• Users Online: 365
  • Print this page
  • Email this page


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2022  |  Volume : 6  |  Issue : 2  |  Page : 266-277

Integrated proteomic, transcriptomic, and genomic analysis identifies fibrinogen beta and fibrinogen gamma as key modulators of breast cancer progression and metastasis


1 Department of Microbiology, Faculty of Applied Sciences, Parul University; Department of Microbiology, Parul Institute of Applied Sciences, Parul University; Department of Paramedical and Health Sciences, Faculty of Medicine, Parul University, Vadodara, Gujarat, India
2 Department of Microbiology, Faculty of Applied Sciences, Parul University; Department of Paramedical and Health Sciences, Faculty of Medicine, Parul University, Vadodara, Gujarat, India
3 Department of Microbiology, Faculty of Applied Sciences, Parul University, Vadodara, Gujarat, India

Date of Submission11-Mar-2022
Date of Acceptance28-May-2022
Date of Web Publication17-Jun-2022

Correspondence Address:
Hemantkumar Patadia
Parul University, Limda, Waghodia, Vadodara - 391 760, Gujarat
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/bbrj.bbrj_61_22

Rights and Permissions
  Abstract 


Background: As per the WHO, more than 2 million new cases are diagnosed with breast cancer and more than 685,000 deaths have been reported in 2020. Incidence of recurrence and metastasis has also risen and poses a challenge for developing new therapies with the identification of newer targets. Methods: The objective of this study is to analyze and integrate various data across genomic, transcriptomic, and proteomic levels to find promising markers linked with tumor progression and metastasis development in breast cancer. This study begins with the extraction of data of differentially expressed proteins and subsequently analyzes their gene expression changes and genomic alterations to integrate all three omics data. We used curated breast cancer datasets of different proteomic experiments from dbDEPC3.0 and TCGA datasets of the Metastatic Breast Cancer Project to compare gene expressions and genomic alterations. We further used cBioportal, GeneMania, GEPIA2, Reactome, and canSAR computational tools for identifying the most significant genes associated with tumor progression and metastasis. Results: Based on false discovery rate, 14 genes were subjected to pathway analysis in Reactome and the top 25 significant pathways were analyzed out of a total of 170 pathways. Our study found fibrinogen gamma (FGG) and fibrinogen beta (FGB) linked to pathways connecting RAS-MAPK, its downstream mutants, integrin signaling, and extracellular matrix remodeling pathways. Conclusion: Survival analysis suggested that FGG (P = 0.0065) and FGB (P = 0.013) have a significant positive correlation along with their stage-wise changes in copy number variations and genomic alterations play a pivotal role in controlling tumor progression and metastasis.

Keywords: Breast cancer metastasis, fibrinogen/fibrinogen gamma/fibrinogen beta, gene expression, proteomic, transcriptomic analysis


How to cite this article:
Patadia H, Priyadarshini A, Gangawane A. Integrated proteomic, transcriptomic, and genomic analysis identifies fibrinogen beta and fibrinogen gamma as key modulators of breast cancer progression and metastasis. Biomed Biotechnol Res J 2022;6:266-77

How to cite this URL:
Patadia H, Priyadarshini A, Gangawane A. Integrated proteomic, transcriptomic, and genomic analysis identifies fibrinogen beta and fibrinogen gamma as key modulators of breast cancer progression and metastasis. Biomed Biotechnol Res J [serial online] 2022 [cited 2022 Dec 9];6:266-77. Available from: https://www.bmbtrj.org/text.asp?2022/6/2/266/347723




  Introduction Top


As per the WHO, more than 2 million new cases are diagnosed with breast cancer and more than 685,000 deaths have been reported globally in 2020. As an estimate, there have been more than 7.8 million women survivors of breast cancer in the past 5 years globally.[1] Breast cancer mortality and treatment outcomes are linked with early diagnosis of the disease and genetic predisposition to mutations in BRCA1 and BRCA2. If the treatment is initiated early, survival probabilities can be significantly improved. Over several decades, medical treatments combined with surgical resections have improved the disease management and quality of life in patients with breast cancer. According to a report in 2018, more than 1.6 lakhs new cases and more than 87,000 deaths were newly reported in India.[2] Further, a projected incidence report of 2020 indicates a total incidence of more than 2 lakh cases of breast cancer and a cumulative risk of 1 in 29 women in India, making it the most prevalent form of cancer.[3] Breast cancer is prevalent at a rate of 1 in 28 women in rural area and 1 in 22 in urban area. It is now the most common type of cancer found in Indian women, with approximately more than 50% of them being diagnosed at stages 3 and 4. Hence, an increased number of efforts have been channelized to reduce the mortality rate of breast cancer across the world. Among others, WHO Global Breast Cancer Initiative[4] aims to decrease this number by 2.5% every year amounting to a total of 40% reduction in mortality by the year 2040.

