Changes of collagen content in lung tissues of plateau yak and its mechanism of adaptation to hypoxia

Experimental animals

This study selected three adult male yaks from Qumalai County (QML-Y, high-altitude yak at 4,500 m), three adult male yaks from Xunhua County (XH-Y, low-altitude yak at 2,600 m), and three adult male cattle from Xunhua County (XH-C, low-altitude yellow cattle at 2,600 m), totaling nine animals. At the same elevation, XH-C served as the control group while XH-Y was the experimental group. For different elevations within the same species, XH-Y was the control group and QML-Y was the experimental group. All animals were clinically healthy. The yaks were euthanized by exsanguination in a slaughterhouse. Lungs were harvested within 30 min of death, and deep tissue samples were collected from the diaphragmatic lobe of the left lung. Some tissues were snap-frozen in liquid nitrogen for long-term storage, while smaller samples were fixed in 4% paraformaldehyde for subsequent analysis. This study was approved by the Institutional Animal Care and Use Committee (IACUC) of Qinghai University, Xining, China (Approval No.: PJ-2023037). All methods adhered to the ARRIVE 2.0 guidelines and the Guidelines for Ethical Review of Animal Welfare in Laboratory Animals (GB/T 35892-2018) of the People’s Republic of China. All local regulations and laws were adhered to. All cattle and yaks used in this study were purchased from local farmers in Qinghai Province, China.

Histologic staining

Masson staining

The paraffin sections were sequentially treated with xylene I for 8 min, xylene II for 8 min, anhydrous ethanol I for 6 min, anhydrous ethanol II for 6 min, 95% ethanol for 6 min, 85% ethanol for 6 min, and 75% ethanol for 5 min, followed by rinsing with running water. The sections were then stained with Weigert’s Iron Hematoxylin Staining Solution for 5 min, washed thoroughly with water, and blued using Masson’s bluing solution before another wash with water. Next, the sections were stained with Masson Lichun red acidic magenta solution for 5 min, rinsed with running water, treated with phosphomolybdic acid solution for 5 min, and then the excess liquid was removed. They were subsequently stained with aniline blue solution for 3 min, followed by rinsing with 1% glacial acetic acid aqueous solution until no blue color remained. Finally, the sections were briefly washed with 95% ethanol, dehydrated with anhydrous ethanol for 3 min, and cleared in xylene III for 5 min, xylene II for 5 min, and xylene I for 5 min before being mounted with neutral gum. Lung tissue sections were stained using the Masson method, resulting in blue-stained collagen fibers and red-stained cytoplasm. All staining solutions were prepared at identical concentrations, and the staining times, temperatures, and washing procedures were standardized. All samples were fixed with the same fixative for an identical duration, and processed with consistent dehydration and clearing procedures. The embedding method was also uniform, ensuring that the tissue sections had the same thickness, thereby maintaining staining consistency across samples.

Observation and measurement

Stained tissue sections were observed and photographed with a Nikon optical microscope (DS-Ri2) and subsequently measured using Image-Pro Plus 6.0 software. The measurement data were analyzed using SPSS 26.0, with results presented as mean ± standard deviation ( $\overline{\mathrm{X}}$ ± SD). A t-test was conducted to assess differences, utilizing SPSS 26.0.

Transcriptome sequencing and analysis

Lung tissue samples were collected from cattle and yaks at the same altitude, as well as from yaks at different altitudes. RNA was extracted using the Trizol method, and mRNA was enriched with Oligo (dT) magnetic beads to construct cDNA libraries. The transcriptome was sequenced using the Illumina HiSeq 2500 platform. Raw data were filtered, and the clean reads were compared to the reference genome for gene annotation, quantitative analysis, and differential expression analysis. In the gene analysis section, we also performed analyses for gene coverage, sequencing randomness, and sequencing saturation. Gene coverage represents the percentage of bases in a gene that are covered by reads, calculated as the ratio of the bases covered by reads to the total bases in the gene’s coding region. To evaluate the randomness of mRNA interruption, we analyzed the distribution of reads across reference genes. Due to variations in the lengths of different reference genes, we normalized read positions to relative positions (the ratio of read position to gene length * 100) and counted reads across different positional ratios. Ideally, with good randomness, reads should be evenly distributed across gene sites. Sequencing saturation analysis assesses whether the sequencing depth of a sample is sufficient to detect most genes. As the sequencing depth (number of reads) increases, the number of detected genes rises until reaching a plateau, indicating saturation in gene detection. The input data for gene differential expression analysis consisted of read count data from gene expression level assessments. The differential expression analysis was performed using DESeq2 and was divided into three main steps: normalization of read counts, calculation of P-values based on the model, and correction for multiple hypothesis testing to obtain the false discovery rate (FDR). Genes with |log2(FoldChange)| > 0.585 and FDR < 0.05 were identified as significant, based on variance analysis results. This set of significant genes was then clustered and analyzed.

To ensure the reliability and reproducibility of the experimental results, this study employed both biological and technical replicates. For biological replication, nine independent samples were collected from local herders and divided into three groups, with each group undergoing three replicates under identical conditions. Technical replication involved conducting all experiments under stringent quality control, maintaining consistent temperature, time, and reagent lot number. The statistical power of this experimental design, calculated in RNASeqPower is 0.95 (https://rodrigo-arcoverde.shinyapps.io/rnaseq_power_calc/). This result illustrates the validity and rigor of our transcriptome sequencing.

