All animal procedures were approved by the Animal Care and Use Committee of Xinjiang Uygur Autonomous Region, China, and were conducted in accordance with the guidelines developed by the China Council on Animal Care and Protocol.

Adult L. yarkandensis and O. cuniculus were selected and collected in December 2019. Six L. yarkandensis was collected from Shaya County, Aksu Prefecture, northwest of the Tarim Basin. And animals were assessed to be adults based on a skull length of greater than 75.50 mm. The average age of 6–8 months and weight of 1.5–1.8 kg. The Animal Laboratory of Tarim University provided six healthy adult rabbits of similar age. These animals were maintained initially in individual cages and had free access to food and drinking water at all times. One week after feeding, the animals were anaesthetized with 20% Ulatan (ethyl carbamate) (0.4 ml/kg) and sacrificed immediately after coma. The kidney was quickly dissected and cut into pieces. The right renal cortex, outer medulla and inner medulla were removed for analyzing the kidney protein, RNA, and the renal cortex for the RNA sequencing. The left kidneys were dissected partially to obtain the cortex, outer medulla and inner medulla, infused with 4% paraformaldehyde (Sigma-Aldrich, Shanghai, China) and then fixed overnight for histologic examination.

Total RNA-seq using the Trizol method and its experimental protocol were utilized. We used GoodView to detect the quality of RNA dyeing on a 1% agarose gel and used nucleic acid protein tester BioSpectrometer (Eppendorf) determination of purity of RNA. The RNA was then reversely transcribed into cDNA using EasyScript^® One-Step gDNA Removal and cDNA Synthesis SuperMix (Transgen, Beijing, China).

A total amount of 1.5 µg RNA per sample was used as input material for the RNA sample preparations. Following the manufacturer’s recommendations, sequencing libraries were generated using NEBNext^® Ultra™ R.N.A. Library Prep Kit for Illumina^® (NEB, United States), and index codes were added to attribute sequences to each sample. The library fragments were purified using the AMPure XP system (Beckman Coulter, Beverly, United States). Then, 3 µL USER Enzyme (NEB, United States) was used with size-selected, adaptor-ligated cDNA at 37°C for 15 min followed by 5 min at 95°C before PCR. Then PCR was performed with Phusion High-Fidelity DNA polymerase, Universal PCR primers, and Index (X) Primer. PCR products were purified (AMPure XP system), and library quality was assessed on the Agilent Bioanalyzer 2100 system.

Raw data (raw reads) of fastq format were first processed through in-house Perl scripts. Clean data (clean reads) were obtained by removing reads containing adapters, ploy-N, and low-quality reads from raw data. Q20, Q30, GC-content, and sequence duplication levels of the clean data were calculated. All downstream analyses were based on clean data with high quality. High-quality RNA sequencing data from the library were assembled using Trinity software. If the same genes identified multiple transcripts, we chose one of the longest transcripts as representative of the gene sequence, hereafter referred to as the unigene sequence.

All unigene sequences were searched for functional annotations in several databases: Nr (NCBI non-redundant protein sequences); Nt (NCBI non-redundant nucleotide sequences); Pfam (Protein family); KOG/COG (Clusters of Orthologous Groups of proteins); Swiss-Prot (a manually annotated and reviewed protein sequence database); KO (KEGG Ortholog database); and GO (Gene Ontology).

The transcriptome spliced at Trinity was used as the reference sequence, and clean reads of each sample were mapped to this. In this process, we used RSEM software (Li and Dewey, 2011) for Bowtie comparisons, obtained the read count number of each sample compared to each gene, and performed Fragments Per Kilobase per Million (FPKM) conversion to analyze the gene expression level.

The read count data were analyzed for gene differential expression using the DEseq R package (1.10.1). DESeq provides statistical routines for determining differential expression using a model based on the negative binomial distribution. The resulting p values were adjusted using Benjamini and Hochberg’s approach for controlling the false discovery rate. Genes with an adjusted p-value < 0.05 as found by DESeq were assigned as differentially expressed.

