Journal of Analytical Methods in Chemistry

Journal of Analytical Methods in Chemistry / 2017 / Article

Research Article | Open Access

Volume 2017 |Article ID 6745932 |

Karolina Pietrowska, Diana Anna Dmuchowska, Paulina Samczuk, Tomasz Kowalczyk, Pawel Krasnicki, Malgorzata Wojnar, Aleksandra Skowronska, Zofia Mariak, Adam Kretowski, Michal Ciborowski, "LC-MS-Based Metabolic Fingerprinting of Aqueous Humor", Journal of Analytical Methods in Chemistry, vol. 2017, Article ID 6745932, 13 pages, 2017.

LC-MS-Based Metabolic Fingerprinting of Aqueous Humor

Academic Editor: Josep Esteve-Romero
Received07 Sep 2016
Accepted06 Dec 2016
Published05 Jan 2017


Aqueous humor (AH) is a transparent fluid which fills the anterior and posterior chambers of the eye. It supplies nutrients and removes metabolic waste from avascular tissues in the eye. Proper homeostasis of AH is required to maintain adequate intraocular pressure as well as optical and refractive properties of the eye. Application of metabolomics to study human AH may improve knowledge about the molecular mechanisms of eye diseases. Until now, global analysis of metabolites in AH has been mainly performed using NMR. Among the analytical platforms used in metabolomics, LC-MS allows for the highest metabolome coverage. The aim of this study was to develop a method for extraction and analysis of AH metabolites by LC-QTOF-MS. Different protocols for AH preparation were tested. The best results were obtained when one volume of AH was mixed with one volume of methanol : ethanol (1 : 1). In the final method, 2 µL of extracted sample was analyzed by LC-QTOF-MS. The method allowed for reproducible measurement of over 1000 metabolic features. Almost 250 metabolites were identified in AH and assigned to 47 metabolic pathways. This method is suitable to study the potential role of amino acids, lipids, oxidative stress, or microbial metabolites in development of ocular diseases.

1. Introduction

Metabolomics aims to identify and (semi-)quantify the small molecule metabolites present in a studied sample [1]. This approach has been extensively applied in biomedical research to find disease biomarkers [1], study therapeutic effects of drugs or natural substances with potential therapeutic capabilities [2], or explore metabolic pathways perturbed by a particular disease or condition [3]. Regarding the samples of mammalian origin, blood (serum/plasma) and urine are most commonly sampled for metabolomics. These types of biological fluids are especially important in the field of biomarkers discovery [1]. However, in studies aiming to explore changes in metabolic pathways evoked by the development of a disease, more specific biospecimens could be more informative. Such specific samples may provide more detailed information about disease pathogenesis and progression, toxicity, novel therapeutic targets, and response to drug administration, or about the effects of nutrition and exercise [4]. Just in the case of cancer has metabolomics been applied to study lung, kidney, prostate, gastric, colorectal, ovarian [5], liver, brain [6], and breast cancer tissues [7]. Among other types of mammalian samples metabolomics studies on cerebrospinal [8] or amniotic [9] fluids, exhaled breath condensate [10], stool, or saliva [11] samples were performed.

One of the human fluids in which the composition of small molecules has not been extensively explored using a metabolomics approach is aqueous humor (AH). AH is a transparent fluid produced by the ciliary body that passes from the posterior chamber through the pupil into the anterior chamber, where it drains out of the eye [12]. It supplies nutrients and removes metabolic waste from avascular tissues in the eye [13]. Homeostasis of AH is required to maintain adequate intraocular pressure as well as optical and refractive properties of the eye [14]. AH is a complex mixture known to contain electrolytes, glucose, urea, antioxidants (e.g., glutathione and ascorbic acid), organic solutes, several proteins (e.g., growth factors and cytokines), and oxygen and carbon dioxide [12, 14]. Until now, only a few articles in which metabolomics has been applied to analyze AH have been published [12, 1518]. Most of those studies were performed on animal models and with the use of nuclear magnetic resonance spectroscopy (NMR) [15–18]. The group of Song et al. have studied the metabolic changes in rabbit AH after glucocorticosteroids administration [15] and the hypotensive effects of glycyrrhizin on a rabbit model of ocular hypertension induced by triamcinolone acetonide [16]. 1H NMR has also been used to study metabolic alterations in AH in the rat glaucoma model induced by intracameral sodium hyaluronate injections [17] and to evaluate the effect of UV-A and UV-B irradiation on the metabolic profile of aqueous humor in rabbits [18]. Currently there is only one metabolomics study on human aqueous humor. In this research, a mass spectrometry (MS) based metabolomics approach was used to explore changes in metabolites observed in patients with different severities of myopia [12]. Metabolomics has a huge potential for studying human AH. It may improve knowledge about the molecular mechanisms of several eye diseases (e.g., cataract, glaucoma, pseudoexfoliation syndrome, or age-related macular degeneration) and even indicate metabolic pathways which can be promising drug targets. Moreover, like tissue metabolomics, sampling AH offers benefits of specificity over other biofluids to enable better, more sensitive characterization at the site of the disease.

MS and NMR are the analytical platforms that dominate within the field of metabolomics. The concept of global analysis of small molecules was first proposed in 1999 by Nicholson et al. who used NMR for this purpose [19]. NMR is a highly selective, nondestructive technique; however it offers lower sensitivity in comparison to MS. On the other hand, detection of metabolites using MS is usually preceded by their separation using liquid chromatography (LC), gas chromatography (GC), or capillary electrophoresis (CE). All three separation techniques in combination with MS detection are used in metabolomics studies [20]. GC-MS is applied to measure volatile and thermally stable analytes and CE-MS is applied for polar and charged molecules, while LC-MS is applied for polar and nonpolar compounds (depending on the type of column used) [21]. Separation step reduces the complexity of the biological sample and allows the analysis of different classes of metabolites at different time points [22]. Among the analytical platforms utilized in metabolomics, LC-MS allows for the highest metabolome coverage [1]. The first technique allowed for detection of about 20 metabolites in AH [16], while, using CE-MS and LC-MS, tens and hundreds of metabolites were detected, respectively [12]. The number of metabolites detected by LC-MS depends on the sample extraction procedure and chromatographic conditions used [23].

