Proteomic analysis of giant panda testicular tissue of different age groups

Source of panda testicular tissues

The normal testicular tissue came from the frozen sample bank of Chengdu Research Base of Giant Panda Breeding (CRBGPB), located in Sichuan Province, PRC. The samples were derived from three male giant pandas: a 22-year-old male panda (H22y) who died of non-testicular causes in old age, an 18-year-old male panda (F18y) with normal tissue sections taken from removed testicles for testicular tumor, and an 8-day-old male panda (J8d), who accidentally died due to rearing issues. All samples were frozen in liquid nitrogen and then transferred to a −80 °C freezer for storage. All procedures in this study were approved by the Institutional Animal Care and Use Committee of CRBGPB (approval no. CRBGPB-2018008) and adhered to the ethical rules and regulations for animal sample use for the People’s Republic of China.

Total protein extraction

The testicular tissue samples from the giant pandas were retrieved from a −80 °C freezer, then cryogenically ground in liquid nitrogen and lysed using a lysis buffer containing 100 mM NH₄HCO₃ (pH 8), 6 M urea, and 0.2% SDS. Subsequently, the lysate underwent 5 min of ultrasonication on ice. Following ultrasonication, the lysate was centrifuged at 12,000 g for 15 min at 4 °C, and the resulting supernatant was carefully transferred to a clean tube. The extracts from each sample were then treated with 10 mM DTT^red (Sigma, St. Louis, MO, USA) for 1 h at 56 °C to achieve reduction, followed by alkylation with an adequate amount of iodoacetamide (Sigma, St. Louis, MO, USA) for 1 h at room temperature in the dark. Following alkylation, the samples were thoroughly mixed with four times the volume of pre-cooled acetone through vortexing and incubated at −20 °C for a minimum of 2 h. Subsequently, the samples underwent centrifugation, and the resulting precipitation was collected. After two washes with cold acetone, the pellet was dissolved in a dissolution buffer containing 0.1 M triethylammonium bicarbonate (TEAB, pH 8.5) and 6 M urea.

Protein quality test

We utilized the Bradford protein quantification kit (Bio-Rad, Hercules, CA, USA) to quantify the protein. Initially, we prepared a bovine serum albumin (BSA) standard protein solution as per the manufacturer’s instructions, with a concentration gradient ranging from 0 to 0.5 µg/µL. Both the BSA standard protein solutions and sample solutions, each at varying dilution multiples, were dispensed into a 96-well plate, with a final volume of 20 µL in each well. This process was repeated three times for each gradient. Subsequently, we swiftly added 180 µL of G250 dye solution to the plate and allowed the mixture to sit at room temperature for 5 min. The absorbance at 595 nm was then measured. We constructed a standard curve using the absorbance values of the standard protein solution and determined the protein concentration of the samples accordingly. For SDS-PAGE gel electrophoresis, we loaded 20 µg of the protein sample onto a 12% gel. The concentrated gel was subjected to electrophoresis at 80 V for 20 min, followed by the separation gel at 120 V for 90 min. The gel was subsequently stained with Coomassie Brilliant Blue R-250 (Amresco, Solon, OH, USA) and destained until the bands were visible.

Trypsin treatment

We took 500 µg of the protein sample and added protein dissolution solution to reach a total volume of 200 µL. Add 5 µL of 1 µg/µL trypsin and one mL of 50 mM TEAB buffer. This was mixed well and incubated at 37 °C overnight for enzymatic digestion. An equal volume of 1% formic acid was added, mixed well, and centrifuged at room temperature at 12,000 g for 5 min. The supernatant was collected and slowly passed through a C18 desalting column. Subsequently, the column was washed three times with wash solution (0.1% formic acid, 4% acetonitrile), and then eluted with an appropriate amount of elution solution (0.1% formic acid, 45% acetonitrile) twice. Finally, the eluted samples were combined and freeze-dried.