Early detection and therapy have been shown to be effective in increasing the life span of patients; however, increased incidences of recurrence and metastasis development have posed severe challenges in the success of prevailing therapeutic modalities. Hence, there is a pressing need to find newer prognostic and therapeutic markers that give alternatives to enhance or find more effective therapeutic options. Many integrated large-scale experimental studies have been conducted to look at mutations, copy number variations (CNVs), and differential gene expression under various experimental conditions, and their datasets are made publicly available for the community to extract newer information. One such study was done to identify molecular signatures that can serve as a therapeutic target in the treatment of mild cognitive impairment, the study uses the disease gene datasets and involves bioinformatics tools to identify the novel targets.[5] The genomic, transcriptomic, and proteomic datasets available for various diseases based on different experimental designs give us a broader outlook to find out novel biomarkers for disease progression and metastasis.[6] Moreover, the expression analysis of gene of interest and miRNA helps to provide the linkages in biological networks and pathways.[7] Thus, they can be targeted as a prognostic factor for early-stage detection of various cancer types and serve as a diagnosis prospective, and can show applicability in drug development.[8],[9] For breast cancer, 24 such datasets from different studies can be accessed through TCGA[10] and cBioportal,[11],[12] thereby increasing the type and number of patient samples being analyzed with respect to various clinical pathological conditions. For this study, we chose to work with provisional data submitted from the “Metastatic Breast Cancer Project” deposited in February 2020.[13],[14] This study has extracted information from 237 samples for tumor and metastatic specimens. Whole-exome sequencing was carried out on all 237 samples, whereas RNA sequencing was carried out on 146. The samples were collected from a total of 180 patients who developed metastatic breast cancer providing currently the most appropriate set of data available for our experimental questions.[13],[14]

In the present study, we manually curated and selected the candidates that were common in normal versus tumor and metastasis. Those 153 proteins served as our query set for further analysis to check their alterations at genomic and transcriptomic levels and to identify their occurrence in different stages of the disease. Further, we scrutinized and found the candidates such as proliferating cell nuclear antigen (PCNA), SERPINA1, topoisomerase II α (TOP2A), bone marrow stromal cells antigen 2 (BST2), and interferon-induced transmembrane protein 1 (IFITM1) that can be the potential targets and also reported earlier to be involved in important core pathways such as cell cycle regulation, protein processing, interferon signaling, and innate immune response to be enriched in metastasis and disease progression. However, there was no significant difference in the survival analysis in high versus low-expression groups of these candidates. Finally, we found three candidates, i.e., fibrinogen gamma (FGG), fibrinogen beta (FGB), and ITGA6 that were upregulated in pathways involved in oncogene signaling, integrin signaling, and extracellular matrix (ECM) remodeling with functional module enrichment such as zymogen activation, regulation of heterotypic cell–cell adhesion, regulation of apoptosis, and regulation of vasoconstriction. Moreover, the pathways that are interlinked with these three candidates can be targeted to increase the effectiveness in the treatment of breast cancer. We hypothesize that FGG and FGB serve as the modulator of the metastatic stage as their upregulation can be studied at the genomic and proteomic levels. In recent past, several studies have shown increased levels of FGG in different types of cancer such as gastric cancer, liver cancer, colon cancer, and ovarian and renal cancers.[15],[16],[17] Moreover, overall survival and disease-free survival are shorter in case of patients with higher levels of plasma fibrinogen in the pretreatment phase.[18]

In one of the studies, FGG was seen to be participating in chemoresistance against anthracyclines used for breast cancer treatment, and hence, tumor cells having high expressions of FGG showed a weaker response to anthracycline chemotreatment for breast cancer.[19] Thus, our findings suggest an important link between RAS signaling, ECM remodeling, integrin signaling, and development of metastasis in breast cancer mediated through the involvement and regulation of FGG and FGB expression. These findings can pave the way for exploring mechanistic insights into validating the role of FGG and FGB or its interactors as prognostic and therapeutic markers in breast cancer metastasis.


  Materials and Methods Top


Materials

●A high-performance computer with high-speed Internet connection, an updated version of browsers that support HTML5 and JavaScript is needed

●Hardware: Due to a large amount of data to be analyzed and visualized, recommended specifications will be a minimum of 4GB DDR2 RAM, Intel Core i7 processor or its equivalent, and a 17-inch screen display with a minimum 1440 × 900 resolution

●Software: Microsoft Office 365 package, Adobe Acrobat package, and Adobe Illustrator

●Data: Required datasets of all breast cancer studies from dbDEPC3.0, cBioportal, and GEPIA2.

Sampling strategy

NIL.

Plan of action

Identification of candidates involved in BRCA metastasis

The protein expression data in breast cancer were obtained from dbDEPC.[20] The compiled data were divided into two groups, normal versus cancer group (BRCA) (all experimental studies in which proteomic analysis has been done under different conditions and on different samples to compare healthy breast tissue and breast tumor) and metastasis group (META) (all experimental studies in which proteomic analysis has been done to compare different responses, interventions, pathological parameters, and clinical parameters between primary breast tumor samples and metastasized samples/conditions), each having subgroups, upregulated protein expression profile (UP) and downregulated protein expression profile (DOWN). The data of two groups were converted from Uniprot to Entrez ID and were checked for common genes using Venny 3.0.[21] A comparison of these groups was done further, which gave the common candidates between two groups collated by expression level individually. Only the upregulated group data were chosen for further analysis, as it may serve as an opportunity for targeted studies. This set of upregulated candidates was then further studied to check transcription level expression changes in the database of BRCA Metastasis in cBioportal. cBioportal is a platform that contains cancer genomics datasets from clinical samples across 20 cancer studies. OncoPrint for patients with metastasis and without metastasis was created and compared side by side. Candidates were shortlisted for next-level analysis.