The aim of this study was to develop a strategy for the extraction of metabolites from AH and their analysis by LC-QTOF-MS which allows for detection of the highest number of metabolites by a single analytical platform. Considering similarity between AH and plasma [14] reversed phase chromatography was selected.

2. Materials and Methods

2.1. Samples

AH samples were collected from the patients undergoing cataract surgery. The study was approved by the Medical Ethics Committee of the Medical University of Bialystok and conformed to the tenets of the Declaration of Helsinki. Collection of the samples was performed following participants’ written informed consent for participation in the study. The anterior chamber of the eye was punctured using a 30 G needle and approximately 100–200 μL of AH was aspirated, transferred to Eppendorf® tubes (Eppendorf, Hamburg, Germany), frozen, and stored at −80°C until the day of analysis.

2.2. Chemicals and Reagents

Purified water was obtained using the Milli-Q Integral 3 system (Millipore SAS, Molsheim, France). LC-MS grade methanol, acetonitrile, formic acid and LC grade acetone, and ethanol were purchased from Sigma-Aldrich Chemie GmbH, Steinheim, Germany. The API-TOF reference mass solution kit (G1969-850001) and tuning solutions, ESI-L low concentration tuning mix (G1969-85000), and ESI-TOF Biopolymer Analysis reference masses (G1969-850003) were purchased from Agilent Technologies, Santa Clara, California, USA.

2.3. Sample Treatment and Analysis

On the day of analysis, AH samples were defrosted on ice. Protein precipitation and metabolite extraction were performed by vortex-mixing (IKA-Werke GmbH & Co. KG, Staufen, Germany) (1 min) one volume of the defrosted sample with the addition of one volume of freeze cold (−20°C) methanol/ethanol (1 : 1) mixture. After extraction, samples were stored on ice for ten minutes and centrifuged (Eppendorf, Hamburg, Germany) at 21,000 ×g for 10 minutes at 4°C. The supernatant was filtered through a 0.22 μm nylon filter (ThermoFisher Scientific, Waltham, Massachusetts, USA). Blank extraction was also prepared and analyzed together with AH samples. Blank extraction was analyzed in triplicate, while analysis of AH sample was five times repeated.

Samples were analyzed by an LC-MS system consisting of 1290 Infinity LC (Agilent, Santa Clara, California, USA) with a degasser, two binary pumps, and a thermostated autosampler coupled to a 6550 Q-TOF-MS detector (Agilent, Santa Clara, California, USA). Analyses were performed in ESI+ and ESI− modes, whereby 2 μL of extracted AH samples was injected into a thermostated (30°C) RP Poroshell 120 EC-C18, 3.0 × 100 mm, 2.7 μm column (Agilent Technologies, Santa Clara, California, USA). The flow rate was 0.5 mL/min with solvent A (water with 0.1% formic acid) and solvent B (acetonitrile with 0.1% formic acid). The gradient started at 1% phase B for the first minute, followed by an increase of phase B to reach 100% in 10 min. After reaching 100% the gradient returned to starting conditions (1% of phase B) in 0.1 min and was maintained at this solvents proportion for 4.9 min in order to reequilibrate the system for the next injection. During the method development longer gradients (12.5, 15, and 20 min) were also tested.

The mass spectrometer was operated in full scan mode from m/z 50–1000. The capillary voltage was set to 3 kV; the drying gas flow rate was 12 L/min at 250°C and gas nebulizer at 45 psig; fragmentor voltage was 225 V for positive and 275 V for negative ionization mode. Data was collected in centroid mode at a scan rate of 2 spectra per second. Accurate mass measurements were obtained by means of calibrant solution delivery using a dual-nebulizer ESI source. A calibrating solution (G1969-85000) containing reference masses at m/z 121.0509 (protonated purine) and m/z 922.0098 (protonated hexakis (1H,1H,3H-tetrafluoropropoxy)phosphazine or HP-921) in positive ion mode or m/z 119.0363 (proton abstracted purine) and m/z 966.0007 (formate adduct of HP-921) in negative ion mode was continuously introduced by an isocratic pump (Agilent, Santa Clara, California, USA) at a flow rate of 0.5 mL/min (1 : 100 split).

2.4. Data Treatment

The raw data collected by the analytical instrumentation was cleaned of background noise and unrelated ions by the Molecular Feature Extraction (MFE) tool in Mass Hunter Qualitative Analysis Software (B.06.00, Agilent, Santa Clara, California, USA). The MFE algorithm uses the accuracy of the mass measurements to group ions related by charge-state envelope, isotopic distribution, and/or the presence of adducts and dimers. The MFE then creates a list of all possible components as represented by the full TOF mass spectral data. Each compound is described by mass, retention time (RT), and abundance. The limit for the background noise was set to 1000 counts for data extraction by MFE and the following adduct settings were applied to identify coeluting adducts of the same feature: +H, +Na, +K in positive ion mode and −H, +HCOO, +Cl for negative ion mode. Dehydration neutral losses were also allowed in both ionization modes. The sample alignment and data filtering were performed using Mass Profiler Professional 12.6.1 (Agilent, Santa Clara, California, USA). Parameters applied for the alignment were 1% for RT and 10 ppm for the mass variation. Data obtained in every step of the study were filtered to keep metabolic features present in every analysis of blank extract and in 100% of the last three analyses of extracted AH sample.

