Quantitative Proteomic Study Unmasks Fibrinogen Pathway in Polycystic Liver Disease Ⅱ
Nov 22, 2023
3. Results
3.1. Kidney and Liver Does Not Follow the Same Cystic Mechanisms
Previous studies showed that early inactivation of the Pkd1 gene (before p12) triggers rapid development of polycystic disease (cystic window), whereas inactivation after p14 leads to late cyst formation after 4–5 months, with a mild phenotype (non-cystic window) [19]. Using the same orthologous model of ADPKD (Pkd1cond/cond; Tam-Cre) [18,19], we investigated whether this window of renal development corresponded in time with a similar window of hepatic development. To draw a direct comparison with our previous studies [37], mice were induced with tamoxifen to inactivate the Pkd1 gene at postnatal days 14 (p14) and 12 (p12) and sacrificed at the age of 30 days (p30). Interestingly, we observed a cystic phenotype at both time points with a different degree of disease (mild and severe), suggesting that the kidney and liver have independent cystic developmental windows (Figure 1a), and/or possible different timings and progression of the disease. No differences were found between males and females at this slaughter age between both windows of inactivation (n = 6 was used for each group of mutants). The p14 mutant animals presented a milder phenotype than p12 mutant group with lower values of cystic index, number of cysts and hepatic function according to ALP serum value (Figure 1b–d). This is the first study showing different developmental windows/mechanisms for Pkd1 liver and kidney deficiency. Those molecular differences will have to be determined in future studies.
Figure 1. Characterization of the cystic liver phenotype at day 30 form p12 and p14 stages. Pkd1 deletion at p12 carries a more severe cystogenesis. (a) Representative macroscopic and microscopic images of hematoxylin–eosin stained from livers of Pkd1cond/cond;Tam-Cre mouse Wild Type (WT, Pkd1cond/cond;Tam-Cre−) and Mutant (KO, Pkd1cond/cond; Tam-Cre+ ) with Pkd1 gene deletion induced by tamoxifen at postnatal day 14 (p14 group) and 12 (p12 group). All mice were sacrificed at p30. Scale bar, 2 mm (upper panel) 100 µm (lower panel). n represents the number of samples per group, and Bd means bile duct. (b,c) Hepatic cystic index and number of cysts of the different phenotypes. (d,e) Blood serum ALP and ALT values. Bars represent means ± SEM in all cases. p < 0.05 by Student’s t-test (two-tail) was considered as a significant result. ns represents not significance. * p < 0.05, ** p < 0.01, *** p < 0.001.
Click to cistanche herba for kidney disease
3.2. Shotgun and SWATH-MS Proteomic Analysis in PLD
As referred in Figure 2, we performed differential proteome analysis at both cystic stages (p12—severe cystic disease- and p14—mild cystic disease-) of Mutants (KO) in comparison to Wild Type (WT) animals, in order to identify and characterize relevant molecular mechanisms undergoing liver cystogenesis and disease progression.
Figure 2. Illustrated workflow scheme. We used an ADPKD murine model (Pkd1cond/cond; Tam‐Cre mice) in which the genetic inactivation of Pkd1 was induced by tamoxifen administration at postnatal day 10 (p10) and p11 (rapid disease progression) or p15 and p16 (delayed disease progression) according to the developmental switch for renal cystogenesis, defining the two groups of the study named p12 and p14 respectively [19]. For both groups, we used an n equal or greater than six wild-type (WT) and Mutant (KO) individuals in each condition, and all animals were sacrificed at p30. We studied the differential proteome of WT and KO livers in both severities of liver cystic disease (Figure 1), using both quantitative proteomic SWATH–MS and shotgun data‐dependent acquisition DDA‐MS analysis. Of the proteins with a significant change in abundance, we used bioinformatics Figure 2. Illustrated workflow scheme. We used an ADPKD murine model (Pkd1cond/cond;Tam-Cre mice) in which the genetic inactivation of Pkd1 was induced by tamoxifen administration at postnatal day 10 (p10) and p11 (rapid disease progression) or p15 and p16 (delayed disease progression) according to the developmental switch for renal cystogenesis, defining the two groups of the study named p12 and p14 respectively [19]. For both groups, we used an n equal or greater than six Wild Type (WT) and Mutant (KO) individuals in each condition, and all animals were sacrificed at p30. We studied the differential proteome of WT and KO livers in both severities of liver cystic disease (Figure 1), using both quantitative proteomic SWATH–MS and shotgun data-dependent acquisition DDA-MS analysis. Of the proteins with a significant change in abundance, we used bioinformatics tools to detect novel therapeutic targets and pathways involved in disease. Finally, we validated those targets using first in silico (DDA vs SWATH analysis) and finally in vivo strategies, as well as RT-qPCR, Western blotting and immunohistochemistry. p means postnatal day
Protein Expression Pattern in Hepatic Cystogenesis First, we performed proteomic mass spectrometry analysis by shotgun data-dependent acquisition DDA-MS or DDA. This analysis allows characterizing the proteome from complete and individual samples, providing the presence/absence of proteins with high sensitivity but without providing their quantification. We exclusively considered all the proteins expressed differentially in more than five independent samples from a total of six samples, for p14 and p12 stages of Wild Type and mutant individuals, respectively (Figure S1). At p14, 1 WT-exclusive (or down-regulated for the disease) and 7 MUTexclusive (up-regulated) proteins were identified, while at p12, eight WT-exclusive and six MUT-exclusive proteins were observed (Figure S1).