Gene Ontology (GO) enrichment analysis of the differentially expressed genes (DEGs) was implemented by the GOseq R package based on the Wallenius non-central hypergeometric distribution (Young et al., 2010) which can adjust for gene length bias in DEGs. We used KOBAS (Mao et al., 2005) software to test the statistical enrichment of DEGs in KEGG pathways. We identified significantly enriched GO terms and signaling pathways (p < 0.05).

SamTools (http://samtools.sourceforge.net/) and BLASTX in BLAST+ were used to compare unigene sequences to the O. cuniculus transcription factor dataset in the AnimalTFDB 3.0 database (http://bioinfo.life.hust.edu.cn/AnimalTFDB) (Hu et al., 2019). BLASTX result files were interpreted and screened, and TF Length >110 amino acids and identity >80% were selected as screening conditions. The TRANSFAC database (https://genexplain.com/transfac/) was adopted to predict the TF corresponding to target gene prediction. Unigene ID exists in each TF and target gene sequence for subsequent analysis.

The DEG KEGG pathway enrichment analysis results determined that the renal water reabsorption signaling pathways were mainly vasopressin-regulated via the proximal tubule bicarbonate recovery pathway. We took DEGs as the target genes, which play a key role in the pathway. We studied the functional predictions and regulatory analysis of transcription factors upstream of the target gene according to the regulatory relationship between the predicted transcription factors and the target gene.

To verify the accuracy and validity of transcriptome sequencing data, reverse transcription cDNA products were amplified by polymerase chain reaction (PCR). Primers targeted AQP1 (Cluster-1445.13691), AQP2 (Cluster-1445.18746), CREB3 (Cluster-1445.10450), ADCY3 (Cluster-1445.26438), FXYD2 (Cluster-1445.18046), HIF1A (Cluster-1445.20583), and NFATc1 (Cluster-1445.8040). (Supplementary Table S1). Polymerase chain reactions consisted of 10 μM of each primer, dNTP Mixture, and TB Green^® Premix Ex TaqTM Ⅱ (Takara Bio, Beijing, China), in a total reaction volume of 20 μl. Quantitative RT-PCR was performed on a CFX96 Touch Deep Well (Bio-Rad, Delaware, American). Conditions were 95°C for 3 min, 45 cycles of 95°C for 5 s, 60°C for 30 s, and 72°C for 10 s. Analysis of relative gene expression was performed using the 2^−ΔΔCt method (Schmittgen and Livak, 2008). Values were normalized to GAPDH values and expressed as gene/GAPDH ratio.

Total proteins were isolated from the kidneys of O. Cuniculus and L. yarkandensis. A 100 mg sample of right kidney tissue was isolated and placed in chilled lysis buffer containing 0.04 M Tris-HCl (PH 7.4), 0.82% NaCl, 1.5% Triton X-100, 0.5% deoxycholic acid sodium salt, 0.1% SDS, a protease inhibitor cocktail (Sigma-Aldrich, Shanghai, China), and 1 mM PMSF. The tissues were homogenized in 1 ml of lysis buffer. The homogenates were placed on ice for 20 min and then were centrifuged at 12,000 × g for 20 min at 4°C. The supernatants were then collected. The total protein concentration was measured using a BCA protein assay reagent kit (Aidlab, Beijing, China) according to the manufacturer’s protocol.

The total proteins were solubilized in Laemmli sample buffer at 70°C for 8 min and then were subjected to SDS-polyacrylamide gel electrophoresis. After transferring to polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Delaware, American), western blotting membranes were stripped and blocked, followed by incubation overnight at 4 °C with antibodies against HIF1A, NFATC1, NF-κB1, ADCY3, ATP1A1, SLC4A4 and GAPDH (Proteintech, Wuhan, China). After washing, the membranes were incubated with horseradish peroxidase (HRP)-labeled anti-rabbit secondary antibody (Proteintech, Wuhan, China) for 1 h at room temperature and then were visualized via enhanced chemiluminescence (SuperSignal, Pierce). Western blotting was scanned using Tanon 5200, and then the labeling density was quantified using Gel-Pro Analyzer4. Integral optical density (IOD) was used to represent the density measurement results of the target protein bands, and then the determination results of the target protein and corresponding GAPDH protein were normalized. Finally, the mean values of L. yarkandensis and O. cuniculus were compared.