2.5. Metabolite Identification

Identification of compounds detected by LC-MS was performed as follows. Accurate masses of features were searched against the METLIN, KEGG, LIPIDMAPS, and HMDB databases, which were simultaneously accessed by CEU Mass Mediator (available on-line search engine, Putative identities were then confirmed by LC-MS/MS using a QTOF (Agilent, Santa Clara, California, USA). Ions corresponding to putatively identified metabolites were targeted for collision-induced dissociation (CID) fragmentation based on the previously determined exact mass and RT. Primary identification was achieved by matching accurate mass and isotopic distribution of the targeted ion. In the second step of identification MS/MS fragmentation spectra were studied to elucidate the structure of the fragmented molecules or were compared with the spectral data of reference compounds available at public databases (HMDB, METLIN, and LIPIDMAPS).

3. Results and Discussion

3.1. Optimization of the Experimental Conditions

The purpose of this study was to develop a method for metabolic fingerprinting of human AH. LC-MS was selected as the analytical platform which allows measurement of the largest number of metabolites. However, during the method development not only the number of metabolites detected but also the reproducibility of measured intensities of metabolic features and the number of features in the blank extractions were taken into account. AH is to some extent similar to blood plasma as it contains plasma proteins and metabolites which are filtered to AH through fenestrated capillaries of the ciliary body stroma via the iris root [14]. Although the total concentration of proteins in AH seems to be low (five times lower than in blood plasma [14]), protein depletion in AH sample was performed before metabolite analysis. This step was used to eliminate potential enzymatic reactions between proteins and metabolites. The lower complexity of the sample is also advantageous as it minimizes matrix effect compared to the analysis of blood plasma, which may affect chromatographic resolution and ionization of metabolites in the ion source [29]. Moreover, injection of the sample containing proteins into LC system with acetonitrile or alcohol as the organic mobile phase may lead to system clogging by precipitated proteins [30]. Consequently, in the first step, different solvents and their mixtures commonly used for simultaneous proteins precipitation and metabolites extraction [31] were tested. One volume of AH was mixed with one volume of the following solutions: acetone, acetonitrile, methanol : ethanol (1 : 1), and methanol/acetone/acetonitrile (1 : 1 : 1). Extractions were performed as described in Materials and Methods. AH extracts and blank extraction samples were analyzed with a 20 min gradient in both ion modes. As it can be seen in Table 1 the numbers of metabolic features detected in each extract were similar. The highest number was detected in the acetone extract (1200 metabolic features), while the lowest was detected in acetonitrile extract (1102 metabolic features). Reproducibility of the method was also evaluated by calculation of coefficient of variation (CV) for intensity of each detected feature. The results were presented as the percentage of the features with CV ≤ 20% or CV ≤ 30%. The most reproducible results (92.8% of metabolic features with coefficient of variation (CV) < 30% and 87.9% with CV < 20%) were obtained with methanol : ethanol extraction. Moreover, this extraction allowed for detection of 1165 metabolic features and was selected as appropriate for AH preparation. In the next step different volumes of methanol : ethanol for protein precipitation and metabolite extraction were tested. Samples were prepared by mixing 1 volume of AH with 0.5, 1, 2, or 3 volumes of methanol : ethanol and analyzed with a 20 min gradient. The lowest number (about one thousand) of metabolic features was detected in the samples prepared in the ratio of AH to methanol : ethanol equal to 2 : 1. The other samples (ratios 1 : 1, 1 : 2, and 1 : 3) gave similar results (from 1074 to 1184 metabolic features); however, the highest reproducibility (94.1% of metabolic features with CV < 30% and 89.8% with CV < 20%) was obtained with 1 : 1 AH to solvent ratio. Moreover, as can be seen in Figure 1, chromatograms recorded for samples prepared in the ratio 1 : 1 were the most intense, which can be crucial for identification of metabolites by MS/MS fragmentation. Consequently, mixing one volume of sample with one volume of methanol : ethanol was selected as an adequate ration for preparation of AH for LC-MS-based metabolic fingerprinting. To decrease the time and consequently the costs of analyses, attempts to shorten the time of analysis were undertaken. The slope of the gradient was increased and different methods in which phase B reached 100% in 10, 12.5, or 15 min were tested. As can be seen in Table 2, the number of metabolic features detected and their repeatability do not differ significantly between the tested methods. The use of a shorter method allows for analysis of larger number of the samples in one sequence, if necessary, offering the potential of this method as a high-throughput screening approach for ocular diseases.

Extraction solutionsNumber of metabolic features detected⁢The % of metabolic features measured with given reproducibility
CV < 30%CV < 20%


In mixed solutions the solvents were used in equal proportions. The numbers of metabolic features detected do not include features present in the blank extraction. Presented results include data from both ion modes.

Method Number of metabolic features detectedThe % of reproducibly measured metabolic features
TotalCV < 30%CV < 20%CV < 30%CV < 20%

15 min14451346128393.1%88.8%
17.5 min13111226118893.5%90.6%
20 min14071319130093.7%92.4%

The numbers of metabolic features detected do not include features present in the blank extraction. Presented results include data from both ion modes.
3.2. Identification of Metabolites

The final method for AH fingerprinting with LC-MS allowed for detection of more than one thousand metabolic features with good repeatability. To prove that detected metabolites are not artifacts, metabolic features measured with CV < 20% in three AH analyses were searched against available internet databases. The list of putatively identified features included almost 250 different metabolites. Compounds identified as drugs (or drug metabolites) which were not taken by the patients in addition to phytometabolites were not included in that list. To give an overview of the metabolic pathways to which putatively identified metabolites belong, a pathway analysis was performed using MetaboAnalyst 3.0 (Figure 2 and Table 3). In total, these metabolites could be assigned to 47 metabolic pathways. Based on the pathway enrichment analysis ( values) and pathway topology analysis (pathway impact values) the importance of each metabolic pathway can be established [32]. As can be seen in Figure 2, based on metabolites detected in AH, phenylalanine metabolism and taurine and hypotaurine metabolism are the most significant metabolic pathways observed using this method. The value of other pathways is also represented on Figure 2, where 20 of the most significant pathways are marked. The complete list of metabolic pathways to which putatively identified metabolites from AH belong is presented in Table 3. The table includes the number of metabolites present in the pathway and detected in AH as well as results of pathway analysis ( value and pathway impact value).

