Blog Archives

Gene Expression Analysis of Cervical Cancer Progression

Gene Expression Analysis of Cervical Cancer Progression

Caitlin Cundiff, M Karthikeyan, Gayathri Suresh Kumar,

Parimala S Devi,Taylor Westrick

Introduction

What is cervical cancer ?

Cervical cancer occurs when abnormal cells on the cervix grow out of control.More than 12000 cases of cervical cancer are diagnosed every year in USA. If one of the high risk types of HPV lingers in the body,it can cause abnormal cells to develop in the cervix.

What causes cervical cancer?

Most cervical cancer is caused by a virus called human papillomavirus, or HPV. Not all types of HPV cause cervical cancer. Some of them cause genital warts, but other types may not cause any symptoms. There are about 100 types of Human Papillomavirus(HPV), and about 70 types or so cause cervical cancer(categorized as “high risk”types).

Cancer cells have the potential to spread from their original location to other parts of the body where they can grow into new tumours. This process is called metastasis. The tumours are also called metastasis (singular) or metastases (plural). Metastases are also called secondary tumours.

 Understanding the usual progression of cancer helps the doctor to predict its probable course, plan treatment and anticipate further care.

Like all cancers, cancer of the cervix is much more likely to be cured if it is detected early and treated immediately.

  • One of the key features of cervical cancer is its slow progression from normal cervical tissue, to precancerous (or dysplastic) changes in the tissue, to invasive cancer.
  • The slow progression through numerous precancerous changes is very important because it provides opportunities for prevention and early detection (through Pap test) and treatment.

  • These opportunities have caused a decline in the incidence of cervical cancer over the past decades in the United States. Still, over 12,000 new cases of cervical cancer occur each year in the U.S., and over 4,000 women die each year from the disease.

Invasive cancer means that the cancer affects the deeper tissues of the cervix and may have spread to other parts of the body. This spread is called metastasis. Cervical cancers don’t always spread, but those that do most often spread to the lungs, the liver, the bladder, the vagina, and/or the rectum.

What is squamous cell carcinoma?

Squamous cell carcinoma is a common form of skin cancer that develops in the thin, flat squamous cells that make up the outer layer of the skin. Squamous cell carcinoma is usually not life-threatening, though it can be aggressive in some cases. Untreated, squamous cell carcinoma can grow large or spread to other parts of your body, causing serious complications.

What is cervical squamous cell carcinoma?

According to the American Cancer Society, in 2010 more than 12,200 American women will be diagnosed with invasive cervical cancer. Squamous cell carcinoma is the most common type of cervical cancer. More than 4,200 women die from cervical cancer each year, but the death rate has been declining by about 4 percent a year. The main reason for this decline is the increased use of the Pap test, which can detect pre-cancerous cells or early-stage cancers when they can be treated most effectively. The five-year survival rate for patients with very early-stage cervical cancer is more than 95 percent.

What characterizes cervical squamous cell carcinoma?

Cervical cells may become pre-cancerous due to the effects of a persistent HPV infection. In a small number of cases, and over a long period of time, usually at least 10 years, pre-cancerous cells may become capable of invasion. There are two types of cervical carcinoma based on the microscopic appearance of the cancer; squamous cell carcinoma accounts for 80 to 90 percent of all cervical cancers, with adenocarcinoma making up 10 to 20 percent. Sometimes, cancers have characteristics of both types; these malignancies are called adenosquamous carcinomas or mixed carcinomas.Early-stage cervical squamous cell carcinoma may not present symptoms; later-stage cancers can cause abnormal vaginal bleeding, increased vaginal discharge, pelvic pain, or pain during sexual intercourse.

What is meant by the stage of the cancer?

Stage 0 is the pre-cancerous or in-situ stage. Stage 1 cervical squamous cell carcinomas are small and confined to the cervix, and stage 4 tumors have spread beyond the cervix. Stages 2 and 3 describe conditions in between these two extremes.

Question formulated – Are there biomarkers involved in the development of cervical cancer that can be used for cancer therapy?

 

Three objectives were set:

1.Analyze differential expression of genes  involved in HPV progression.