After dissecting, the kidney tissues of L. yarkandensis and O. cuniculus were fixed in 4% paraformaldehyde. For histological analysis, the cortex, outer medulla and inner medulla of the left kidney were paraffin-embedded with the usual paraffin embedding method. The paraffin-embedded cortex, outer medulla, and inner medulla were sectioned at 6 µm for immunohistochemical staining of ADCY3, ATP1A1 and SLC4A4 deparaffinized by washing in xylene three times for 10 min each. This was followed by rehydration through a series of ethanol washes from 100% to 70% ethanol. The slides were then placed in methanol containing 0.5% hydrogen peroxide for removing the endogenous peroxidase activity. Non-specific binding was blocked by incubating the slides for 1 h at room temperature in 5% BSA. The cortex, outer medulla, and inner medulla sections were incubated with antibodies against ADCY3 (Proteintech, 19492-1-AP, 1:100), ATP1A1 (Proteintech, 14418-1-AP, 1:240), and SLC4A4 (Proteintech, 11885-1-AP, 1:500) overnight at 4°C. The sections were rinsed in 0.1 M PBS (PH 7.2–7.4) and then incubated for 30 min at room temperature with HRP-labeled goat anti-rabbit secondary antibody (Proteintech, Wuhan, China). The sections were washed with PBS, incubated with diaminobenzidine for 6 min, and then washed again. The tissue sections were stained using hematoxylin (Sigma-Aldrich, Shanghai, China) for 40 s and then washed under running water for 5 min. The primary antibody was substituted with PBS, and the controls underwent a similar procedure as described before. The immunostained sections were observed and photographed on a microscope (Motic BA600-4, Beijing, China). The captured images were analyzed in the IpWin32 software for each group’s IOD and area of immunopositive cells. The difference in the average density (IOD/Area) between the groups was compared.

Statistical analysis software Graphpad Prism was used to calculate the means and standard errors. A student’s t-test was used to assess the differences between groups. A p-value of less than 0.05 was considered statistically significant.

We constructed six samples containing three biologically repetitive cDNA libraries from O. cuniculus and L. yarkandensis. We sequenced the libraries using an Illumina high-throughput sequencing platform. The six samples were named OC_1, OC_2, and OC_3 (O. cuniculus), and LY_1, LY_2, and LY_3 (L. yarkandensis). For the sequencing data of these 6 samples, Raw Reads were filtered to remove the low-quality data. The remaining clean reads data showed little difference, and the readings were 21.92, 20.22, 21.30, 19.94, 21.86, and 21.93 million, respectively (Supplementary Table S2). The sequences based on clean reads were analyzed and spliced by Trinity. The longest transcript of each gene was selected as the unigene sequence for a total of 42,998 unigene sequences of L. yarkandensis; 15,413 unigene sequences ranged from 300 to 500 bp and 16,648 unigene sequences exceeded 1,000 bp (Supplementary Table S3, Figure S1).

To obtain the potential functional information of unigene sequences, we searched and annotated seven databases, including Nr, Nt, Pfam, KOG/COG, Swiss-Prot, KEGG, and GO. The proportion of unigene sequences on annotations ranged from 10.76% to 78.84%. Among the 42,998 unigene sequences, NR and NT databases annotated 20,010 (46.53%) and 33,038 (76.83%) sequences (Supplementary Table S4).

We used GO Terms and KEGG Pathways to analyze the functionality of the unigenes; 15,263 unigenes were annotated with GO terms. These belong to 43 functional groups, distributed in three main categories: molecular function, biological process, and cellular component (Figure 1). In the biological process category, cellular process, metabolic process, biological regulation, and regulation of biological process are the functions of many genes. In the cellular component category, intracellular, cellular anatomical entity, and protein-containing complex are the main functions. Among the molecular functions, binding, catalytic activity, and transporter activity annotate the functions of many genes (Supplementary Table S5).