PathwayNumber of metabolites in pathwayNumber of metabolites detected in AH valuePathway impact

Nitrogen metabolism3960.0343240
Aminoacyl-tRNA biosynthesis7580.0224820.05634
Phenylalanine, tyrosine, and tryptophan biosynthesis2740.327930.008
Valine, leucine, and isoleucine biosynthesis2740.218090.06784
Phenylalanine metabolism4550.159790.28993
Taurine and hypotaurine metabolism2030.0597510.35252
Biotin metabolism1120.0734150.13008
Arginine and proline metabolism7760.107560.18287
Pantothenate and CoA biosynthesis2730.829290.18014
Beta-alanine metabolism2830.808690.06625
Glycine, serine, and threonine metabolism4840.0586610.00118
Glyoxylate and dicarboxylate metabolism5040.768780.14716
Tryptophan metabolism7950.225620.18481
Pyrimidine metabolism6040.698730.02251
Alanine, aspartate, and glutamate metabolism2420.314270.20703
Ascorbate and aldarate metabolism4530.231760.11744
Sphingolipid metabolism2520.526650.09061
D-Arginine and D-ornithine metabolism810.777370
Tyrosine metabolism7640.015380.04724
Glycolysis or gluconeogenesis3120.0801310.09576
Pentose phosphate pathway3220.433660.02181
Vitamin B6 metabolism3220.0052520.0303
Glycerolipid metabolism3220.301380.0412
Pyruvate metabolism3220.407180.28201
D-Glutamine and D-glutamate metabolism1110.327930.02674
Methane metabolism3420.759860.01696
Glycosylphosphatidylinositol- (GPI-) anchor biosynthesis1410.489740.0439
Valine, leucine, and isoleucine degradation4020.327930.06442
Sulfur metabolism1810.336730.03307
Citrate cycle (TCA cycle)2010.174270.09024
Caffeine metabolism2110.327930
Selenoamino acid metabolism2210.355010.00321
Thiamine metabolism2410.878810
Lysine biosynthesis3210.859270.09993
Terpenoid backbone biosynthesis3310.829290
Ubiquinone and other terpenoid-quinone biosynthesis3610.015380
Glutathione metabolism3810.560910.03608
Glycerophospholipid metabolism3910.964540.12753
Butanoate metabolism4010.0959180.08516
Nicotinate and nicotinamide metabolism4410.0024840
Histidine metabolism4410.2580.13988
Primary bile acid biosynthesis4710.54410.00822
Lysine degradation4710.592720.14675
Purine metabolism9220.709890.00969
Pentose and glucuronate interconversions5310.740980
Cysteine and methionine metabolism5610.808690.01649
Amino sugar and nucleotide sugar metabolism8810.864130.00265

Pathway impact and p value were obtained from metabolic pathway analysis performed with MetaboAnalyst 3.0.

Identification of metabolites forwarded for pathway analysis was putative that, according to the Metabolomics Standards Initiative (MSI), is the lowest level of metabolite identification [33]. To increase the confidence of identification to level 2 [33], MS/MS fragmentation spectra of putatively identified signals were acquired. Obtained fragmentation spectra were compared to spectra available in databases (METLIN, HMDB) or in the literature [3, 34, 35]. Based on that information, the identity of over fifty metabolites was confirmed. Identified compounds are summarized in Tables 46 including retention time, theoretical mass and error of measured mass, MS/MS fragments, average intensity, and calculated coefficient of variation. Except metabolites four pharmacological substances (Table 4) present in eye drops or eye gels, which were administrated to the patients, were also detected in AH. Identified metabolites could be grouped into lipid compounds (fatty acids, acylcarnitines, sphingolipids, and fatty acid amides) and amino acids which are included in Table 5. The rest of the identified metabolites are summarized in Table 6.

CompoundRT [min]Monoisotopic mass [Da]Mass error [ppm]FragmentsAbundance [counts]CV [%]

Dexpanthenol3.1205.13145P: 188.128, 170.117, 102.054, 76.075, 74.023, 69.069, 58.0640
3.1205.13141N: 174.114, 156.103, 126.092, 102.056, 74.061, 72.046, 71.05, 44.0141
Timolol4.4316.15692P: 261.102, 244.076, 188.048, 74.06, 56.0490
Tropicamide4.1284.15253P: 267.149, 255.149, 135.0921
Proparacaine4.8294.19433P: 222.113, 178.087, 136.039, 100.1121

P or N indicates polarity in which metabolite was detected, positive or negative, respectively. Abundance is a value representing metabolite intensity calculated by MFE algorithm.