Fractionation by chromatography

First, we prepared mobile phase A (2% acetonitrile, 98% water, adjusted to pH =10 with ammonia) and mobile phase B (98% acetonitrile, 2% water, adjusted to pH =10 with ammonia). Then the lyophilized powder in mobile phase A, followed by centrifugation at 12,000 g for 10 min at room temperature. An L-3000 HPLC system with a Waters BEH C₁₈ column (4.6 × 250 mm, 5 µm) was utilized at a column temperature of 50 °C. Next, we collected 1 tube per minute, combined them into 10 fractions, and lyophilized them. Finally, we reconstituted each fraction with a solution containing 0.1% formic acid.

Liquid chromatography-mass spectrometry analysis

First, we prepared a mobile phase A (100% water, 0.1% formic acid) and mobile phase B (80% acetonitrile, 0.1% formic acid). Then the lyophilized powder in 10 µL of mobile phase A, followed by centrifugation at 15,000 rpm for 20 min at 4 °C. Next, we took 1 µg of the supernatant for injection in the LC-MS system. The EASY-nLCTM 1200 Nano UHPLC system was utilized with a custom-made 2 cm × 75 µm, 3 µm pre-column, and a custom-made 15 cm × 150 µm, 1.9 µm analytical column. The Q Exactive™ HF-X mass spectrometer was employed with a Nanospray Flex™ (ESI) ion source. The ion spray voltage was set to 2.3 kV, and the ion transfer tube temperature to 320 °C. For mass spectrometry, a data-dependent acquisition mode was used with a full scan range of m/z 350–1,500. The first-stage mass resolution was set to 60,000 (at 200 m/z), the C-trap maximum capacity to 3 × 10$\stackrel{ˆ}{}$6, and the maximum injection time to 20 ms. The top 40 ions were selected from the full scan based on ion intensity for higher-energy collision dissociation (HCD) fragmentation in the second stage. The second-stage mass resolution was 15,000 (at 200 m/z), C-trap maximum capacity at 5 × 10$\stackrel{ˆ}{}$4, maximum injection time at 45 ms, peptide fragmentation collision energy at 27%, threshold intensity at 2.2 × 10$\stackrel{ˆ}{}$4, and dynamic exclusion set to 20s. Finally, the raw mass spectrometry data file (.raw) was generated.

Database retrieval and bioinformatics analysis

Mass spectrometry data analysis was performed using Proteome Discoverer 2.2 software. The utilized database was X101SC19103253-Z01_ailuropoda-melanoleuca-uniprot.fasta (35,787 sequences). The search parameters were as follows: type of quantification was Precursor Quantification, the instrument was QE HF-X, with Precursor Mass Tolerance of ±10 ppm and Fragment Mass Tolerance of ±0.02 Da. Carbamidomethyl of cysteine was set as a static modification, while oxidation of methionine was considered a dynamic modification. Acetylation was designated as an N-terminal modification, and trypsin was chosen as the enzyme for digestion, allowing for a maximum of 2 missed cleavage sites. Peptide Spectrum Matches (PSMs) with a confidence level above 99% were designated as reliable PSMs. Proteins containing at least one unique peptide segment were considered credible proteins. Only the trustworthy peptide spectra and proteins were retained, and false discovery rate (FDR) validation was performed, eliminating peptides and proteins with an FDR exceeding 1%. KEGG (Kyoto Encyclopedia of Genes and Genomes, http://www.genome.jp/kegg), COG (Cluster of Orthologous Groups of proteins), and databases such as Gene Ontology (GO, http://www.geneontology.org) and Domain (IPR) using Interproscan software for functional annotation.

Proteomic profiling of testis of three pandas

In this study, proteomic evaluation was used to study the changes in protein abundance of different aged giant pandas, to determine the markers of male reproductive performance for future studies.

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (https://proteomecentral.proteomexchange.org) via the iProX partner repository with the dataset identifier PXD051003. The data access link on iProX is: https://www.iprox.cn/page/project.html?id=IPX0008444000. Figure 1A illustrates the experimental workflow for proteomic analysis. As a result, we identified a total of 139,039 peptides and 11,435 proteins (Table S1), the total protein numbers of the three giant panda testicular tissue samples were 10,090, 10,178, and 10,176, respectively (Fig. 1B). In addition to that, the Venn diagram (Fig. 1C) shows that 9,268 proteins were identified in common. Furthermore, J8d, F18y, and H22y each uniquely identified 298, 116, and 126 proteins, respectively.