For KEGG Pathways, 11,253 unigenes were annotated. The top 10 pathways are signal transduction (1,689 sequences), immune system (836 sequences), transport and catabolism (763 sequences), endocrine system (688 sequences), Cellular community-eukaryotes (601 sequences), signaling molecules and interaction (536 sequences), translation (441 sequences), folding, sorting and degradation (415 sequences), and nervous system (413 sequences) (Figure 2). KEGG analysis shows that 26 unigene sequences are proximal tubule bicarbonate reclamation pathways; 48 unigene sequences are involved in vasopressin-regulated water reabsorption pathways.

Gene expression levels of unigene sequences were identified and expressed by FPKM. Our study used two species treatment groups to analyze the ReadCount of gene expression level analysis. L. yarkandensis and O. cuniculus had 22,910 and 23,290 expressed genes, respectively. Among them, 9,010 and 9,390 genes were unique to L. yarkandensis and O. cuniculus, respectively, and the two species shared 13,900 genes (Figure 3A).

The up-regulation distribution and p-value changes of differentially expressed genes (DEGs) are shown in the volcano diagram: 6,610 up-regulated genes and 5,727 down-regulated genes were expressed in L. yarkandensis (p < 0.05 and |log2FoldChange| > 1) (Figure 3B). These DEGs reflect the significant difference in gene expression level between L. yarkandensis and O. cuniculus.

To further determine the range of DEGs related to water reabsorption, DEGs were first positioned using GO Enrichment. The largest difference between L. yarkandensis and O. cuniculus are intracellular (1,163 DEGs), cellular nitrogen compound metabolic process (1,149 DEGs), ion binding (1,002 DEGs), biosynthetic process (905 DEGs), and organelle (875 DEGs). The results showed that L. yarkandensis and O. cuniculus differed in fundamental biological processes, including ion binding, intracellular, and biosynthetic process (Supplementary Table S6). Also, 1,559 DEGs were located as KEGG Pathways. The largest differences between L. yarkandensis and O. cuniculus are pathways in cancer (113 DEGs), PI3K-Akt signaling pathway (111 DEGs), focal adhesion (87 DEGs), regulation of actin cytoskeleton (68 DEGs), rap1 signaling pathway (62 DEGs), and endocytosis (60 DEGs). The results showed that L. yarkandensis and O. cuniculus had different expression levels in cancer, actin cytoskeleton, and some biological tissues (Supplementary Table S7).

According to BLAST comparison results of unigenes and the O. cuniculus transcription factor database. We predicted 1,159 transcription factors from unigene, of which 232 transcription factors were differentially expressed (Supplementary Table S8).

After obtaining predicted transcription factors, we searched for the Symbol ID of the transcription factor and corresponding downstream target genes of each transcription factor in the TRANSFAC database. Among the 232 differentially expressed transcription factors predicted, 20 existed in the database, and 1,137 corresponding target genes were predicted. Each transcription factor and target gene sequence has a corresponding unigene sequence, which was used for subsequent analysis (Supplementary Table S9).

To study the water reabsorption capacity of L. yarkandensis, we studied DEGs regulating water reabsorption via pathways: the vasopressin-regulated water reabsorption pathway and the proximal tubule bicarbonate reclamation pathway. From the two pathways, it was found that five DEGs (AQP1, AQP2, CREB3, ADCY3, and FXYD2) were significantly up-regulated (4 genes) or down-regulated (1 gene). These DEGs may play a key role in water reabsorption in their respective pathways. By cross-comparing the data of 5 water reabsorption DEGs and transcription factors and their target genes, we obtained related transcription factors that may regulate these 5 differential genes. Since a gene can be predicted to have multiple transcription factors, we selected two DEGs with significant differential expression, namely HIF1A and NFATc1, as the target transcription factors (Supplementary Table S10).