CompoundRT [min]Monoisotopic mass [Da]Mass error [ppm]FragmentsAbundance [counts]CV [%]Metabolic pathway or relation to eye diseases

Acetylcarnitine1.2203.11586P: 145.049, 85.028, 60.087FAO
Butyrylcarnitine3.4231.14714P: 173.08, 85.028, 60.084FAO
Hexanoylcarnitine4.6259.17842P: 99.082, 85.0272FAO
Octanoylcarnitine5.6287.20975P: 229.141, 127.11, 85.0291FAO
Decanoylcarnitine6.4315.2411P: 85.028, 60.0813FAO
Palmitic amide6.7255.25622P: 102.089, 88.075, 57.069, 43.0521Endocannabinoids metabolism
Stearamide7.6283.28751P: 88.07615Endocannabinoids metabolism
Hexadecasphinganine6.6273.26684P: 256.263, 106.086, 88.075, 57.0694Sphingolipids metabolism
Hydroxysphinganine6.7317.2931P: 300.29, 256.264, 88.075, 57.06915Sphingolipids metabolism
5 : 1 dicarboxylic fatty acid1.2130.02662N: 85.03, 41.03917Fatty acid metabolism
Suberic acid4.6174.08923N: 111.08, 83.0516Fatty acid metabolism
Nonanedioic acid5.1188.10493N: 169.091, 125.097, 97.0645Fatty acid metabolism
Sebacic acid5.6202.12053N: 183.101, 139.1124Fatty acid metabolism
Undecanedioic acid6.1216.13623N: 197.12, 153.133Fatty acid metabolism
Undecanedicarboxylic acid7.0244.16752N: 225.15, 181.161Fatty acid metabolism
Arginine1.0174.1175P: 158.093, 130.097, 116.071, 70.065, 60.0551Arginine and proline metabolism, urea cycle, Canavan disease, DEDs [24]
Glutamine1.0146.06914P: 130.049, 101.07, 102.054, 84.044, 56.049, 41.03819Nitrogen, arginine, proline, pyrimidine, alanine, aspartate, glutamate, glutamine and purine metabolism, aminoacyl-tRNA biosynthesis
1.0146.06911N: 128.036, 127.051, 109.041, 101.071, 99.056, 84.046, 74.025, 58.03, 41.9985
Histidine1.0155.06955P: 110.072, 93.044, 83.061Nitrogen, alanine, and histidine metabolism, aminoacyl-tRNA biosynthesis
Phenylalanine3.0165.0794P: 149.059, 131.048, 120.08, 103.0530Nitrogen and phenylalanine metabolism; aminoacyl-tRNA, phenylalanine, tyrosine, and tryptophan biosynthesis; DEDs [24]
3.0165.0797N: 147.045, 103.055, 72.0092
Tryptophan3.6204.08990N: 186.057, 159.093, 142.067, 116.05, 74.0241Nitrogen, tryptophan, glycine, serine, and threonine metabolism; aminoacyl-tRNA, phenylalanine, tyrosine, and tryptophan biosynthesis

P or N indicates polarity in which metabolite was detected, positive or negative, respectively. Abundance is a value representing metabolite intensity calculated by MFE algorithm. FAO: fatty acid oxidation; DEDs: dry eye disorders.

CompoundRT [min]Monoisotopic mass [Da]Mass error [ppm]FragmentsAbundance [counts]CV [%]Metabolic pathway or relation to eye diseases

Spermidine0.8145.15794P: 129.14, 112.111, 84.035, 75.091, 72.081, 58.0641Alanine metabolism
Hydroxyglutaric acid1.2148.03724N: 129.021, 103.04, 101.024, 85.031, 57.0356Butanoate metabolism
Aconitic acid1.2174.01640N: 129.021, 111.01, 85.03, 41.0414Glyoxylate and dicarboxylate metabolism
Citric acid1.2192.0271N: 129.02, 111.01, 87.01, 85.0315Glyoxylate and dicarboxylate metabolism, TCA
Ascorbic acid1.2176.03211N: 127.004, 115.004, 87.009, 71.014, 59.0145Glutathione, ascorbate and aldarate metabolism
Pyroglutamic acid1.2129.04263N: 82.0281Glutathione metabolism
Indoleacrylic acid3.6187.06336P: 170.06, 146.06, 142.065, 15.0547Microbial metabolism [25, 26]
Hydroxyphenyllactic acid3.7182.05793N: 163.04, 135.045, 119.05, 107.049, 92.99, 72.99, 44.9962Tyrosine metabolism
Phenylacetylglutamine4.1264.1110N: 145.062, 127.0520Phenylalanine metabolism
Trimethylamine N-oxide1.075.06846P: 58.065, 42.033, 30.0321Methane metabolism
p-Cresol4.6108.05754N: 106.042, 105.033, 92.028, 77.0382Degradation of aromatic compounds
p-Cresol sulfate4.65188.01431N: 107.05, 79.9572Degradation of aromatic compounds
Uric acid1.2168.02833P: 151.094, 141.046 Purine metabolism
1.2168.02835N: 124.016, 96.021, 69.01, 41.9990
Aminosalicylic acid or hydroxyanthranilic acid5.2153.04266P: 136.039, 108.044, 80.050Tryptophan metabolism
Quinic acid1.1192.06340N: 173.046, 127.041, 93.035, 85.03, 59.014, 44.9971Phenylalanine, tyrosine, and tryptophan biosynthesis
Choline1.0103.09972P: 60.081, 45.0332Phospholipid biosynthesis, DEDs [24]
Creatine1.1131.06957P: 90.0550Arginine and proline metabolism
Betaine1.0117.0796P: 59.073, 58.0652Methionine metabolism
Indole3.6117.05785P: 117.057, 100.112, 91.0540Tryptophan metabolism; phenylalanine, tyrosine, and tryptophan biosynthesis
Pantothenic acid3.2219.11072N: 146.081, 99.045, 88.04, 71.05, 44.01419Alanine metabolism; pantothenate and CoA biosynthesis
Hydroxybenzaldehyde4.5122.03683N: 120.022, 92.0271Degradation of aromatic compounds
Glycolic acid1.076.0163N: 72.994, 47.0137Glycine, serine, threonine, glyoxylate, and dicarboxylate metabolism; pentose phosphate pathway
Threonic acid1.0136.03720N: 117.02, 89.025, 75.009, 59.0141Ascorbate and aldarate metabolism
Hydroxypyruvic acid1.0104.0113N: 59.014, 41.002, 31.0190Glycine, serine, threonine, glyoxylate, and dicarboxylate metabolism
Glyceric acid1.1106.02662N: 75.009, 59.013, 56.9983Glycerolipid, glyoxylate, and dicarboxylate metabolism; pentose and glucuronate interconversion
Carnitine1.0161.10525P: 103.039, 102.091, 85.028, 60.081, 43.0171Glycine, serine, and threonine metabolism
Trigonellinamide1.0136.06374P: 94.064, 92.0491Nicotinate and nicotinamide metabolism
Pyrrolidonecarboxylic acid1.0129.04261P: 84.044, 56.049, 41.0383Glutathione metabolism
Taurine1.0125.01470N: 79.9583Nitrogen, taurine, and hypotaurine metabolism; primary bile acid biosynthesis
Acetylcholine1.1145.11037P: 87.0437, 60.08, 43.0171Phospholipid biosynthesis, DEDs [24]
Dimethylarginine1.1202.14305P: 158.129, 116.07, 88.086, 70.065, 46.0640Protein methylation process
Acetyl-histidine1.1197.08004P: 180.081, 156.076, 152.081, 110.071,1Nitrogen metabolism
Hypoxanthine1.2136.03852P: 119.035, 110.034, 94.039, 92.057, 82.039, 55.0283Purine metabolism
Lactic acid1.290.03171N: 43.01913Pyruvate metabolism, uveitis [27], ocular hypertension [28]
Glutamylleucine3.6260.13722N: 130.0893Nitrogen metabolism
Indoxylsulfuric acid4.2213.00963N: 132.046, 80.964, 79.9576Tryptophan metabolism

