reviewed: a New “Omics” platform technology for
5
systems biology and implications for natural products research, J Nat Prod. 68 (2005)
6
1813-1820.
7
[7] J. Xuan, G. Pan, Y. Qiu, L. Yang, M. Su, Y. Liu, J. Chen, G. Feng, Y. Fang, W. Jia,
8
Q. Xing, L. He, Metabolomic profiling to identify potential serum biomarkers for
9
schizophrenia and risperidone action, J Proteome Res. 10 (2011) 5433-5443.
an
us
cr
ip t
1
[8] S. Rezzi, Z. Ramadan, L.B. Fay, S. Kochhar, Nutritional metabonomics:
11
applications and perspectives, J Proteome Res. 6 (2007) 513-525.
12
[9] R.D. Beger, J. Sun, L.K. Schnackenberg, Metabolomics approaches for
13
discovering biomarkers of drug-induced hepatotoxicity and nephrotoxicity, Toxicol
14
Appl Pharm. 243 (2010) 154-166.
16 17 18
d
te
Ac ce p
15
M
10
[10] C. Ma, K. Bi, M. Zhang, D. Su, X. Fan, W. Ji, C. Wang, X. Chen, Metabonomic study of biochemical changes in the urine of Morning Glory Seed treated rat, J Pharm Biomed Anal. 53 (2010) 559-566. [11] S. Peng, L. Yan, J. Zhang, Z. Wang, M. Tian, H. Shen, An integrated
19
metabonomics and transcriptomics approach to understanding metabolic pathway
20
disturbance induced by perfluorooctanoic acid, J Pharm Biomed Anal. 86 (2013)
21
56-64.
22
[12] X. Wang, H. Wang, A. Zhang, X. Lu, H. Sun, H. Dong, P. Wang, Metabolomics
11 Page 11 of 21
Study on the Toxicity of Aconite Root and Its Processed Products Using Ultrap-
2
erformance Liquid-Chromatography/ Electrospray-Ionization Synapt High-Definition
3
Mass Spectrometry Coupled with Pattern Recognition Approach and Ingenuity
4
Pathways Analysis, J Proteome Res. 11 (2012) 1284–1301.
5
[13] U.P. Steinbrecher, P.H. Pritchard, Hydrolysis of phosphatidylcholine during
6
LDLoxidation is mediated by platelet-activating factor acetylhydrolase, J Lipid Res.
7
30 (1989) 305-315.
8
[14] O. Shemesh, H. Golbetz, J.P. Kriss, B.D. Myers, Limitations of creatinine as a
9
filtration marker in glomerulopathic patients, Kidney Int. 28 (1985) 830-838.
an
us
cr
ip t
1
[15] M.W. Su, S.S. Chang, C.H. Chen, C.C. Huang, S.W. Chang, Y.C. Tsai, C.F. Lam,
11
Preconditioning renoprotective effect of isoflurane in a rat model of virtual renal
12
transplant, J Surg Res. doi: 10.1016/j.jss.2014.02.035.
13
[16] C.U. Niemann, N.J. Serkova, Biochemical mechanisms of nephrotoxicity:
14
application for metabolomics, Expert Opin Drug Metab Toxicol. 3 (2007) 527-544.
16 17 18
d
te
Ac ce p
15
M
10
[17] N.J. Serkova, U. Christians, Biomarkers for toxicodynamic monitoring of immunosuppressants: NMR-based quantitative metabonomics of the blood, Ther Drug Monit. 27 (2005) 733-737. [18] M. Shoeb, U.C. Yadav, S.K. Srivastava, K.V. Ramana, Inhibition of Aldose
19
Reductase Prevents Endotoxin-Induced Inflammation by Regulating Arachidonic
20
Acid Pathway in Murine Macrophages, Free Radic Biol Med. 51 (2011) 1686-1696.
21
[19] C. Pompeia, T. Lima, R. Curi, Arachidonic acid cytotoxicity: can arachidonic
22
acid be a physiological mediator of cell death? Cell Biochem Funct. 21(2003) 97-104.