The raw data obtained from the mass spectrometry needed to be controlled by a series of quality controls after the database was searched, including peptide length distribution, precursor ion mass tolerance distribution, unique peptide number distribution, protein coverage distribution, and protein molecular weight distribution. Firstly, each mass spectrometer has its measurement range, limiting the detectable peptide length. Peptides that are too long or too short cannot be detected by the mass spectrometer. The distribution of peptide lengths mainly falls between 7 and 25 residues (Fig. 2A), indicating the appropriate selection of protease enzymes. Secondly, there is a difference between the measured and theoretical molecular weights of peptides, which is important for evaluating mass spectrometer performance and result quality. The peak concentration around 0 indicates minimal mass deviation (Fig. 2B). Furthermore, after comparing identified peptide sequences with a protein database, proteins with identical peptide segments were grouped. Unique peptides within each group contribute to the protein’s specificity and reliability. A shallower curve indicates a higher number of unique peptides, signifying a more reliable identification of proteins (Fig. 2C). For a given identified protein, a higher count of supporting peptides enhances its credibility. Hence, the coverage of identified proteins indirectly reflects the overall accuracy of the results. The chart demonstrates consistent coverage across intervals, indicating high identification accuracy (Fig. 2D). Additionally, the (0, 0.1) interval has the highest count of peptides, suggesting the highest credibility within this range. Finally, the distribution of protein molecular weights is a crucial indicator of identified protein sizes. The chart displays a distribution of proteins across a wide range of molecular weights from 0 to above 100 kDa (Fig. 2E), signifying a broad spectrum of identified proteins.

Functional analysis of identified testis proteins

Currently, commonly used functional databases for annotation include Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Cluster of Orthologous Groups of Proteins (COG). These databases were utilized to functionally annotate proteins identified in the giant panda testicular tissue, aiming to understand the functional characteristics of different proteins and facilitate insights into the functionality of giant panda testicular tissue. As depicted in the Venn diagram (Fig. 3A), the total number of proteins annotated by GO, KEGG, COG, and IPR are 7,924, 10,790, 4,527, and 10,442, respectively. Among them, 3,808 proteins were commonly annotated. Additionally, KEGG, COG, and IPR individually annotated 332, 1, and 16 proteins, respectively. GO analysis was performed on the proteins identified from the testicular tissues of giant pandas to extract functional information about the proteins. Analysis results of the top 10-degree values in each category were selected and presented in histograms (Fig. 3B). Among them, the top five enrichment biological processes were: protein phosphorylation, redox processes, proteolysis, signal transduction, and metabolic processes. The function of the enriched molecule is mainly the binding reaction, such as protein binding, ATP binding, etc. Metabolic processes are key processes that maintain the activity of testicular tissues. Metabolites are not only essential for cell signaling and energy transfer but are also directly influenced by the physiological and pathological changes in tissues and accurately reflect the physiological changes.

The cluster of COG is the database for the orthologous classification of proteins and the proteins that make up each COG are assumed to be derived from a common protein ancestor (Li et al., 2016a). The COG classification of proteins is of great significance for functional annotation and evolutionary research (Li et al., 2016b). The identified proteins were divided into 26 COG categories, with the top group being General function prediction only (group R) (Fig. 3C). Translation, ribosomal structure and biogenesis (group J), and signal transduction mechanisms (group T) tied for second place. Posttranslational modification, protein turnover, and chaperones (group O) were third. There were 4,527 proteins annotated based on the COG database. However, about 60.41% of the identified proteins were not assigned as functional annotations.