To verify the accuracy and validity of transcriptome sequencing results and the reliability of prediction results of transcription factors and their target genes, we used quantitative RT-PCR to detect the mRNA expression levels. These genes include AQP1 (Cluster-1445.13691), AQP2 (Cluster-1445.18746), CREB3 (Cluster-1445.10450), ADCY3 (Cluster-1445.26438), FXYD2 (Cluster-1445.18046), HIF1A (Cluster-1445.20583), and NFATc1 (Cluster-1445.8040). We then compared the expression level of the transcriptome sequencing results and analysis. As shown in Figures 4, 7 genes determined by the RNA-seq experiment had a good correlation with the experimental results of RT-qPCR. Among them, AQP1, AQP2, ADCY3, CREB3, HIF1A, and NFATc1 were up-regulated in L. yarkandensis and FXYD2 was down-regulated. This finding indicates that our RNA-seq sequencing results are reliable.

According to the analysis of RT-qPCR data, ADCY3 was significantly up-regulated in the kidney of L. yarkandensis compared with O. cuniculus. Literature has shown that AQP2 expression level is related to transcription factor NF-κB1 (Albertoni Borghese et al., 2018). In the proximal tubule bicarbonate reclamation pathway, ATP1A1 and SLC4A4 proteins also play a critical role. In order to further analyze the regulatory mechanism of the proximal tubule bicarbonate reclamation pathway and vasopressin-regulated water reabsorption pathway, protein abundance of SLC4A4, ATP1A1 and transcription factor NF-κB1 were measured by western blot.

To further analyze whether the protein abundance of ADCY3, ATP1A1, and SLC4A4, and transcription factors HIF1A, NFATc1, and NF-κB1, western blot was used to detect the protein expression levels of ADCY3, ATP1A1, SLC4A4, HIF1A, NFATc1, and NF-κB1 in the kidney of both L. yarkandensis and O. cuniculus. As shown in Figure 5, GAPDH was used as the internal reference protein, and there was a band at 36 KDa incubated with a GAPDH antibody. There was a band at 121 KDa when SLC4A4 antibody was cultured, and integral optical density (IOD) analysis of western blot results showed that the protein expression level of SLC4A4 in the kidney of L. yarkandensis was significantly up-regulated (p < 0.001) (Figure 5A). The HIF1A antibody was incubated with a band at 120 kDa, and IOD analysis showed that the HIF1A protein expression level in the kidney of L. yarkandensis was up-regulated (p < 0.05) (Figure 5B). The ADCY3 antibody was incubated with a band at 180 kDa, and IOD analysis showed that the ADCY3 protein expression level in the kidney of L. yarkandensis was up-regulated (p < 0.05) (Figure 5C). The ATP1A1 antibody was incubated with a band at 100 kDa, and IOD analysis showed that the ATP1A1 protein expression level in the kidney of L. yarkandensis was up-regulated (p < 0.05) (Figure 5D). The NFATc1 antibody was incubated with a band at 101 kDa, and IOD analysis showed that the NFATc1 protein expression level in the kidney of L. yarkandensis was up-regulated (p < 0.01) (Figure 5E). The NF-κB1 antibody was incubated with a band at 105 kDa, and IOD analysis showed that the NF-κB1 protein expression level in the kidney of L. yarkandensis was up-regulated (p < 0.05) (Figure 5F) (Supplementary Figure S2).

Immunohistochemistry data show that positive signals of the ADCY3 protein are mainly located in the basolateral membrane of collecting duct cells, and positive staining is light yellow or brownish-yellow. IHC mean optical density analysis showed that compared with O. cuniculus, the ADCY3 protein was strongly labeled in the basolateral membrane of the renal cortical collecting duct (CCD) of L. yarkandensis (Figures 6A,D), with significantly up-regulated expression (p < 0.001). Strong markers were detected in the basolateral membrane of the outer medullary collecting duct (OMCD) (Figures 6B,E), which was significantly up-regulated (p < 0.05). Strong markers were detected in the basolateral membrane of the renal inner medullary collecting duct (IMCD) (Figures 6C,F), presenting an extremely significant up-regulation (p < 0.01) (Figure 6).