Deregulatory gene expression analysis

The shortlisted candidates obtained from level 1 were analyzed under three different conditions:

  1. Boxplot was plotted in GEPIA 2[22] to check differential expression in BRCA dataset for tumor versus normal at mRNA level
  2. Query genes were checked for mutual exclusivity and co-occurrence in cBioportal
  3. Expression volcano plots and enrichment graphs were created in cBioportal to identify significant changes in altered and unaltered groups.


Network analysis and pathway analysis

Using GeneMania,[23] a concentric circle network diagram was drawn and function enriched based on false discovery rate (FDR) and coverage, important network pathways were selected. For doing pathway analysis, Reactome (database of biological pathways)[24] was used. Based on P values, candidates were shortlisted.

Biological functions and survival analysis

Graphs of disease-free survival analysis were plotted using GEPIA 2[22] and curated based on P values. The candidates were checked for their biological functions in GeneCards (a human gene database),[25],[26] which has a compilation of various types of information about the genes.

Tumor stage-specific variations and copy number variations

The shortlisted candidates were used as an individual input in canSAR[27] to study their role in stage-wise progression of breast cancer disease via violin plot. To identify the gain, loss, amplification, and deletion of shortlisted gene candidates, we also studied the CNV in breast cancer from canSAR. The CNV was observed stage wise and results were drawn out.

Statistical analysis

For data analysis during differential gene expression, mutual exclusivity, co-occurrence, enrichment graphs, and network and pathway analysis, P < 0.05 is considered the minimum threshold for significance in our analysis. Any deviations from this value are denoted in the appropriate sections.


  Results Top


The primary objective of our study is to identify differentially expressed candidates in breast cancer metastasis which can be linked to RAS signaling pathways, thus involved with oncogene addiction of malignant transformation processes and affecting overall survival. The pipeline of analysis is outlined as a schematic diagram in [Figure 1].
Figure 1: Flowchart diagram depicting the workflow and scheme of data collection, filtering, processing, and analysis

Click here to view


Identification of differentially expressed proteins related to metastasis in breast cancer (BRCA)

The tumor development, stage-wise progression, and subsequent metastasis during breast cancer project an important challenge in the intervention and management of effective treatment plans. Since the incidences of recurrence and metastasis development in BRCA are widespread, we aimed at identifying candidates specifically involved with metastasis development in BRCA. To keep open the possibility of proposing newer targets for therapeutic interventions, we started our search from the set of differentially expressed proteins (DEPs) from datasets of various proteomics studies available on dbDEPC 3.0. Out of 6866 BRCA DEPs curated across different experimental designs such as normal versus cancer, cancer (stage X) versus cancer (stage Y), BRCA (no treatment vs. treatment), and BRCA metastasis, we chose to compare DEPs from studies of normal versus cancer (2574) and BRCA Metastasis (1021). We excluded data of studies done on triple-negative breast cancer subtypes since this condition is only checked for the former comparator. In the first condition (normal vs. cancer), we found 938 DEPs downregulated and 1636 DEPs upregulated. Similarly in the second condition (BRCA metastasis), we found 485 DEPs downregulated and 536 DEPs upregulated. Many of the candidates were represented more than once in both the conditions across different studies. Hence, we further created a matrix of comparison of upregulated and downregulated DEPs by converting the gene symbols into unique Entrez ID identifiers. As shown in [Figure 2]a, we found 155 common DEPs upregulated in both the conditions (BRCA UP and META UP). Further, we found 169 common DEPs upregulated in BRCA (BRCA UP) and downregulated in metastasis (META DOWN). Similarly, we found 129 common DEPs downregulated in BRCA and upregulated in META and 126 DEPs downregulated across BRCA DOWN and META DOWN. Since upregulated proteins present a probable option of serving as interventional targets, we chose to further analyze the DEPs of the first quadrant in [Figure 2]a at the transcriptomic level. We performed OncoPrint analysis on RNA set data obtained from the “The Breast Cancer Metastatic project (Provisional, Feb 2020)”[13],[14] of 146 samples submitted to TCGA using the query platform of cBioportal. When the track identifiers were set to PATH sample in metastatic setting and clinical identifiers set to MedR Ever Metastatic Sites, we generated a heatmap of 153 unique genes showing differential expression in individual patient samples when the Z-score threshold was set to ± 1.5. As whown in [Figure 2]b, many samples show alteration frequencies in metastatic as well as nonmetastatic settings. Hence, we compared genes that are affected in patients with metastasis versus patients without metastasis to identify candidates having significantly higher alteration frequency in the former condition. We shortlisted the top 14 candidate genes having alteration frequency in the range of 8%–17% [Figure 2]c and difference in percentage in the range of 6%–17% [Supplementary Table 1]. [Alteration frequency = percentage of number of patients in which gene expression is affected out of all the patients analyzed as obtained from OncoPrint analysis].
Figure 2: Identification of candidates upregulated during tumor development and metastasis in breast cancer. All the common differentially expressed proteins under up- and downregulated conditions were extracted (a) and queried in cBioPortal-OncoPrint to identify candidate genes which showed differential expression (b). Further, the common 14 sets of candidate genes were analyzed for their alteration frequency across clinical samples (c)

