NMR-Based Metabolomics of Non-Human Primate Ocular Tissues: A Dataset

1. Summary

Metabolomics is a discipline dedicated to identifying and quantifying small molecules, known as metabolites, within biological samples such as cells, tissues, or fluids. These molecules encompass the initial, intermediate, and final products of cellular processes, collectively offering a snapshot of the metabolic state in a biological system. Metabolomics serves as a potent tool for comprehending intricate biological systems and plays a pivotal role in the creation of novel diagnostic, therapeutic, and environmental solutions.

Differences in metabolomic compositions signify alterations in specific metabolic pathways, offering insights that can contribute to the development of disease diagnosis methods and treatment strategies. However, obtaining human tissues, especially ocular tissues, poses challenges. Sampling of control human ocular tissues is often difficult or impossible, leading to a reliance on tissues (lens, cornea, aqueous (AH), and vitreous (VH) humors) obtained from cadavers several hours post-mortem in the majority of human metabolomic ophthalmological studies [1,2,3]. Anaerobic reactions and cell lysis caused by disruptions in cell energy status and osmoregulation lead to rapid and significant changes in the metabolomic composition of biological tissues and fluids, causing post-mortem bias in data [4,5,6,7].

An alternative approach to studying human eye diseases involves the use of experimental animals. Rodents, particularly mice and rats, are commonly employed in ophthalmological studies, especially those related to the lens [8,9,10]. However, a comparison of metabolomic compositions between human and animal lenses suggests that rodents offer limited suitability as models for human ophthalmological diseases. Earlier studies in our lab [2,3,9] have demonstrated substantial differences in the sets of cytoprotective compounds between human and rat lenses. This underscores the critical need to identify an animal model with a lens metabolomic composition closely resembling that of the human lens.

Monkeys and apes, being the species genetically closest to humans, exhibit anatomical structures and functional capabilities in their eyes that resemble those of humans. Given the similarity in the anatomy and functionality of the human and monkey eyes, nonhuman primates are often regarded as the optimal animal models for studying human eye diseases [11]. Primates have proven effective in investigating ocular diseases and developing treatment methods, encompassing areas such as amblyopia, cataract, and glaucoma [12,13,14,15]. However, the metabolomic composition of nonhuman primate ocular tissues remains largely unidentified. While occasional references exist regarding the presence and abundance of individual metabolites in monkey ocular tissues, such as UV filters in the monkey lens [16,17], a comprehensive quantitative metabolomic analysis is lacking. In this study, we conducted a thorough measurements of the quantitative metabolomic profiles of four tissues from a crab-eating macaque (Macaca fascicularis) and rhesus macaque (Macaca mulatta), including blood serum, lens, VH, and AH. The resulting quantitative metabolomic dataset holds significant relevance across a diverse range of fundamental and applied scientific fields, including applications in animal modeling of human diseases, evaluation of medical formulations, and evolutionary studies.

2. Data Description

The dataset consists of two main parts:

1. Raw instrumental data – spectra from ¹H Nuclear Magnetic Resonance facility (description of the acquisition protocol is provided below in the methods section).

2. Absolute concentrations of metabolites in samples. Quantitative data presented in the table (csv format), where the columns correspond to samples (n=45), and the rows correspond to metabolites (n=99). Concentrations in the table are given in nmol per gram of wet tissue weight.

3. Metadata describing samples.

2.1. Primate Specimen Description

The following specimen encompass the dataset:

The samples were collected from 18 animals, 15 specimens of Crab-eating macaque (Macaca fascicularis) and 3 specimens of Rhesus macaque (Macaca mulatta), and include intravital serum (7 samples), post-mortem serum (6 samples), lens (11 samples), AH (10 samples), and VH (11 samples). Table 1 summarizes the description of the samples used in the present study.

2.2. Raw Instrumental NMR Data and Metabolite Assignment

The metabolomic compositions of macaque tissues were determined using NMR method. Typical ¹H NMR spectra with annotated signals corresponding to identified metabolites are presented in Figure 1 (serum), Figure 2 (AH), Figure 3 (VH), and Figure 4 (lens).

2.3. Quantitative data overview

The majority of the data obtained correspond to young healthy specimens of Crab-eating macaque (M. fascicularis). These data can be considered as the basic metabolomic profiles of the corresponding tissues. The dataset also includes the data on specimens of Rhesus macaque (M. mulatta), and on old animals. This allows for the estimating of the influence of age and taxonomy on metabolic processes in the ocular tissues. The comparison of metabolomic profiles of the serum samples obtained intravital and 15 minutes post-mortem makes it possible to evaluate fast thermal and enzymatic post-mortem reactions occurring in the monkey body.

To overview the data on the concentrations of metabolites in the samples and gain insight into the dataset, we performed preliminary exploratory statistical analysis. Figure 5 shows a pie chart revealing the 10 top most abundant metabolites in the dataset across all samples in average (in four tissue types and two species) – Lactic acid, myo-Inositol, Glucose, Glutamine, Creatine, Glutamic acid, 3-Hydroxykynurenine O-β-D-Glucoside (3-OHKG, UV-filter [18]), Glycerol, Taurine, Alanine – comprising ca. 87% of the total metabolite content.

Figure 6 shows score plot of first two principal components from principal component analysis (PCA) which describes the dataset by revealing differences and similarities in samples depending on tissue type and species. The first principal component (PC1) accounts for 43.9% of the total variance, PC2 – 15.1%. The majority of the samples belonging to lenses of M. fascicularis are rather distant along PC1 from other tissues. Although there are two samples of M. fascicularis lenses that are much closer to M. mulatta lenses.