As shown in Figure 7, the IHC positive signals of ATP1A1 protein are located in the proximal basolateral membrane of the PCT and the thin limbs (TL) of the renal medulla, and the positive staining is light yellow or brown-yellow. Compared with O. cuniculus, strong markers of ATP1A1 protein were detected in the basolateral membrane of PCT in the renal cortex of L. yarkandensis (Figures 7A,D), and their expression was significantly up-regulated (p < 0.05). Strong markers were detected in the basolateral membrane of the proximal tubule straight of OM in L. yarkandensis (Figures 7B,E) and they were significantly up-regulated (p < 0.05). Strong markers were detected in the TL of the inner medulla in L. yarkandensis (Figures 7C,F) and were significantly up-regulated (p < 0.05).

As shown in Figure 8, the positive IHC signal of SLC4A4 protein was in the basolateral membrane of PCT, and the positive staining was light yellow or brownish-yellow. Compared with O. cuniculus, the SLC4A4 protein was strongly labeled in the basolateral membrane of PCT in the kidney of L. yarkandensis (Figures 8A,B) and was significantly up-regulated (p < 0.001). However, in the staining of the renal medulla, SLC4A4 was not expressed in O. cuniculus and L. yarkandensis. The gene was only expressed in the PCT basolateral membrane of the renal cortex.

According to transcriptome sequencing data and DEGs analysis results, L. yarkandensis and O. cuniculus shared 42,998 unigene sequences, including 6,610 genes were significantly up-regulated, 5,727 genes were significantly down-regulated, and 30,540 genes were not differentially expressed. GO enrichment analysis and KGEE pathway enrichment analysis found that many essential cell functions, biological processes, and pathways of organisms have different degrees of gene expression. According to the sequence analysis of DEGs, 232 differentially expressed transcription factors were identified, among which 5 genes and 7 transcription factors were found in relation to both vasopressin-regulated water reabsorption pathways and proximal tubule bicarbonate reclamation pathways. According to the difference in expression of transcription factors, gene screening and literature support, 3 transcription factors and 7 genes were selected to experiment.

In proximal tubule bicarbonate reclamation pathway, previous studies have shown that the primary function of the AQP1 proximal renal tubule is to absorb more than 70% of the water from the glomerular filtrate (Katsura et al., 1995). The expression levels of AQP1 and AQP2 in the kidneys of L. yarkandensis were higher than those of O. cuniculus, which may be related to the increased water permeability of renal tubules of L. yarkandensis (Zhang et al., 2019a). Therefore, the high expression of AQP1 in L. yarkandensis will likely increase the permeability of renal tubules to water. Quantitative RT-PCR results of the mRNA level showed that the AQP1 was up-regulated, and the FXYD2 was down-regulated. Western blotting showed that the SLC4A4 protein and ATP1A1 protein were highly expressed in the kidney of L. yarkandensis. Our previous experiment found that the AQP1 protein is significantly overexpressed in the kidney of L. yarkandensis (Zhang et al., 2019a). Immunohistochemical analysis showed that ATP1A1 protein is in the basolateral membrane of PCT and TL of the renal medulla, and the SLC4A4 protein was in the basolateral membrane of renal PCT.

ATP1A1 protein can establish and maintain a transmembrane electrochemical gradient for Na⁺ and K⁺, whereas ATP1A1 variants may alter cation channel abnormalities and cause the loss of Na-K-ATPase function (Čechová et al., 2016). This may lead to dysfunction of tubular reabsorption (Lin et al., 2021). FXYD2 is a regulatory subunit and has been shown to inhibit Na-K-ATPase activity (Mayan et al., 2018). The restriction of FXYD2 in the distal nephron may play an essential role in basolateral Na-K-ATPase and trans-epithelial sodium reabsorption in the principal cells (Floyd et al., 2010). The presence of Na-K-ATPase with high Na⁺ affinity may be conducive to effective reabsorption of Na⁺ in kidney segments with high Na⁺ load. In conclusion, the up-regulated expression of ATP1A1 and down-regulated expression of FXYD2 together increase Na-K-ATPase activity, accelerate the exchange rate between Na⁺/K⁺ in proximal tubules, and maintain ion concentration in proximal tubules to maintain cell homeostasis.