P or N indicates polarity in which metabolite was detected, positive or negative, respectively. Abundance is a value representing metabolite intensity calculated by MFE algorithm. TCA: tricarboxylic acid cycle; DEDs: dry eye disorders.
3.3. Relationship between Identified Metabolites and Ocular Abnormalities

Considering the number of metabolic pathways to which detected metabolites belong (Table 3 and Figure 2) and diversity of the identified metabolites (Tables 46), the proposed method for AH fingerprinting could be useful to study several eye diseases. Amino acids and lipids have already been linked with myopia [12], glaucoma [17], or ocular hypertension [16]. Proteomics studies on AH showed that APOC1 (one of apolipoproteins, proteins involved in lipid transport and metabolism) was overexpressed in patients with Coats’ disease [36], while another apolipoprotein (APOD) was found decreased in AH of patients with pseudoexfoliation syndrome [14]. Consequently, our method for AH fingerprinting could also be used to study metabolic changes in AH related to these diseases providing additional information about lipid metabolism in Coats’ disease and pseudoexfoliation syndrome. Among other metabolites detected in AH (Table 6) several molecules with antioxidative properties (e.g., citric acid, ascorbic acid, quinic acid, pantothenic acid, betaine, or taurine) or related to oxidative stress (e.g., pyroglutamic acid or hypoxanthine) were identified. Oxidative stress contributes to several eye diseases, which have been reviewed recently [37]. The knowledge about the role of oxidative stress in development and progression of these diseases could also be extended by application of this method. Some of the identified metabolites (e.g., p-cresol, p-cresol sulfate, or indoxylsulfuric acid) are known to be linked with gut microbiota [25, 26]. Currently there is only one scientific report indicating the contribution of gut bacteria to eye disease. Horai et al. demonstrated in a mouse model of spontaneous uveitis that a microbiota-dependent signal activates retina-specific T cells in the gut lamina propria that precedes clinical onset of the autoimmune uveitis, important cause of visual impairment in humans [38]. Metabolomics of AH may also help to evaluate a possible role of microbial metabolites in pathogenesis of eye diseases. The list of identified metabolites includes also such molecules which have already been reported as altered in different eye disorders. Changes in choline, acetylcholine, arginine, and phenylalanine levels were observed in tears of humans with dry eye disorders (DEDs) in comparison to control group [24], while alterations in lactate have been found in uveitis [27] or ocular hypertension [28].

4. Conclusions

Currently in only one study metabolic fingerprinting with LC-QTOF-MS was used for global measurement of metabolites in AH. In this report AH was 5 times diluted with water and 20 μL of the sample was analyzed by LC-QTOF-MS with reversed phase (RP) chromatographic separation [12]. In the present study for the first time several protocols for AH sample preparation before metabolic fingerprinting analysis were tested. Different solvents for simultaneous protein precipitation and metabolites extraction from AH were used. The best results were obtained with a mixture of methanol : ethanol (1 : 1). In the final method AH sample was two times diluted and 2 μL of extracted sample was analyzed using RP chromatography with QTOF-MS detection. Using the method, over 1000 metabolic features were reproducibly measured from which almost 250 were identified putatively and the identity of over 50 was confirmed by MS/MS fragmentation. Identified metabolites were assigned to 47 metabolic pathways. This method revealed the extent of the potential role of amino acids, lipids, oxidative stress, or microbial metabolites in development of eye diseases.

Competing Interests

The authors declare that there is no conflict of interests regarding the publication of this manuscript.


This work was supported by the Medical University of Bialystok, Poland (Grant no. N/ST/ZB/16/001/1157). Clinical Research Centre is a part of the Centre for Innovative Research—the Leading National Research Center in Poland (KNOW 2012–2017). This study was conducted with the use of equipment purchased by Medical University of Bialystok as part of the RPOWP 2007–2013 Funding, Priority I, Axis 1.1, Contract no. UDA-RPPD.01.01.00-20-001/15-00 dated 26.06.2015.