Click here to view



Deregulatory gene expression analysis shows differential regulation of candidates at transcription and translational levels suggesting an important role in tumor development and metastasis

In our study, the changes across genomic, transcriptomic and proteomic levels were integrated to find relevant candidate genes. Since we started with the identification of differentially expressed proteins which are consistently upregulated during BRCA tumor development and BRCA metastasis, we subsequently investigated the expression levels of these genes in the initial stage of tumor development as compared to the tissue-matched normal samples. Analysis of all 153 candidate genes represented a classic volcano plot of log ratio of alteration on x-axis versus significance represented by -log 10 P value on y-axis [Figure 3]a. The cumulative analysis of 14 gene set signatures for differential gene expression analysis in GEPIA2 using a log2FC cutoff: 1 and P value cutoff: 0.01 shows the differential expression of the group to be higher across n = 1085 tumor tissues as compared to n = 291 normal tissues [Figure 3]a. However, individual box-plot analysis of each of the genes reveals three contrasting outputs [Figure 3]b. First, the genes such as BST2, GLO1, PCNA, SERPINA1, and TOP2A are significantly upregulated in tumor tissues as compared to normal tissues. Second, the genes such as ATP5B, COL19A1, ENO2, FGB, FGG, and IFITM1 have no significant difference in gene expression between tumor and normal tissues. Third, the genes such as ALDH2, DPYSL2, and ITGA6 are significantly downregulated in tumor tissues as compared to normal tissues. Furthermore, while comparing the expression of 14 genes among themselves in a pairwise fashion, the analysis tested 91 pairs between 14 tracks across 146 patient samples based on P value derived from the one-sided Fisher's exact test and q value derived from Benjamini–Hochberg FDR correction procedure in OncoPrint (cBioportal) and found 10 significant pairs which shows co-occurrence [Supplementary Table 2]. The FGB-FGG and ATP5B-PCNA show the highest tendency to co-occur. It is notable that the same set of proteins was found to be upregulated in conditions of tumor development and metastasis in our preliminary search which categorizes into three different conditions of comparative gene expression in the tumor development stage suggesting important roles of transcriptional, posttranscriptional, and translational regulation governing the formation of their protein products. These results indicate that our queried set of genes is differentially regulated at transcriptional and translational levels suggesting an important role in the process of tumor development and metastasis.
Figure 3: Gene expression analysis of candidates between normal and tumor tissues. To find significantly altered genes, a volcano plot was generated of the entire query of upregulated genes (153) and box plot for the set of 14 genes (a). Individual box plots for each gene in normal versus tumor tissues are shown in (b)

Click here to view



Network and pathway analysis reveals enrichment of specific modular functions and linkage with regulations of RAS-MAPK pathway and modulation of extracellular matrix and Integrin signaling

As shown in [Figure 4]a, the above-mentioned 14 genes were used as an input for generating a multilayer network based on co-expression, physical interaction, pathways, colocalization, and genetic interaction in GeneMania assigning the weights automatically based on query gene using linear regression. The network construction was limited to default 10 attributes and 20 additional interactors to extract meaningful information out of our network. This resulted in 264 interaction pairs of which 54.22% of the network was mapped by coexpression, 38.56% was mapped by physical interaction, 6.29% was mapped by pathways, 0.84% by colocalization, and 0.09% by genetic interactions [Figure 4]a. Functional enrichment analysis (FEA) of the genes in the network was done based on the GO categories and q values from a FDR corrected hypergeometric test using the Benjamini–Hochberg procedure along with the coverage ratios for the number of annotated genes in the displayed network versus the number with that annotations in the genome. Seventy-fiv such FEA categories are listed in [Supplementary Table 3] with a q-value cutoff of 0.1. As shown in [Figure 4]a, four categories represented in the network are (i) zymogen activation (FDR = 7.34, e-3/coverage = 4/35), (ii) regulation of heterotypic cell–cell adhesion (FDR = 1.64 e-2/coverage = 3/24), (iii) negative regulations of apoptotic signaling pathway (FDR = 1.64e-2/coverage = 5/159), and (iv) regulation of vasoconstriction (FDR = 1.64e-2/coverage = 3/24). Based on the strongest correlation obtained between FGG and FGB in section 3.2 above, the FGG centric network is highlighted in [Figure 4]a.
Figure 4: Network and pathway analysis for modular functions enrichment with gene significance. Network of queried 14 genes with an overlay of selected modular functions is shown (a). Overlay and details of all enriched pathways of selected genes are shown in (b and c). c(i) Overlay of enriched modular function in MAP2K and MAPK activation circuit. c(ii) Overlay of enriched modular function downstream of BRAF mutations. c(iii) Integration of enriched BRAF signaling with MAP2K/MAPK signaling circuit