Heatmap showing relative metabolite abundances in the dataset as well as hierarchical clustering of both samples and metabolites is presented in Figure 7. Heatmap can be used to identify samples or metabolites that are unusually high or low in the dataset. Heatmap confirms observation from PCA that 6 lenses of M. fascicularis are rather distant from other tissues (upper right red rectangle).

3. Methods.

3.1. Sample collection

The research was carried out in line with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the European Union Directive 2010/63/EU on the protection of animals used for scientific purposes, and approved by bioethics committee of the Research Institute of Medical Primatology (Record 80/1 from 02.12.2021).

We obtained intravital (n=7) and post-mortem blood samples from Crab-eating macaques (Macaca fascicularis) (n=8) by venipuncture using a syringe with a 21G needle. Within 10-15 minutes after the blood collection, blood samples were centrifuged (3000×g, 10 min), and the plasma obtained was transferred into separate vials. The plasma samples were immediately frozen and kept at -70° C until analysis.

Euthanasia was performed for Crab-eating macaques (Macaca fascicularis) (n=8) and Rhesus macaques (Macaca mulatta) (n=3) by intravenous administration of 5.0 ml of Anestofol 5% (Interpharm, Russia) with preliminary general anesthesia by intravenous injection of 0.10 ml/kg of Xila (2% xylazine) (Interchemie Werken "de Adelaar" BV, the Netherlands ) and 0.05 ml/kg of Zoletil® (Virbac Sante Animale, France). The samples of AH were taken from the anterior chamber by the cornea puncture with the 21G needle. Then the cornea was cut along the iris border, the lens was extracted, and VH was collected with the 21G needle. The samples of AH, VH, and lens were frozen and kept at -70° C.

3.2. Sample preparation

The process of preparing the lens samples was carried out following the detailed procedure described in [19]. Typical lens weight was 90-100 mg, typical plasma volume was 300 µL, VH volume was 600 µL, AH volume was 70 to 200 µL. The lenses were homogenized in glass vials using a TissueRuptor II rotor-stator homogenizer (Qiagen, Venlo, Netherlands) with 1600 µL of cold MeOH (-20°C). AH, VH, and plasma samples were extracted without prior homogenization. The extraction was performed by water/methanol/chloroform mixture as described in [19]. The metabolomic extracts were vacuum-dried for further analysis.

3.3. NMR measurements

For NMR measurements, the dried extracts were re-dissolved in 600 μL of D₂O containing 2 × 10^–5 M of sodium 4,4-dimethyl-4-silapentane-1-sulfonic acid (DSS) as an internal standard and 50 mM of deuterated phosphate buffer (pH 7.2). The ¹H NMR measurements were conducted at the "Mass Spectrometric Investigations" Center of Collective Use, SB RAS using an AVANCE III HD 700 MHz NMR spectrometer (Bruker BioSpin, Rheinstetten, Germany). The concentrations of metabolites in the samples were determined by the peak area integration relatively to the internal standard DSS, and then recalculated into metabolite concentrations in the tissue (for the lens, in nmoles per gram of the tissue wet weight, and for AH, VH, and serum, in nmoles per milliliter). The exceptions were UV filters, for which LC-MS and LC-OD methods were used.

3.4. LC–MS and LC-OD Measurements

LC-MS and LC-OD analyses were performed as described in [3]. The metabolomic extracts were re-dissolved in 100 μL of 10 mM ammonium formate and 0.1% formic acid solution in H₂O. The hydrophilic interaction liquid chromatography (HILIC) method was used for LC separation of the samples, utilizing a TSKgel Amide-80 HR column (4.6 × 250 mm, 5 μm) on an UltiMate 3000RS chromatograph (Dionex, Germering, Germany). A diode array UV-vis detector (DAD, LC-OD) with a 190–800 nm spectral range was used with a flow cell. Concentrations of UV filters 3-Hydroxykynurenine (3-OHKN), 3-Hydroxykynurenine O-β-D-Diglucoside (3-OHKDG), 4-(2-Amino-3-Hydroxyphenyl)-4-Oxobutanoic acid O-β-D-Glucoside (AHBG), 4-(2-Amino-3-Hydroxyphenyl)-4-Oxobutanoic acid O-β-D-Diglucoside (AHBDG), Deaminated 3-OHKG (4-(2-Amino-3-Hydroxyphenyl)-4-Oxocrotonic acid O-β-D-Glucoside, 3-OHCKAG), and Methylated 3-OHKG (Me-3-OHKG) were measured by LC-OD at the 360 nm wavelength assuming that their absorption coefficients at 360 nm are similar, ε(360) = 4.5 × 10³ M^-1cm^-1 [3,20].

3.5. Identification and quantification metabolites

Metabolite identification was conducted by analyzing their NMR spectra, which were sourced from literature, databases (HMDB, METLIN, BMRB, and SpectraBase), and our in-house NMR library [2,21,22] and AMDB [23]. To identify unknown signals, we either used LC-OD, LC-MS, and LC-MS/MS or fractionated the metabolomic extract by LC and conducted NMR analysis on each fraction, as detailed in [21]. The baseline processing, identification, and integration of spectral NMR peaks (quantification) were done using the MestReNova v12.0 (Mestrelab Research, A Coruna, Spain). The final table contains 89 reliably identified and quantified metabolites. The raw data could contain more information on yet unidentified metabolites.