12 Page 12 of 21
[20] M. Chen, M. Su, L. Zhao, J. Jiang, P. Liu, J. Cheng, Y. Lai, Y. Liu, W. Jia,
2
Metabonomic Study of Aristolochic Acid-Induced Nephrotoxicity in rats, J Proteome
3
Res. 5 (2006) 995-1002.
4
[21] M.T. Park, J.A. Kang, J.A. Choi, C.M. Kang, T.H. Kim, S. Bae, S. Kang, S. Kim,
5
W.I. Choi, C.K. Cho, H.Y. Chung, Y.S. Lee, S.J. Lee. Phytosphingosine induces
6
apoptotic cell death via caspase 8 activation and Bax translocation in human cancer
7
cells, Clin Cancer Res. 9 (2003) 878-885.
8
[22] C. Ma, K. Bi, D. Su, W. Ji, M. Zhang, X. Fan, C. Wang, X. Chen, Serum and
9
kidney metabolic changes of rat nephrotoxicity induced by Morning Glory Seed, Food
an
us
cr
ip t
1
Chem Toxicol. 48 (2010) 2988-2993.
11
[23] J. Yang, X. Zhao, X. Liu, C. Wang, P. Gao, J. Wang, L. Li, J. Gu, S. Yang, G. Xu,
12
High Performance Liquid Chromatography-Mass Spectrometry for Metabonomics:
13
Potential Biomarkers for Acute Deterioration of Liver Function in Chronic Hepatitis
14
B, J Proteome Res. 5 (2006) 554-561.
16 17 18
d
te
Ac ce p
15
M
10
[24] S. Koizumi, S. Yamamoto, T. Hayasaka, Y. Konishi, M. Yamaguchi-Okada, N. Goto-Inoue, Y. Sugiura, M. Setou, H. Namba, Imaging Mass Spectrometry Revealed The Production Of Lyso-Phosphatidylcholine In The Injured Ischemic Rat Brain, Neuroscience. 168 (2010) 219-225.
19
[25] T.D. Vuong, K.S. de, R.R. de, T.J. Rabelink, H.A. Koomans, J.A. Joles, Albumin
20
restores lysophosphatidylcholine-induced inhibition of vasodilation in rat aorta,
21
Kidney Int. 60 (2001) 1088-1096.
22
Figure Captions
13 Page 13 of 21
Fig 1. Effect of ACV on BUN and Scr levels. (A) Changes in BUN levels. (B)
2
Changes in Scr levels. Data represents mean ± SD (** p < 0.01, compared with the
3
control group).
4
Fig 2. Result of multivariate statistical analysis. (A) PCA score plot of the LA, HA
5
and NS groups. (B) PLS-DA score plot of the HA and NS groups. (C) PLS-DA score
6
plot of the LA and NS groups. (D) S-plot of PLS-DA model between the HA and NS
7
groups. (E) Loading plot of PLS-DA model between the HA and NS groups.
us
cr
ip t
1
Ac ce p
te
d
M
an
8
14 Page 14 of 21
1
Highlights
2
• Sixteen plasma metabolites were identified as changed following acyclovir
6 7 8 9
ip t
metabonomics.
cr
5
• Four disturbed metabolic pathways due to acyclovir were identified by
• Kidney injury model was built following intraperitoneal injection of two doses of
us
4
administration.
acyclovir.
• Metabonomics technology can provide information for toxic mechanism of
an
3
acyclovir.
M
10
Ac ce p
te
d
11
17 Page 15 of 21
Ac
ce
pt
ed
M
an
us
cr
i
*Graphical Abstract
Page 16 of 21
Identified potential biomarkers in different groups.
tR (min)
metabolite
obsd [M+H]+
calcd [M+H]+
1 2 3 4
0.89 0.92 0.93 5.17
Ureab Creatinineb Prolineb Tryptophanc
61.0398 114.0661 116.0704 205.0968
61.0396 114.0662 116.0706 205.0972
3.3 -0.9 -0.9 -2.0
CH4N2O C4H7N3O C5H9NO2 C11H12N2O2
↑ ↑ ↓ ↓
5
11.64
305.2474
305.2475
-0.3
C20H32O2
↑
LA and HA
6
7.99
Arachidonic acidb Phytosphingosine
-2.8
C18H39NO3
↓
LA and HA
7
12.66
LPC(15:0)c
8
9.46
9
9.92
formula
M an
318.3003
ce pt
c
318.2994
error (ppm)
us
no
The group identified this biomarkerd LA and HA LA and HA LA and HA LA and HA
ed
Table 1.