Click here to view



[Figure 4]b depicts an over-representation analysis generated with the help of a statistical hypergeometric distribution test that determines whether certain Reactome pathways are enriched in our submitted data. The test produces a probability score, which is corrected for FDR using the Benjamini–Hochberg method. In the given analysis, all 14 identifiers were found in the Reactome, hitting 170 pathways by at least one of them. The list of 25 most relevant pathways sorted by P values can be found in [Supplementary Figure 1]. Out of these pathways, FGG and FGB were mapped to the maximum number of pathways suggesting them to be important hubs in the network. We have represented pathways interlinked between RAS signaling and ECM organization. The pathways enriched toward ECM remodeling and integrin signaling are GRB2:SOS provides linkage to MAPK signaling for integrins (R-HSA-354194), p130Cas linkage to MAPK signaling for integrins (R-HSA-372708), ECM organization (R-HSA-1474244), and Integrin signaling (R-HSA-354192). The pathways enriched toward connection with RAS signaling are signaling by high-kinase activity BRAF mutants (R-HSA-6802948), MAP2K and MAPK activation (R-HSA-5674135), and signaling by RAF1 mutants (R-HSA-9656223). [Figure 4]c represents a compiled schematic of different nodules in the RAS signaling pathway, and the modules affecting RAF and MAP2K-MAPK activation are represented in [Figure 4]c (i), those affecting signaling by RAF1 mutants in [Figure 4]c (ii), and those representing signal transduction by high-kinase activity of BRAF mutants in [Figure 4]c (iii). In addition to FGG and FGB, the other interactors in these pathway cascades were also reflected in our original data sets of differentially expressed proteins and differentially expressed genes (data not shown) suggesting a systemic involvement and interlinking of these pathways during metastasis promotion. Since FGG, FGB, SERPINA1, ITGA6, and ALDH2 are represented across multiple pathways, they can be considered putative targets for further modulation and management of the disease.



Survival analysis of putative targets and their biological functions in the context of fibrinogen gamma and fibrinogen beta

Genes that are present in similar pathways may have a strong correlation in their expression profile and may affect the overall disease-free survival of the patients when perturbed. To test this, we retrieved disease-free survival curves for FGG, FGB, SERPINA1, ITGA6, and ALDH2. Except FGG and FGB, there is no significant change in the overall disease-free survival as seen by calculation of significance on Hazard's ratioand log-rank P value between high-expression groups and low-expression groups. As shown in [Figure 5]a, differential expression of both FGG and FGB has a significant impact on the percentage of disease-free survival and duration. Rest of the genes' differential expression yields insignificant/minimal differences (data not shown). Moreover, there is a strong positive correlation seen in the expression profile of FGG and FGB as well as FGG and FGA seen by Spearman's correlation and Pearson coefficient [Figure 5]b. Rest gene pairs do not show any correlation in their expression (data not shown). FGA, FGB, and FGG stand for fibrinogen alpha chain, beta chain, and gamma chain, respectively. They encode subunits of fibrinogen that are proteolytically cleaved by thrombin and play a role in clot formation. There are many congenital disorders reported due to mutation in these components and the proteolytic activation leading to fatality. However, increasing light of knowledge in the regulation of expression and post transcriptional regulation of these genes reveals their involvement in many different pathways and also cancer.
Figure 5: Survival analysis for genes fibrinogen beta and fibrinogen beta along with intergenic correlation analysis. High- and low-expression group of fibrinogen beta and fibrinogen beta has a significant difference in overall disease-free survival (a) and their expression is positively correlated with each other (b)

Click here to view


The same can be seen through differential expression of FGG, FGB, and FGA transcripts across different stages of breast cancer and the changes in their CNV profile [Figure 6]. Further regulation in their function with the context of cancer and metastasis can be achieved by the transcription factors that regulate their expression for example, FGA gene promoter contains binding sites for transcription factors E47, Evi-1, FOXO3 and its isoforms, ISGF-3, Pbx1a, STAT3 and Tal-1. Similarly, FGB gene promoter contains binding sites for transcription factors NF-AT and its variants, Pax-4a, POU3F2 and FGG gene promoter contains binding sites for transcription factors E-47, FOXJ2, STAT3, Tal-1, USF-1 and USF-2. Transcription factor STAT3 which regulates the transcriptions of FGB and FGG genes has been shown previously to be associated with EGFR and MAPK3 pathways. Similarly, transcription factor Tal-1 is associated with acute lymphoblastic leukemia reported earlier,[25],[26] which regulates the transcriptions of FGB and FGG genes. Evi-1 (MECOM) TF-regulating FGA gene is found to be associated with acute myeloid leukemia.[25],[26] This establishes their instrumental role in the contribution of breast cancer development, progression, and metastasis.
Figure 6: FG, fibrinogen beta, FGA expression in different stages of BRCA and their copy number variation profiling. Despite being found as upregulated differentially expressed proteins, expressions of fibrinogen beta, fibrinogen beta, FGA are lower across different stages (a), also reflected in their copy number variation profiling (b)