2.Analyze differential expression of genes in cervical cancer tumorigenesis

3.Identify common  biomarkers with a potential cause-effect relationship with cervical cancer that can be used for gene targeted therapy .

Data sets used

Data set 1:

Pre-invasive and invasive cervical squamous cell carcinomas

Involved three types of cervical cells of Homo sapiens :

a)Normal

b)High-Grade Squamous Intraepithelial Lesion

c)Invasive Squamous Cell Carcinoma

http://www.genticel.com/web/en/35-hpv-and-cervix-carcinoma.php

Data set 2

Cervical cancer tumorigenesis

—Involved three types of cervical cells of Homo sapiens :

a) Cervical Cancer Cell Line

b) Normal Cervical Tissue

c) Cervical Cancer Primary Tumor

Methodology

Two data sets were retrieved from NCBI GEO database, record #GDS3292 and #GDS3233. The first represents the preinvasive and invasive squamous cell carcinoma while the second cervical cancer tumorigenesis. The data values  were log  2  normalized. and analyzed with JMP Genomics.

 For each data set, three comparisons were done across all cell types. Up regulated and down regulated genes were determined through differential gene expression using volcano plots for each comparison. Further analysis was done with clustering through PCA and heat maps. Pathway analysis was then carried out with KEGG and functional annotation through GORILLA and Revigo. Conclusions were drawn according to results.

Figure 1- Summary of Methods: Presentation of basic methods followed in the project.

Determining the differential genes of both data set 1 and data set 2:

Data Set 1

Pre invasive and invasive cervical squamous cell carcinoma

The samples were analyzed using the JMP Genomics Basic Expression Workflow module. The following analysis was carried out –   variance check, a heat map, and a principle component analysis. In addition  a volcano plot was generated to determine the down regulated and up regulated genes . Each image below  contains  a figure legend that describes the general information and  observations we observed in the  graph.

Figure A.1- Variance Check for Data Set 1 Code represents the elements that contribute to a differential expression due to the difference in type of sample. Array corresponds to the  differences due to the microarray chip; this represents a background and should  be low. Residual indicates differences due to unknown biological elements.

 Figure A.2- Heat Map of Data Set 1:

Blue represents a normal cell of the set , red a high grade squamous intraepithelial lesion, and green an invasive squamous cell carcinoma.  These colors correlate with the samples to the left of the heat map. Within the heat map, blue represents a  negative correlation and red a positive correlation. The  uppermost left box  shows a positive correlation with itself (normal cell – normal cell) where the uppermost right box shows a negative correlation between high grade squamous intraepithelial lesion- invasive squamous cell carcinoma.

Figure A.3- PCA of samples in data set 1:

Blue represents a normal cell of the set , red a high grade squamous intraepithelial lesion, and green an invasive squamous cell carcinoma. Each type of sample clusters with its like cell samples. This represents comparable samples across the data set in relation to overall gene expression for all genes of like type.

Figure A.4- Volcano plot of data set 1:

Three treatments were generated comparing each cell type to the other. In all three plots , anything above the red line represents a significant gene. Anything to the left (below 0) signifies downregulation and anything to the right (>0) upregulation. Theses genes were identified and noted for further analysis.

Figure A.5- Differential Genes: Above are the genes identified to be upregulated and downregulated for each treatment.

Analysis of Differentially expressed  Genes

Each treatment comparison (ex normal cell – squamous cell carcinoma) was analyzed with two components: a principal component analysis and a heat map. Only one treatment is shown for each data set.

Data Set 1

Preinvasive and invasive cervical squamous cell carcinoma-normal

Figure A2.1- PCA of Invasive squamous cell carcinoma-normal cell(Upregulated/Downregulated): Red represents downregulated genes, blue insignificant genes, and green upregulated genes. A high clustering was observed for the downregulated genes with only a slight clustering of the upregulated genes.

Figure A2.2-Heat map of squamous cell carcinoma-normal cell

Blue represents insignificant genes, red downregulated genes, and green upregulated genes. These colors identify with the column to the left of the heat map. Within the heat map, blue represents a negative correlation and red a positive correlation. The first box shows a positive correlation between downregulated-down regulated, The second box shows a negative relationship between downregulated-upregulated.