  1. R. Bujak, W. Struck-Lewicka, M. J. Markuszewski, and R. Kaliszan, “Metabolomics for laboratory diagnostics,” Journal of Pharmaceutical and Biomedical Analysis, vol. 113, pp. 108–120, 2015. View at: Publisher Site | Google Scholar
  2. J. Godzien, M. Ciborowski, S. Angulo et al., “Metabolomic approach with LC-QTOF to study the effect of a nutraceutical treatment on urine of diabetic rats,” Journal of Proteome Research, vol. 10, no. 2, pp. 837–844, 2011. View at: Publisher Site | Google Scholar
  3. M. Ciborowski, J. L. Martin-Ventura, O. Meilhac et al., “Metabolites secreted by human atherothrombotic aneurysms revealed through a metabolomic approach,” Journal of Proteome Research, vol. 10, no. 3, pp. 1374–1382, 2011. View at: Publisher Site | Google Scholar
  4. E. J. Want, P. Masson, F. Michopoulos et al., “Global metabolic profiling of animal and human tissues via UPLC-MS,” Nature Protocols, vol. 8, no. 1, pp. 17–32, 2013. View at: Publisher Site | Google Scholar
  5. D. B. Liesenfeld, N. Habermann, R. W. Owen, A. Scalbert, and C. M. Ulrich, “Review of mass spectrometry-based metabolomics in cancer research,” Cancer Epidemiology Biomarkers and Prevention, vol. 22, no. 12, pp. 2182–2201, 2013. View at: Publisher Site | Google Scholar
  6. R. Beger, “A review of applications of metabolomics in cancer,” Metabolites, vol. 3, no. 3, pp. 552–574, 2013. View at: Publisher Site | Google Scholar
  7. C. Denkert, E. Bucher, M. Hilvo et al., “Metabolomics of human breast cancer: new approaches for tumor typing and biomarker discovery,” Genome Medicine, vol. 4, no. 4, article 37, 2012. View at: Publisher Site | Google Scholar
  8. R. Mandal, A. C. Guo, K. K. Chaudhary et al., “Multi-platform characterization of the human cerebrospinal fluid metabolome: a comprehensive and quantitative update,” Genome Medicine, vol. 4, no. 4, article 38, 2012. View at: Publisher Site | Google Scholar
  9. M. Orczyk-Pawilowicz, E. Jawien, S. Deja et al., “Metabolomics of human amniotic fluid and maternal plasma during normal pregnancy,” PLoS ONE, vol. 11, no. 4, Article ID e0152740, 2016. View at: Publisher Site | Google Scholar
  10. A. Peralbo-Molina, M. Calderón-Santiago, F. Priego-Capote, B. Jurado-Gámez, and M. D. Luque de Castro, “Metabolomics analysis of exhaled breath condensate for discrimination between lung cancer patients and risk factor individuals,” Journal of Breath Research, vol. 10, no. 1, Article ID 016011, 2016. View at: Publisher Site | Google Scholar
  11. J. H. Wang, J. Byun, and S. Pennathur, “Analytical approaches to metabolomics and applications to systems biology,” Seminars in Nephrology, vol. 30, no. 5, pp. 500–511, 2010. View at: Publisher Site | Google Scholar
  12. C. Barbas-Bernardos, E. G. Armitage, A. García et al., “Looking into aqueous humor through metabolomics spectacles—exploring its metabolic characteristics in relation to myopia,” Journal of Pharmaceutical and Biomedical Analysis, vol. 127, pp. 18–25, 2016. View at: Publisher Site | Google Scholar
  13. Y. Ji, X. Rong, H. Ye, K. Zhang, and Y. Lu, “Proteomic analysis of aqueous humor proteins associated with cataract development,” Clinical Biochemistry, vol. 48, no. 18, pp. 1304–1309, 2015. View at: Publisher Site | Google Scholar
  14. A. A. Kliuchnikova, N. I. Samokhina, I. Y. Ilina et al., “Human aqueous humor proteome in cataract, glaucoma, and pseudoexfoliation syndrome,” Proteomics, vol. 16, no. 13, pp. 1938–1946, 2016. View at: Publisher Site | Google Scholar
  15. Z. Song, H. Gao, H. Liu, and X. Sun, “Metabolomics of rabbit aqueous humor after administration of glucocorticosteroid,” Current Eye Research, vol. 36, no. 6, pp. 563–570, 2011. View at: Publisher Site | Google Scholar
  16. Z. Song, Y. Gong, H. Liu, Q. Ren, and X. Sun, “Glycyrrhizin could reduce ocular hypertension induced by triamcinolone acetonide in rabbits,” Molecular Vision, vol. 17, pp. 2056–2064, 2011. View at: Google Scholar
  17. A. Mayordomo-Febrer, M. López-Murcia, J. M. Morales-Tatay, D. Monleón-Salvado, and M. D. Pinazo-Durán, “Metabolomics of the aqueous humor in the rat glaucoma model induced by a series of intracamerular sodium hyaluronate injection,” Experimental Eye Research, vol. 131, pp. 84–92, 2015. View at: Publisher Site | Google Scholar
  18. M.-B. Tessem, T. F. Bathen, J. Čejková, and A. Midelfart, “Effect of UV-A and UV-B irradiation on the metabolic profile of aqueous humor in rabbits analyzed by 1H NMR spectroscopy,” Investigative Ophthalmology and Visual Science, vol. 46, no. 3, pp. 776–781, 2005. View at: Publisher Site | Google Scholar
  19. J. K. Nicholson, J. C. Lindon, and E. Holmes, “‘Metabonomics’: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data,” Xenobiotica, vol. 29, no. 11, pp. 1181–1189, 1999. View at: Publisher Site | Google Scholar
  20. D. González-Peña, D. Dudzik, C. Colina-Coca et al., “Multiplatform metabolomic fingerprinting as a tool for understanding hypercholesterolemia in Wistar rats,” European Journal of Nutrition, vol. 55, no. 3, pp. 997–1010, 2016. View at: Publisher Site | Google Scholar