Click here to view



  Discussion Top


Increasing incidences of tumor recurrence, treatment resistance, and metastasis are posing major threats to improve the prognosis of disease management and outcome for patients with breast cancer.[28] The problem is pervasive irrespective of the pathological subtypes, stage/grade, age of diagnosis, and sites of metastasis in breast cancer.[28] Many studies are ongoing that explore the system's view or personalized approach to determine the best possible treatment option for advanced-stage breast cancer. However, the multigenicity and phenotypic heterogeneity pose a great challenge in improving the success ratio of such efforts. Recently, many such large-scale studies have paved ways to look at the disease at genomic, transcriptomic, and proteomic levels in breast cancer.[10] The data generated out of the studies serves as a valuable resource of information to identify biomarkers that may be helpful in (1) our understanding of the disease, (2) molecular changes occurring across different patient cohorts, (3) correlation of genetic subtypes and clinical characteristics, (4) predicting response and sensitivity toward adjuvant and/or immunotherapy, and (5) discovery of novel biomarkers for better prognosis and newer therapeutic targets.

In this study, we aimed to explore differentially expressed proteins that correlate with tumor progression and metastasis in breast cancer and further investigate their alterations at genomic and transcriptional levels. We found 2574 DEPs in normal versus cancer studies (938 – downregulated, 1636 – upregulated) and metastasis studies (485 – downregulated, 536 – upregulated) of BRCA. After filtering for unique identifiers and creating a 2 × 2 matrix for finding overlapping sets of up- and downregulated proteins in the two conditions, we chose to further explore 155 DEPs in upregulated conditions across both the transformational stages. From this list of proteins, our study found 14 sets of DEGs based on their expression profiling in metastatic versus nonmetastatic tumor samples. Due to the inherent presence of tumor heterogeneity, these DEGs showed nonproportionate correlation in tumor versus normal transcriptomic profiling. Some of the candidates such as ALDH2, DPYSL2, and ITGA6 which showed upregulated protein levels during tumorigenesis and metastasis were downregulated at transcription level during tumorigenesis as compared to tissue types normal sample. This suggests a negative transcriptional but positive post transcriptional and/or translational regulation during the corresponding phenotypic changes in BRCA. On the contrary, other candidates such as BST2, GLO1, PCNA, SERPINA1, and TOP2A show consistent upregulation at the transcriptional and translational stages. Interestingly, the candidates ATP5B, COL19A1, ENO2, FGB, FGG, and IFITM1 showed no significant difference in gene expression of normal versus tumor samples despite their proteins being upregulated. This hints at a possible posttranscriptional regulation of these candidates to be essential for tumor progression and metastatic transition. Network analysis and functional annotations of these candidates reveal important significantly functional clusters such as zymogen activation, antibacterial humoral response, protein processing, regulation of heterotypic cells and adhesion, negative regulation of apoptotic signaling pathway, regulation of vasoconstriction/hemostasis/coagulation, cell cycle control, glycolysis, and other metabolic processes. On the other hand, these modules were enriched for some common cancer signaling pathways such as (i) GRB2:SOS, (ii) MAPK signaling for integrins, (iii) p130Cas signaling for integrins, (iv) MAP2K and MAPK activation, (v) signaling by RAF1 and BRAF mutants, (vi) cell cycle control pathways for G0 and early G1, (vii) Mitotic G1 and G1/S transition, (viii) SUMOylation of DNA replication proteins. A significant enrichment was also found for specific immune response pathways such as interferon alpha/beta signaling, MyD22 deficiency (TLR2/4), IRAK4 deficiency (TLR2/4), regulation, and disease associated with TLR signaling. Most interestingly, our attention was drawn to pathways that regulate hemodynamics such as platelet degranulation, fibrin clot formation, and pathways connected to ECM remodeling such as integrin signaling and integrin cell surface interactions. We also found that some previously annotated prognostic factors across different cancers such as PCNA and DNA TOP2A were included in the identified pathways (find reference PCNA, TOP2A as prognostic factors in diff cancers). Tumor heterogeneity is an important aspect of cancer development that makes it feasible to explore the role of novel candidates related to the prognosis of breast cancer patients. Our study hinted at the possibility of BST2 and IFITM1 to be interesting to explore further. However, BST2 and IFITM1 present poor prognostic correlation as seen from Kaplan–Meier survival curves in high and low expressed groups. Remarkably, we found other core genes linked to ECM remodeling that may promote metastasis of breast cancer cells. Since many triggers can lead to epithelial-mesenchymal transition (EMT), we focused on genes enriched in our module linked to core functions of ECM. Genes such as COLN19A1, ITGA6, FGG, and FGB are mapped to multiple Reactome pathways which interlink ECM remodeling and different stages of downstream signaling to RAS. To identify which of them could be potential markers, we analyzed the expression of these genes with disease-free survival outcomes. We found only FGG and FGB to be significantly correlated with prognostic outcomes in patients with breast cancer. However, the roles of COL19A and ITGA6 in cellular remodeling cannot be neglected. Cancer such as prostate cancer, lung cancer, hepatocellular carcinoma, and pancreatic cancer showed differential expression of FGG and FGB that can serve as a potential tool to predict the prognosis of disease and helps in early diagnosis.[29] One of the studies also showed that by activating EMT, FGG serves as a modulator for the migration and invasion of hepatocellular carcinoma cells.[30] Additional evidence proved that elevated expression of FGG in tissue prognosticates poor survival estimation of breast cancer disease. The role of STAT3 as an activator in FGG regulation was confirmed by infecting parental MCF7 cells with an activated STAT3 construct.[19]