Data Set 2         Cervical cancer tumorigenesis-normal

Fig. B.1. Variance Analysis for Data set 2:

The code represents the difference in the differential expression caused by the  difference in the samples taken, the array represents the variation caused by the microarray chip, and the residual represents the effect of the unknown biological elements.

Fig. B.2. Clustering of data samples across all genes for Data set 2:

The three kinds of samples are cervical cancer cell line( red), normal cell(blue),cervical cancer primary tumor(green).  In the heat map, the red and blue represent positive correlation and negative correlation respectively.  The boxes marked 1, 2, 3 and 5 indicate the positive correlation. The 1st box indicates positive correlation amongst itself (i.e. cervical cancer cell line- cervical cancer cell line). The box numbered 4 indicates a negative correlation between the normal cell and the cervical cancer primary tumor.

Fig. B.3. Principle Component Analysis of Data set 2:

The three kinds of samples are cervical cancer cell line( red), normal cell(blue),cervical cancer primary tumor(green).  Each sample type is clustered with like cell samples. Within each stage of cervical cancer tumorigenesis, the gene expression is similar across each cluster.

Fig. B.4. Volcano plots of Data set 2:

Volcano plots of each of the types of cells were plotted. The red line refers to the cut off for differential significance. The points above the line are significant genes. The points to the right of 0 are up-regulated and the points to the left of 0 are down-regulated genes.

Fig. B.5. List of Up-regulated and down-regulated genes in Data set 2

Fig. B.6. Principle Component Analysis of cervical cancer primary tumor- normal

Red represents the down-regulated genes and blue represents the up-regulated genes. A higher frequency of clustering is observed in the up-regulated genes than the down-regulated genes.

Fig. B.7. Heat map of Genes of cervical cancer primary tumor–normal:

The three kinds of samples are cervical cancer cell line( red), normal cell(blue),cervical cancer primary tumor(green).  In the heat map, red represents the down-regulated genes and blue represents the up-regulated genes. The boxes 1 and 2 show a negative correlation between the cervical cancer primary tumor cells and the normal cells.

GENE ONTOLOGY ANALYSIS

Genes can be clustered according to their biological process , molecular function and cellular component using Gene Ontology databases.
The following figures illustrates the biological functions the upregulated and the downregulated genes are enriched in.

Cervical Cancer Primary Tumour vs Normal

The Upregulated Genes

Figure GO:1- Gene Ontology Clusters of Go Process for the genes upregulated in the Cervical Cancer Primary Tumour vs Normal pair

From Figure GO:1  one can observe that majority of the upregulated genes are broadly involved in the complement activation pathways more specifically in immune processes.A fraction of them are also involved in regulating cellular response to acid , collagen fibril organization and multicellular organismal catabolism.

Figure GO:2- Gene Ontology Clusters of Go Function for the genes upregulated in the Cervical Cancer Primary Tumour vs Normal pair

The enrichment of the growth factor binding GO terms tends to justify the involvement of the tumour upregulated genes in tumorigenesis and uncontrolled cell growth . Few other genes pull up the antigen binding GO term and the cell structural integrity GO term.

Selection_001

Figure GO:3-Gene Ontology Component of Go Function for the genes up regulated in the Cervical Cancer Primary Tumour vs Normal pair

The upregulated genes are expressed mainly in the extracellular region .Significant expression is seen fibrillar collagens too !

The Downregulated  Genes

Figure GO:4-Gene Ontology Clusters of Go Process  for the genes downregulated in the Cervical Cancer Primary Tumour vs Normal pair

Tumorigenesis  seems to stall cell differentiation , wound healing and peptide cross linking

Figure  GO:5-Gene Ontology Clusters of Go Component for the genes downregulated in  the Cervical Cancer Primary Tumour vs Normal pair

The downregulation of these genes happens in the cornified envelope , a region of the stratum corneum .

Invasive Squamous Cell Carcinoma vs Normal

The Upregulated Genes

Figure GO:6-Gene Ontology Clusters of Go Process for the genes upregulated in the Invasive Squamous Cell Carcinoma vs Normal pair

The genes that are upregulated are primarily involved in inflammatory response.