Fluorescence quantitative PCR

Real-time quantitative PCR was performed using cDNAs from lung tissues of yak and cattle collected from high and low altitudes. Relative mRNA levels were measured using 7.5 μl of 2× Universal Blue SYBR Green qPCR Master Mix, 1.5 μl of 2.5 μM gene primers (both upstream and downstream), 2.0 μl of the reverse transcription product (cDNA), and 4.0 μl of nuclease-free water. GAPDH served as the internal reference gene, with the primer sequences listed in Table 1. The mRNA expression of COL1A1, COL1A2, and COL6A2 genes in lung tissues was analyzed using the 2^−∆∆CT method for relative quantification. An analysis of variance was conducted using SPSS 26.0 software to identify significant differences.

Masson staining results of yak and cattle lung tissues

The blue-stained regions of the lung tissue sections revealed collagen fiber structures. Collagen fibers were more abundant in the lung tissues of yaks at low altitude compared to both cattle at low altitude and yaks at high altitude (Fig. 1A). Quantitative analysis revealed that the percentage of collagen fibers in the lung tissues of yaks at low altitude was significantly higher than in cattle (P = 0.000) (Fig. 1B). Additionally, the percentage of collagen fibers in the lung tissues of yaks at high altitude was significantly lower than in yaks at low altitude (P = 0.008) (Fig. 1C).

Overall statistics of differential genes between groups

The results revealed 3,684 differentially expressed genes in yaks compared to cattle in low-altitude areas, including 2,189 up-regulated genes and 1,495 down-regulated genes (Fig. 2A). Additionally, there were 2,326 differentially expressed genes when comparing yaks from high and low-altitude areas, with 1,340 up-regulated genes and 986 down-regulated genes (Fig. 2B).

Fluorescence quantitative PCRvalidation analysis

In the low-altitude comparison between yaks and cattle, relative quantitative analysis was performed on the COL1A1, COL1A2 and COL6A2 genes using cattle as the standard (Fig. 3A). The COL1A1 gene showed no significant differential expression between the two groups (P = 0.247), while the COL1A2 gene exhibited significant differential expression (P = 0.049), and the COL6A2 gene displayed highly significant differential expression (P = 0.006).

In the comparison between high-altitude and low-altitude yaks, the COL1A1, COL1A2 and COL6A2 genes were analyzed using relative quantitative methods with low-altitude yaks as the reference (Fig. 3B). The COL1A1 gene showed significant differential expression between the two groups (P = 0.029), while the COL1A2 (P = 0.001) and COL6A2 (P = 0.000) genes exhibited highly significant differential expression. The quantitative detection results were consistent with the transcriptome sequencing trends.

Gene ontology enrichment analysis

Gene Ontology (GO) analysis revealed that low-altitude yaks were enriched in biological processes such as bioadhesion, cell adhesion, and extracellular matrix organization, as well as cellular components including cell periphery, extracellular matrix, and collagen-containing extracellular matrix, compared to low-altitude cattle (Fig. 4A); and high-altitude yaks were enriched in similar biological processes and cellular components compared to low-altitude yaks (Fig. 4B). These findings indicate that differentially expressed genes in lung tissues are involved in cell adhesion and collagen synthesis in both groups.

Kyoto encyclopedia of genes and genomes enrichment analysis

Differentially expressed genes in the lung tissues of yaks and cattle at low altitude were annotated using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Analysis of the top 20 pathways with the most significant differences (smallest Q-value) revealed that these genes were enriched in signaling pathways such as ECM-receptor interaction and protein digestion and absorption (Fig. 5A). Similarly, differentially expressed genes in yak lung tissues from both high and low altitude regions were annotated using the KEGG database. Analysis of the top 20 pathways with the most significant differences (smallest Q-value) showed enrichment in signaling pathways including ECM-receptor interaction, protein digestion and uptake, focal adhesion, and cell adhesion molecules (Fig. 5B). These results suggest that the differentially expressed genes in the lung tissues of yaks and cattle at low altitude, as well as yaks at both high and low altitudes, are enriched in pathways related to collagen synthesis.

Screening of differentially expressed collagen genes

Based on existing studies and GO annotation, 19 significantly different collagen genes were identified in the lung tissues of yak and cattle at low altitude. Among these, 18 genes were up-regulated and one gene was down-regulated in yak lung tissues (Table 2). Additionally, 23 significantly different collagen genes were found in yak lung tissues at both high and low altitudes, with all of these genes being down-regulated at high altitude (Table 3).

Significant differential gene expression changes in collagen

The differentially expressed collagen genes in the two groups were categorized into three types: mesenchymal collagen synthesis-related genes, basement membrane collagen synthesis-related genes, and peripheral cellular collagen synthesis-related genes. Specific values of relative gene expression are presented in Supplemental Material 1. Among these, interstitial collagen synthesis-related genes such as COL1A1, COL1A2, COL2A1, COL3A1, COL9A2, COL12A1, COL14A1, COL16A1, COL21A1, COL23A1, and COL26A1 were highly expressed in the lung tissues of yak from low-altitude areas compared to both cattle from low-altitude areas and yak from high-altitude areas (Fig. 6A). Basement membrane collagen synthesis-related genes, including COL4A1, COL4A2, COL4A4, COL4A5, COL4A6, COL8A1, and COL18A1 were also highly expressed in the lung tissues of yak from low-altitude areas compared to cattle from low-altitude areas and yak from high-altitude areas (Fig. 6B). Peripheral collagen synthesis-related genes such as COL5A1, COL5A2, COL5A3, COL6A1, COL6A2, COL6A3, and COL15A1 were highly expressed in the lung tissues of yak from low-altitude areas compared to cattle from low-altitude areas and yak from high-altitude areas. In contrast, COL6A5 was more highly expressed in the lung tissues of yak from high-altitude areas compared to those from low-altitude areas and cattle (Fig. 6C).