Most of the bicarbonate is reabsorbed from the proximal tubules using NHE (mainly NHE3) for H⁺ in exchange for sodium (Wang et al., 2001; Li et al., 2019). The secreted H⁺ binds with HCO₃ ⁻ in the lumen and is converted into CO₂ and H₂O (Purkerson and Schwartz, 2007). NHE3 was no differentially expressed in L. yarkandensis and O. cuniculus, and the protein encoded by the NHE3 gene was normally expressed in both L. yarkandensis and O. cuniculus. CO₂ and H₂O in the lumen enter the proximal tubule cells through AQP1 and are converted into HCO₃ ⁻ and H⁺ by CAⅡ. The protein NBCe1 encoded by the SLC4A4 gene regulates Na⁺-HCO₃ ^- cotransport in the basolateral proximal tubules, an essential step in bicarbonate reabsorption (Kurtz and Zhu, 2013). HCO₃ ⁻ and H⁺ enter the blood at a ratio of 3:1 through NBCe1 to complete the process of reabsorption (Soleimani et al., 1987). Therefore, the up-regulated expression of the SLC4A4 protein in the kidney of L. yarkandensis can accelerate the bicarbonate reclamation process.

In the vasopressin-regulated water reabsorption pathway, RT-qPCR results showed that AQP2 was up-regulated, the ADCY3 gene was significantly up-regulated, and the CREB3 gene was up-regulated to some extent. Western blotting showed that the ADCY3 protein was highly expressed in the kidneys of L. yarkandensis, and our previous experiment found that AQP2 protein was highly expressed in the kidneys of L. yarkandensis (Zhang et al., 2019a). Immunohistochemical analysis showed that the ADCY3 protein is highly expressed in CCD and MCD. These results suggest that AVP-mediated water reabsorption mainly occurs in the renal CD. According to the expression of related genes and proteins, the up-regulated expression of ADCY3 in L. yarkandensis can regulate the up-regulated expression of CREB3 in the nucleus of CD cells, increasing the efficiency of phosphorylated CREB3 to induce AQP2 gene transcription and increase AQP2 protein expression level (Brown, 2003). The up-regulated expression of ADCY3 may induce the AQP2 protein to accelerate the transfer from the intracellular storage vesicles to the apical plasma membrane through the cAMP-PKA-AQP2 level, increasing the number of AQP2 proteins on the apical plasma membrane, enhanced the water reabsorption capacity of CD cells in the kidney of L. yarkandensis. Based on the above studies, we speculated that L. yarkandensis could promote water reclamation and maintain water in the body.

HIF1A induces glycolysis by facilitating pyruvate dehydrogenase kinase-1 (PDK-1) activity and actively inhibits mitochondrial function and oxygen utilization (He et al., 2014). In the nucleus pulposus, HIF1A could also significantly maintain the expression levels of AQP1 and AQP5 (Johnson et al., 2015). According to RT-qPCR results, HIF1A was up-regulated in the kidney of L. yarkandensis, and western blotting analysis showed that HIF1A protein was highly expressed. Transcriptomic data were used to predict that HIF1A has some regulatory effect on FXYD2. Since the down-regulated expression of FXYD2 and up-regulated expression of ATP1A1 jointly promote the increase of NA-K-ATPase activity, it is speculated that the up-regulated expression of HIF1A can reduce the expression of the FXYD2 gene–thus indirectly regulating the increase of NA-K-ATPase activity. In the vasopressin-regulated water reabsorption pathway, transcription factor HIF1A has some regulatory effects on the CREB3 gene. Therefore, the up-regulated expression of HIF1A can regulate the up-regulated expression of CREB3, thereby increasing the transcription and expression of AQP2 and increasing the water reabsorption capacity of CD cells.