cr
ip t
Table(s)
Content variancea
482.3241
-1.7
C23H48NO7P ↑
LA and HA
LPC(16:1)c
494.3236
494.3241
-1.0
C24H48NO7P ↓
HA
LPC(16:0)c
496.3405
496.3398
1.4
C24H50NO7P ↓
LA and HA
Ac
482.3233
MS/MS
205.1 [M+H]+ 188.1 [M+H-NH2]+
318.3 [M+H]+ 300.2 [M+H-H2O]+ 256.2 [M+H-C2H5O2]+ 482.3 [M+H]+ 464.3 [M+H-H2O]+ 405.3 [M+H-C3H10NO]+ 299.3 [M+H-C5H13NO4P]+ 184.1 [M+H-C18H35NO2]+ 494.3 [M+H]+ 476.3 [M+H-H2O]+ 184.1 [M+H-C19H35NO2]+ 125.0 [M+H-C22H42NO3]+ 104.0 [M+H-C20H40NO4P]+ 496.3 [M+H]+ 478.3 [M+H-H2O]+
Page 17 of 21
ip t cr LPC(20:5)c
542.3223
542.3241
11
10.71
LPC(20:3)c
546.3556
546.3554
12
11.35
LPC(20:2)c
548.3706
13
12.98
LPC(20:1)c
14
10.40
15
12.98
-3.3
us
9.78
C28H48NO7P ↓
ed
M an
10
ce pt
548.3711
HA
0.4
C28H52NO7P ↓
HA
-0.9
C28H54NO7P ↓
LA and HA
550.3867
1.3
C28H56NO7P ↓
LA and HA
LPC(22:5)c
570.3556
570.3554
0.4
C30H52NO7P ↓
HA
LPC(22:4)c
572.3695
572.3711
-2.8
C30H54NO7P ↓
HA
Ac
550.3874
313.2 [M+H-C5H13NO4P]+ 184.1 [M+H-C19H37NO2]+ 104.1 [M+H-C20H42NO4P]+ 542.3 [M+H]+ 524.3 [M+H-H2O]+ 259.1 [M+H-C17H33NO2]+ 185.1 [M+H-C19H36NO3P]+ 126.0 [M+H-C26H42NO3]+ 546.4 [M+H]+ 528.4 [M+H-H2O]+ 184.1 [M+H-C18H36NO4P]+ 125.1 [M+H-C26H46NO3]+ 548.4 [M+H]+ 530.4 [M+H-H2O]+ 184.1 [M+H-C23H41NO2]+ 149.1 [M+H-C24H48NO3]+ 104.1 [M+H-C24H46NO4P]+ 550.4 [M+H]+ 532.4 [M+H-H2O]+ 185.1 [M+H-C23H43NO2]+ 105.1 [M+H-C24H48NO4P]+ 570.4 [M+H]+ 552.3 [M+H-H2O]+ 184.1 [M+H-C21H41NO3P]+ 125.0 [M+H-C28H47NO3]+ 572.4 [M+H]+ 554.4 [M+H-H2O]+ 258.2 [M+H-C13H32NO5P]+
Page 18 of 21
ip t LPC(22:2)c
576.4017
576.4024
-1.2
cr
13.42
C30H58NO7P ↓
a
M an
us
16
HA
184.1 [M+H-C21H42NO3P]+ 576.4 [M+H]+ 558.4 [M+H-H2O]+ 184.1 [M+H-C25H45NO2]+ 166.1 [M+H-C25H45O4]+ 104.1 [M+H-C26H50NO4P]+
↑, content increased; ↓, content decreased. bconfirmed by commercial standards. cconfirmed by MS/MS information. d the groups in
Ac
ce pt
ed
which biomarkers were identified.
Page 19 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Fig.1
Page 20 of 21
Ac
ce
pt
ed
M
an
us
cr
i
Fig.2
Page 21 of 21