FGB is inactivated by matrix protein (MMP13)[31] and FGG is inactivated by MMP13 and MMP14.[32] Similarly, FGA is inactivated by MMP8, MMP12, MMP13, and MMP14.[31] Since MMPs have a known role to play in ECM modulation, the inactivation of FGG, FGB, and FGA by them can also be indicated by their significant involvement in breast cancer metastasis. Thus, our study through integrated genomic-transcriptomic and proteomic analysis reveals the important role of FGG and FGB in breast cancer tumorigenesis, progression, and metastasis.


  Conclusion Top


Increasing evidence has demonstrated the role of FGG and FGB in ECM modulation and controlling different facets of cancer progression and response to therapy in hepatocellular cancer, small cell lung cancer, non-small cell lung cancer, and prostate cancer. Our study highlights their importance as prognostic markers and modulators of metastasis by integrating data from genomic alterations, transcriptomic studies of different datasets, and curated proteomic analysis through different experiments. In addition, interactors of these candidates and their intertwined pathways could be targets of modulation and intervention to improve clinical outcomes of breast cancer treatments. We plan to explore mechanistic insights of these candidates in breast cancer in future.

Limitations of the study

Along with the potential of exploring multiple datasets generated from different experiments for extracting new information, there are certain limitations to this study. The proteomic data set are limited to manually curated experiments and do not offer a scope of reanalysis of raw data. Moreover, the metastatic breast cancer dataset used is an ongoing prospective study, which implies that the addition of newer data and clinical samples can further add newer potential candidates if the same study is repeated again. However, the present limitations do not diminish the scope and findings of the present study.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
World Health Organization. Breast Cancer; 2020. Available from: https://www.who.int/news-room/fact-sheets/detail/breast-cancer. [Last accessed on 2021 Mar 26].  Back to cited text no. 1
    
2.
Kulka J, Szász AM, Németh Z, Madaras L, Schaff Z, Molnár IA, et al. Expression of tight junction protein claudin-4 in basal-like breast carcinomas. Pathol Oncol Res 2009;15:59-64.  Back to cited text no. 2
    
3.
Breast Cancer India. Breast Cancer Statistics: India versus the World; 2018. Available from: https://www.breastcancerindia.net/statistics/stat_global.html. [Last accessed on 2020 May 25].  Back to cited text no. 3
    
4.
World Health Organization. The Global Breast Cancer Initiative (GBCI). World Health Organization; 2021. Available from: https://www.who.int/publications/m/item/the-global-breast-cancer-initiative-gbci. [Last accessed on 2021 Oct 26].  Back to cited text no. 4
    
5.
Singh V, Mishra V. Identification of molecular signatures and pathways to identify novel therapeutic targets in mild cognitive impairment: Insights from a systems biomedicine perspective. Biomed Biotechnol Res J 2021;5:451-7.  Back to cited text no. 5
  [Full text]  
6.
Dwivedi V, Bharadwaj S, Mohanty P, Gupta U. Bioinformatics tools for genomic and evolutionary analysis of infectious agents. Biomed Biotechnol Res J 2018;2:163.  Back to cited text no. 6
  [Full text]  
7.
Habibi B, Novin MG, Salehpour S, Novin MG, Yeganeh SM, Nazarian H. Expression analysis of genes and MicroRNAs involved in recurrent implantation failure: New noninvasive biomarkers of implantation. Biomed Biotechnol Res J 2022;6:145-55.  Back to cited text no. 7
  [Full text]  
8.
Pande A. Co-regulatory network of transcription factor and microrna: A key player of gene regulation. Biomed Biotechnol Res J 2021;5:374-9.  Back to cited text no. 8
  [Full text]  
9.
Yadav R. Gene expression analysis to network construction for the identification of hub genes involved in neurodevelopment. Biomed Biotechnol Res J 2021;5:425-34.  Back to cited text no. 9
  [Full text]  
10.
Tomczak K, Czerwińska P, Wiznerowicz M. The Cancer Genome Atlas (TCGA): An immeasurable source of knowledge. Contemp Oncol (Pozn) 2015;19:A68-77.  Back to cited text no. 10
    
11.
Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401-4. Erratum in: Cancer Discov. 2012;2:960.  Back to cited text no. 11
    
12.
Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 2013;6:pl1.  Back to cited text no. 12
    
13.
Wagle N. Abstract 5371: The Metastatic Breast Cancer Project: Partnering with patients to accelerate progress in cancer research. Cancer Res 2018;78:5371.  Back to cited text no. 13
    
14.
Wagle N. Abstract OT-18-01: The metastatic breast cancer project: Generating the clinical and genomic landscape of metastatic breast cancer through patient-partnered research. In: San Antonio Breast Cancer Virtual Symposium; 2020. p. 1538-7445.  Back to cited text no. 14
    