Next, we used a method that allowed us to quantify differentially expressed proteins. We performed a SWATH–MS quantitative proteomic analysis on the very same samples that we used for DDA, obtaining ~1500 proteins per sample and more than 2000 proteins in each study group. First, we performed total area sum (TAS) normalization to evaluate that our samples followed a normal distribution pattern, as shown in Figures S2 and S3. Next, we studied the protein with significant differences between Wild Type and Mutant groups. Table 1 shows the proteins with a significant change in abundance with a two-fold increase (up-regulated) or decrease (down-regulated), and a significant adjusted p-value (according to the parametric Student’s t-test) in WT vs MUT samples at p14 and p12. In total, six proteins showed significant differences in protein abundance between the WTp14 and MUT-p14 groups (five up-regulated proteins and one down-regulated proteins). Between WT-p12 and MUT-p12, 26 proteins (20 proteins up-regulated and 6 proteins down-regulated) showed significant differences (Figure 3a and Table 1). Graphically, these variations can be observed through volcano plots, which were generated by plotting the log(2)-fold changes for all proteins identified against their −log(10) p-value (Figure 3b). Significant differences among protein abundance levels can be visualized in the heat map with individualized values according to the areas of the spectral library, and its clustering level in the cluster heat map analysis (Figures 3c and S4). These clusters detected by the SWATH–MS analysis help us to identify those qualitatively separated samples, to be considered for the quantitative analysis (Figure S5). An unsupervised multivariate statistical analysis was performed using principal component analysis (PCA) to compare the data between samples (Figure S6). The heatmap and PCA data demonstrated reproducibility among the sample triplicates as the three replicates closely clustered in both analyses. We can observe both in the PCA and in the heatmap cluster analysis (Figures S5 and S6) how some samples did not group correctly, overlapping between both clusters. Nonetheless, we can see the tendency to separate data in Wild Type and Mutant clusters. The lower number of proteins with significant differences in p14 vs p12 could be justified by the different degree of severity (Figure 2), suggesting that the number of proteins and pathways increases based on the severity and disease status of the cystic phenotype.
3.3. Clustering, Pathway Enrichment and Protein-Protein Interaction Analysis of Gene Expression in PLD
To establish the function related to the differentially expressed proteins identified by SWATH-MS analysis, we used several bioinformatic tools and forms of analyses. Gene Ontology (GO) terms and pathways analyses were performed using FunRich [39,40], String [41] and Reactome [42]. The proteins were sorted in FunRich gene enrichment analysis. Regarding the biological process, the most enriched group of proteins for p14 group (mild cystic) were related to fatty acid transport and prostaglandin biosynthesis, and for p12 (severe cystic) group with fibrinogen/fibrin activity, cell–cell/matrix adhesion and metabolic process (Figure 4a, upper panel). ANXA2 and the fibrinogen complex were the two most enriched cellular components in p14 and p12 groups, respectively. Furthermore, the extracellular space was the cellular component with the highest percentage of proteins in both groups (Figure 4a, lower panel). Similarly, the same analysis was established with string analysis, identifying different metabolic processes as the main biological process, and the extracellular space as the most enriched cell component; consistent with GO results (Figure S7a). Finally, we repeated the analysis with the Reactome platform, finding that metabolism and immune system were the most enriched pathways (Figure S7b).