KEGG is a pathway-based categorization of orthologous genes that provides biochemical metabolic pathway and signal transduction pathway analyses for a gene or protein dataset (Zhai et al., 2020). In this study, 33 pathways were mapped, mainly related to cellular processes, environmental information processing, genetic information processing, metabolism, and organismal systems (Fig. 3D). The most annotated pathway was signal transduction (980 proteins). Signal transduction proteins play a crucial role in the testes, where they mediate the actions of hormones and other signaling molecules, which are essential for testicular function, including spermatogenesis (the production of sperm), Leydig cell function (testosterone production), and the overall regulation of the testicular microenvironment (Luo et al., 2022; Smith & Walker, 2014). Some of the other major pathways include transport and catabolism, cell growth and death, associated with lipid/carbohydrate/amino acid metabolism, translation, folding, sorting, and degradation pathways.

IPR and subcellular localization analysis

Protein domains are reusable segments of proteins that play an important role in protein evolution. As shown in Fig. 3E, this study divided the identified proteins into 21 domains, of which the top three domains were: the protein kinase domain (319 proteins), WD40 repeat (237), and WD40-repeat-containing domain (215). The cell of an organism is a highly ordered structure, and the intracellular layer can be divided into different organelles or cellular regions according to different spatial distributions and functions. There were 15 types of subcellular localization of the identified proteins (Fig. 3F), the most common of which was nucleus protein, accounting for 32.89%. The second and third largest were cytoplasm protein (17.98%) and plasma membrane protein (10.48%), respectively.

GO analysis of age-specific unique proteins in three different age groups

Based on our previous research, we have identified the presence of certain unique proteins in the testicular tissues of giant pandas at three distinct age stages (specifically 8 days, 18 years, and 22 years old). To gain a deeper understanding of the biological significance of these proteins and their roles in the reproductive system, we conducted a functional enrichment analysis on these unique proteins. The results of this analysis will help us to elucidate their involvement in biological processes, their molecular functions, and their cellular locations. This comprehensive approach aims to provide insights into the specific functions of these proteins and their potential implications for the reproductive health and fertility of giant pandas. Figure 4A represents the GO analysis of unique proteins from the F8d sample, which has enriched a total of 20 biological processes. The top five enriched biological processes are Cell morphogenesis, Growth, Heart development, Glycoprotein metabolic process, and Cell junction organization. Figure 4B illustrates the GO analysis for unique proteins in the F18y sample, revealing enrichment in seven biological processes: Progesterone-mediated oocyte maturation, Estrogen signaling pathway, Response to alcohol, Oxidative phosphorylation, Membrane lipid biosynthetic process, Regulation of G protein-coupled receptor signaling pathway, and Modulation of chemical synaptic transmission. Finally, Fig. 4C represents the GO analysis of unique proteins in the H22y sample, which has enriched a total of three biological processes: Response to bacterium, Positive regulation of calcium ion transport, and Metabolism of water-soluble vitamins and cofactors.

Protein quantification and differential analysis

Proteome Discoverer 2.2 first obtained the relative quantification value of each PSM in each sample based on the peak area of the original lower spectrum and then corrected the relative quantification value of the unique peptide according to the quantification information of all PSMs contained in the identified unique peptide. Then it corrected the relative quantification value of each protein according to the quantitative information of all unique peptides contained in each protein. The results showed that the top five proteins with the highest abundance among the proteins identified by H22y were D2HU49, G1LEJ5, P18970, D2H496, and D2HVQ7. The five most abundant proteins identified by F18y were G1LEJ5, D2H496, D2HU49, G1LX08, and G1MDU7. In addition, the top five proteins identified from J8d were G1LEJ5, D2HU49, P18970, G1LVM5, and G1LP52 (Table S2). Ultimately, nine unique proteins were highly abundant in the three pandas, and we analyzed their abundance patterns from young to old times and found that these proteins were more abundant in the young and old periods, except for G1LEJ5, which had the opposite characteristics (Table S2). Among them (Table 1), G1LEJ5, D2HU49, and D2H496 are all uncharacterized proteins, which have potential research prospects and have a certain significance for the study of testicular tissue in giant pandas.