Figure GO:7-Gene Ontology Clusters of Go Process for the genes upregulated in the Invasive

Squamous Cell Carcinoma vs Normal pair

The Upregulated Genes

Figure GO:8-Gene Ontology Clusters of Go Component for the genes upregulated in the Invasive Squamous Cell Carcinoma vs Normal pair

The Downregulated Genes

Figure GO:9-Gene Ontology Clusters of Go Process for the genes down regulated in the Invasive Squamous Cell Carcinoma vs Normal pair

Figure GO:10 -Gene Ontology Clusters of Go Function  for the genes down regulated in the Invasive Squamous Cell Carcinoma vs Normal pair

These functionally annotated genes were then routed for pathway analysis to study all the biological networks that these enriched genes affect .

Pathway Analysis

—Figure PA:1- In our gene expression analysis, protein C3 was up-regulated in squamous cells in high-grade squamous intraepithelial lesions and invasive squamous cell carcinoma. Complement Component C3 plays a key role in activation of the complement pathway of the adaptive immune system.

—Figure PA:2- In our gene expression analysis, iC3b, a protein by product of the complement pathway, was up-regulated in squamous cells, in high-grade squamous intraepithelial lesions, and invasive squamous cell carcinoma.

Conclusion

—In our analysis —potential cancer biomarkers were found, but further analysis is needed to find statistically significant causal genes. However,  our analysis identified statistically significant up regulation of proteins involved in the pathophysiological immune response due to HPV and cervical cancer progression. Pathway analysis showed Complement Component 3 protein upregulation of squamous cells in both high-grade squamous intraepithelial lesion and invasive squamous cell carcinomas. Also upregulated in the precancerous and cancerous stages of cervical cancer was the by product protein iC3b of the complement pathway. Since, more than 99% of cervical cancers are caused to HPV, we conclude that the upregulation of proteins and by products of the complement pathway are due to the viral carcinogenesis nature of cervical cancer. Some patients’ immune systems can clear HPV before it progresses to cervical cancer. However, most HPV if left untreated do progress to cervical cancer. Since there is a clear immune response to the virus both in the precancerous and cancerous state, it is clear the body has strong defense mechanisms against this disease. Therefore, future studies should consider the factors that are involved in activation of the complement pathway and the success of patients that can clear the disease before cancer progression.

References:

1.Zhai Y, Kuick R, Nan B, Ota I et al. Gene expression analysis of preinvasive and invasive cervical squamous cell carcinomas identifies HOXC10 as a key mediator of invasion. Cancer Res 2007 Nov 1;67(21):10163-72. PMID: 17974957

2.Scotto L, Narayan G, Nandula SV, Arias-Pulido H et al. Identification of copy number gain and over expressed genes on chromosome arm 20q by an integrative genomic approach in cervical cancer: potential role in progression. Genes Chromosomes Cancer 2008 Sep;47(9):755-65. PMID: 18506748

3.http://www.webmd.com/cancer/cervical-cancer/

4.http://www.cancer.ca/en/cancer-information/cancer-101/what-is-cancer/how-cancer-spreads/?region=on

5.http://www.genticel.com/web/en/35-hpv-and-cervix-carcinoma.php

6.http://www.emedicinehealth.com/script/main/art.asp?articlekey=5018

7.Eran Eden*, Roy Navon*, Israel Steinfeld, Doron Lipson and Zohar Yakhini. “GOrilla: A Tool For Discovery And Visualization

  of   Enriched GO Terms in Ranked Gene Lists”, BMC Bioinformatics 2009, 10:48.

8.Supek F, Bošnjak M, Škunca N, Šmuc T. “REVIGO summarizes and visualizes long lists of Gene Ontology terms” PLoS ONE 2011. doi:10.1371/journal.pone.0021800

9.Kanehisa, M., Goto, S., Sato, Y., Furumichi, M., and Tanabe, M.; KEGG for integration and interpretation of large-scale molecular datasets. Nucleic Acids Res. 40, D109-D114 (2012)

Advertisements