  21. M. Marcinkiewicz-Siemion, M. Ciborowski, A. Kretowski, W. J. Musial, and K. Kaminski, “Metabolomics—a wide-open door to personalized treatment in chronic heart failure?” International Journal of Cardiology, vol. 219, pp. 156–163, 2016. View at: Publisher Site | Google Scholar
  22. A. Alonso, S. Marsal, and A. Julià, “Analytical methods in untargeted metabolomics: state of the art in 2015,” Frontiers in Bioengineering and Biotechnology, vol. 3, article 23, 2015. View at: Publisher Site | Google Scholar
  23. O. Yanes, R. Tautenhahn, G. J. Patti, and G. Siuzdak, “Expanding coverage of the metabolome for global metabolite profiling,” Analytical Chemistry, vol. 83, no. 6, pp. 2152–2161, 2011. View at: Publisher Site | Google Scholar
  24. C. Galbis-Estrada, S. Martinez-Castillo, J. M. Morales et al., “Differential effects of dry eye disorders on metabolomic profile by 1H nuclear magnetic resonance spectroscopy,” BioMed Research International, vol. 2014, Article ID 542549, 7 pages, 2014. View at: Publisher Site | Google Scholar
  25. W. R. Wikoff, A. T. Anfora, J. Liu et al., “Metabolomics analysis reveals large effects of gut microflora on mammalian blood metabolites,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 10, pp. 3698–3703, 2009. View at: Publisher Site | Google Scholar
  26. E. Bone, A. Tamm, and M. Hill, “The production of urinary phenols by gut bacteria and their possible role in the causation of large bowel cancer,” The American Journal of Clinical Nutrition, vol. 29, no. 12, pp. 1448–1454, 1976. View at: Google Scholar
  27. S. P. Young, M. Nessim, F. Falciani et al., “Metabolomic analysis of human vitreous humor differentiates ocular inflammatory disease,” Molecular Vision, vol. 15, pp. 1210–1217, 2009. View at: Google Scholar
  28. Q. C. Ngumah, S. D. Buchthal, and R. F. Dacheux, “Longitudinal non-invasive proton NMR spectroscopy measurement of vitreous lactate in a rabbit model of ocular hypertension,” Experimental Eye Research, vol. 83, no. 2, pp. 390–400, 2006. View at: Publisher Site | Google Scholar
  29. C. Bylda, R. Thiele, U. Kobold, and D. A. Volmer, “Recent advances in sample preparation techniques to overcome difficulties encountered during quantitative analysis of small molecules from biofluids using LC-MS/MS,” Analyst, vol. 139, no. 10, pp. 2265–2276, 2014. View at: Publisher Site | Google Scholar
  30. H. Pereira, J.-F. Martin, C. Joly, J.-L. Sébédio, and E. Pujos-Guillot, “Development and validation of a UPLC/MS method for a nutritional metabolomic study of human plasma,” Metabolomics, vol. 6, no. 2, pp. 207–218, 2010. View at: Publisher Site | Google Scholar
  31. S. J. Bruce, I. Tavazzi, V. Parisod, S. Rezzi, S. Kochhar, and P. A. Guy, “Investigation of human blood plasma sample preparation for performing metabolomics using ultrahigh performance liquid chromatography/mass spectrometry,” Analytical Chemistry, vol. 81, no. 9, pp. 3285–3296, 2009. View at: Publisher Site | Google Scholar
  32. J. Xia and D. S. Wishart, “MetPA: a web-based metabolomics tool for pathway analysis and visualization,” Bioinformatics, vol. 26, no. 18, pp. 2342–2344, 2010. View at: Publisher Site | Google Scholar
  33. L. W. Sumner, A. Amberg, D. Barrett et al., “Proposed minimum reporting standards for chemical analysis: Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI),” Metabolomics, vol. 3, no. 3, pp. 211–221, 2007. View at: Publisher Site | Google Scholar
  34. M. Ciborowski, J. Teul, J. L. Martin-Ventura, J. Egido, and C. Barbas, “Metabolomics with LC-QTOF-MS permits the prediction of disease stage in aortic abdominal aneurysm based on plasma metabolic fingerprint,” PLoS ONE, vol. 7, no. 2, Article ID e31982, 2012. View at: Publisher Site | Google Scholar
  35. J. Piszcz, E. G. Armitage, A. Ferrarini et al., “To treat or not to treat: metabolomics reveals biomarkers for treatment indication in chronic lymphocytic leukaemia patients,” Oncotarget, vol. 7, no. 16, pp. 22324–22338, 2016. View at: Publisher Site | Google Scholar
  36. Q. Yang, H. Lu, X. Song, S. Li, W. Wei, and R. Yousefi, “iTRAQ-based proteomics investigation of aqueous humor from patients with coats' disease,” PLoS ONE, vol. 11, no. 7, Article ID e0158611, 2016. View at: Publisher Site | Google Scholar
  37. J. Kruk, K. Kubasik-Kladna, and H. Y. Aboul-Enein, “The role oxidative stress in the pathogenesis of eye diseases: current status and a dual role of physical activity,” Mini Reviews in Medicinal Chemistry, vol. 16, no. 3, pp. 241–257, 2015. View at: Google Scholar
  38. R. Horai, C. R. Zárate-Bladés, P. Dillenburg-Pilla et al., “Microbiota-dependent activation of an autoreactive T cell receptor provokes autoimmunity in an immunologically privileged site,” Immunity, vol. 43, no. 2, article 3135, pp. 343–353, 2015. View at: Publisher Site | Google Scholar

Copyright © 2017 Karolina Pietrowska et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Related articles

No related content is available yet for this article.
 PDF Download Citation Citation
 Download other formatsMore
 Order printed copiesOrder

Related articles

No related content is available yet for this article.

Article of the Year Award: Outstanding research contributions of 2020, as selected by our Chief Editors. Read the winning articles.