15.
Pichler M, Hutterer GC, Stojakovic T, Mannweiler S, Pummer K, Zigeuner R. High plasma fibrinogen level represents an independent negative prognostic factor regarding cancer-specific, metastasis-free, as well as overall survival in a European cohort of non-metastatic renal cell carcinoma patients. Br J Cancer 2013;109:1123-9.  Back to cited text no. 15
    
16.
Pedrazzani C, Mantovani G, Salvagno GL, Baldiotti E, Ruzzenente A, Iacono C, et al. Elevated fibrinogen plasma level is not an independent predictor of poor prognosis in a large cohort of Western patients undergoing surgery for colorectal cancer. World J Gastroenterol 2016;22:9994-10001.  Back to cited text no. 16
    
17.
Repetto O, De Re V. Coagulation and fibrinolysis in gastric cancer. Ann N Y Acad Sci 2017;1404:27-48.  Back to cited text no. 17
    
18.
Perisanidis C, Psyrri A, Cohen EE, Engelmann J, Heinze G, Perisanidis B, et al. Prognostic role of pretreatment plasma fibrinogen in patients with solid tumors: A systematic review and meta-analysis. Cancer Treat Rev 2015;41:960-70.  Back to cited text no. 18
    
19.
Liu YL, Yan ZX, Xia Y, Xie XY, Zhou K, Xu LL, et al. Ligustrazine reverts anthracycline chemotherapy resistance of human breast cancer by inhibiting JAK2/STAT3 signaling and decreasing fibrinogen gamma chain (FGG) expression. Am J Cancer Res 2020;10:939-52.  Back to cited text no. 19
    
20.
Yang Q, Zhang Y, Cui H, Chen L, Zhao Y, Lin Y, et al. dbDEPC 3.0: The database of differentially expressed proteins in human cancer with multi-level annotation and drug indication. Database (Oxford) 2018;2018:bay015.  Back to cited text no. 20
    
21.
Oliveros J. An Interactive Tool for Comparing Lists with Venn's Diagrams. Available from: https://bioinfogp.cnb.csic.es/tools/venny/index.html.  Back to cited text no. 21
    
22.
Tang Z, Kang B, Li C, Chen T, Zhang Z. GEPIA2: An enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res 2019;47:W556-60.  Back to cited text no. 22
    
23.
Warde-Farley D, Donaldson SL, Comes O, Zuberi K, Badrawi R, Chao P, et al. The GeneMANIA prediction server: Biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res 2010;38:W214-20.  Back to cited text no. 23
    
24.
Fabregat A, Sidiropoulos K, Viteri G, Forner O, Marin-Garcia P, Arnau V, et al. Reactome pathway analysis: A high-performance in-memory approach. BMC Bioinformatics 2017;18:142.  Back to cited text no. 24
    
25.
Safran M, Dalah I, Alexander J, Rosen N, Iny Stein T, Shmoish M, et al. GeneCards Version 3: The human gene integrator. Database (Oxford) 2010;2010:baq020.  Back to cited text no. 25
    
26.
Safran M, Rosen N, Twik M, Barshir R, Stein TI, Dahary D, et al. The GeneCards Suite Chapter, Practical Guide to Life Science Databases (2022). pp 27-56.  Back to cited text no. 26
    
27.
Mitsopoulos C, Di Micco P, Fernandez EV, Dolciami D, Holt E, Mica IL, et al. canSAR: Update to the cancer translational research and drug discovery knowledgebase. Nucleic Acids Res 2021;49:D1074-82.  Back to cited text no. 27
    
28.
American Cancer Society. Breast cancer facts & Figures 2017-2018. In: Breast Cancer Facts & Figures. PA: American Cancer Society; 2017. p. 1-44. Available from: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/breast-cancer-facts-and-figures/breast-cancer-facts-and-figures-2017-2018.pdf%0Ahttp://www.breastcancer.org/risk/factors/woman. [Last accessed on 2022 Feb 11].  Back to cited text no. 28
    
29.
Ghezzi F, Cromi A, Siesto G, Giudici S, Serati M, Formenti G, et al. Prognostic significance of preoperative plasma fibrinogen in endometrial cancer. Gynecol Oncol 2010;119:309-13.  Back to cited text no. 29
    
30.
Zhang X, Wang F, Huang Y, Ke K, Zhao B, Chen L, et al. FGG promotes migration and invasion in hepatocellular carcinoma cells through activating epithelial to mesenchymal transition. Cancer Manag Res 2019;11:1653-65.  Back to cited text no. 30
    
31.
Klein T, Bischoff R. Physiology and pathophysiology of matrix metalloproteases. Amino Acids 2011;41:271-90.  Back to cited text no. 31
    
32.
Hiller O, Lichte A, Oberpichler A, Kocourek A, Tschesche H. Matrix metalloproteinases collagenase-2, macrophage elastase, collagenase-3, and membrane type 1-matrix metalloproteinase impair clotting by degradation of fibrinogen and factor XII. J Biol Chem 2000;275:33008-13.  Back to cited text no. 32
    


    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
References
Article Figures

 Article Access Statistics
    Viewed436    
    Printed44    
    Emailed0    
    PDF Downloaded58    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]