There are four NFATc proteins in the NFATc subtype transcription factor family, i.e., NFATc1 (NFAT2), NFATc2 (NFAT1), NFATc3 (NFAT4), and NFATc4 (NFAT3) (Hogan et al., 2003). NFATc usually collaborates with AP-1 to regulate gene transcription and the transcription of inflammatory cytokines such as interleukin-2 (IL-2) (Macián et al., 2001; Reppert et al., 2015). NFATc3-KO mice medulla AQP2 mRNA and protein expression levels are reduced, and integrin-linked kinase (IKL) can adjust NFATc transcription activity and through the influence of AQP2 NFATc/AP-1 promoter (Hatem-Vaquero et al., 2017). Other studies have shown that NO can enhance Ca₂ ⁺-induced NFATc activation, synergistic with Ca₂ ⁺, to improve AQP2 mRNA and protein expression levels in mouse papillae (Albertoni Borghese et al., 2011). Our study found that NFATc1 was up-regulated in mRNA and protein levels. By contrast, the CREB3 gene and AQP2 protein were up-regulated in the kidneys of L. yarkandensis, suggesting that the NFATc1 transcription factor could promote the transcription of AQP2.

Studies have shown that NF-κB can release P65 at the κB site and increase P50 and P52 monomer binding to inhibit AQP2 transcription under high osmosis (high tension) (Hasler, 2009). In renal hypertension, decreased AVP sensitivity and increased NF-κB activity in renal marrow collecting tubes may lead to decreased AQP2 expression (Albertoni Borghese et al., 2018). The NF-κB1 gene and protein expression were up-regulated in the kidneys of L. yarkandensis, which seemed to inhibit the expression of the AQP2 protein. However, another transcription factor, NFATc1, was up-regulated, and the expression level was about 2 times higher than the NF-κB1 gene in protein expression. We hypothesize that the kidney of L. yarkandensis requires a high expression of the AQP2 protein on the CD cell membrane so that NFATc1 enhances the expression of the AQP2 protein. The transcription binding site of NF-κB1 in AQP2 is located upstream of the transcription binding site of NFATc1 (Hasler, 2009), is stimulated by high expression of AQP2, and leads to the high water content of urine in the urinary lumen; that part of NF-κB1 gene expression is up-regulated. The inhibition of the expression of the partial AQP2 protein in the urinary lumen reduces urine volume and reduces the high degree of kidney penetration. In conclusion, We speculate that L. yarkandensis can increase the water reclamation efficiency—may be a key adaptation to the arid desert environment.

We speculate that the regulation of transcription factors and water reabsorption by L. yarkandensis kidneys is based on previous studies. When the body is exposed to drought and water shortage, the transcription factor HIF1A may be activated to inhibit the expression of FXYD2. At the same time, ATP1A1 is highly expressed, thus jointly enhancing the expression of NA-K-ATPase in proximal tubule cells. The up-regulated expression of the AQP1 protein and NBCe1 protein encoded by the SLC4A4 gene improves the reabsorption capacity of proximal tubule cells to water (Figure 9). This increases the water utilization rate in L. yarkandensis.

On the other hand, the activated transcription factor HIF1A promotes the expression of the CREB3 gene in the CD nucleus, and the highly expressed transcription factor NFATc1 also increases the transcription of the AQP2 gene, leading to the high expression of ADCY3 as regulated by AVP, and increases the cAMP level. These changes increase the synthesis of AQP2 protein in different directions and are likely to accelerate the transfer of AQP2 protein to the apical plasma membrane. Only the NF-κB1 transcription factor limited AQP2 protein synthesis, and together with the transcription factor NFATc1, AQP2 protein expression was in a dynamically balanced range (Figure 10). These adaptations enhance efficient water utilization, helping L. yarkandensis cope with the Tarim Basin’s arid desert environment.