Protein–protein or cluster analysis based on string analysis was also performed in order to investigate possible protein–protein interactions (interactions with experimental evidence). Numerous protein interactions were identified in different groups (Figure S8), but one cluster at p14 and another at p12 were the most relevant based on the number of interactions between proteins and the level of expression (Figure 4b). The p14-cluster contains annexin A2 protein (ANXA2_P07356) associated with several cellular functions, such as fibrinolysis, cell motility (in epithelial cells), a protein related to actin cytoskeleton and cell-matrix interactions [43], interacting with two other proteins involved in the intracellular lipid transport (FABP1_P12710 and FABP5_Q05816). Interestingly, another relevant group of proteins was identified in the p12-cluster with a very strong protein–protein interaction between several fibrinogens (FGL1_Q71KU9, FIBB_Q8K0E8, FIBA_E9PV24 and FIBGG_Q8VCM7). Fibrinogens are involved in hepatocyte growth, and they mediate blood platelet spreading, interstitial collagen and fibrotic lesions [44]. Furthermore, additional proteins were found to interact with fibrinogens (Figure S8), such as serum amyloid p-component (SAMP_P12246), hemopexin (HEMO_Q91X72) or hydroxy acid oxidase 2 (HAOX2_Q9NYQ2). Consistently with string results, ANXA2 and fibrinogen complexes were also the most enriched cellular components identified by FunRich analysis (Figure 4a). Interestingly, when comparing the data obtained by DDA and SWATH analysis, we confirmed that several proteins identified by SWATH were also identified by DDA analysis, including several related to fibrinogen complex proteins (SAMP/Apcs and S10A9/S100a9; Table S3).
3.4. Validation of SWATH–MS Analysis Unmasks the Fibrinogen Complex as a New Molecular Mechanism Related to PLD
Based on these data, we selected from the SWATH–MS analysis (Figure S9) the altered proteins related to the fibrinogen complex at p12 and p14 stages for in vivo validation with RT-qPCR, Western blotting and immunohistochemistry, to establish their confirmation as possible targets of disease. Eight out of 10 of these selected proteins were validated at mRNA level by RT-qPCR (Figure 5); one up-regulated at the 14-stage (ANXA2_P07356) and at p12-group (SAMP_P12246, FGL1_Q71KU9, ILK_O55222, S10A9_P31725, FIBB_Q8K0E8, HEMO_Q91X72, FIBA_E9PV24 and FIBG_Q8VCM7), and one down-regulated at the p12 group (HAOX2_Q9NYQ2). Interestingly, gene expression of the two proteins with more distance or fewer interactions to the fibrinogen group (ILK and S10A9) were the only ones that were not validated (Figure 5c).
Up-regulation of the fibrinogen complex (ANXA2, SAMP, FIBB, FIBA, FIBG and HAOX2 proteins) was further confirmed by Western blotting (WB) analysis (Figure 6). Overall, gene expression and WB were highly consistent with SWATH-MS proteomic data. Finally, we also performed immunohistochemistry validation of the fibrinogen complex in mutant and polycystic liver samples (Figure 7). Fibrinogen staining was strongly up-regulated in cyst-lining epithelial cells and bile duct dilatations (Figure 7a). These results unmask, for the first time, the fibrinogen complex as a possible pathway related to hepatic cystogenesis and as a possible future therapeutic target for PLD.
Figure 5. Validation of possible PLD targets by RT-qPCR. Gene expression of 8 to 10 selected targets correlated with SWATH-MS data. RT-qPCRs of up-regulated selected p14 target annexin A2 (Anxa2) (a); up-regulated p12 targets amyloid p component, serum (Apcs, SAMP gene), fibrinogen-like 1 (Fgl1), fibrinogen beta chain (Fgb), hemopexin (Hpx), fibrinogen alpha chain (Fga), fibrinogen gamma chain (Fgg) (b), integrin linked kinase (Ilk) and S100 calcium binding protein A9 (S100a9) (c); and down-regulated p12 target hydroxy acid oxidase 2 (Hao2) (d). n = 6 was used for each protein group. Gapdh was used as a housekeeping gene. Bars represent means ± SEM. Student’s t-test with two-tail was used and a value of p < 0.05 was considered significant. ns: not significant (p ≥ 0.05), * p < 0.05, ** p < 0.01, *** p < 0.001.
Figure 7. Fibrinogen complex is up-regulated in the cystic epithelia. (a,b) Representative images of fibrinogen immunohistochemistry in bile duct (upper panel) and liver parenchyma (lower panel) (a) and its quantification (b). n = 6 was used for each group. The immunohistochemistry analysis showed an up-regulation of fibrinogen through cystic epithelia. The samples corresponded to p12 group. Scale bars represent 100 µm. Bd means bile duct. Bars represent means ± SEM. Student’s t-test with two-tail was used and a value of p < 0.05 was considered significant. *** p < 0.001.
Supportive Service Of Wecistanche-The largest cistanche exporter in the China:
Email:wallence.suen@wecistanche.com
Whatsapp/Tel:+86 15292862950
Shop For More Specifications Details:
https://www.xjcistanche.com/cistanche-shop
