Background
Metastatic brain tumors are the most common type of brain tumor in adults and are associated with a poor survival of patients (median survival time = 3–6 months) [
1]. A total of 40–50% of brain metastases originate from lung cancer [
2]. Studies of differential gene expression between brain metastases and primary lung cancer have suggested that many genes may be involved in the brain metastasis of lung cancer. Using a cDNA microarray approach, more than 200 genes, including genes encoding plasma membrane proteins, antigen proteins, and cytoskeletal proteins, have been found to be differentially expressed between a metastatic brain tumor and a lung adenocarcinoma [
3]. These genes function in cell interaction, attachment, and motility.
Actinin, alpha 4 (ACTN4), a nonmuscle cytoskeleton protein, has been frequently reported to be associated with cell motility and cancer metastasis. Honda
et al. have suggested that cytoplasmic ACTN4 increases cell motility and is associated with a high metastatic potential and a poor prognosis of cancer based on their studies on multiple cancer cell lines, including lung cancer cell lines, and 61 patients with early-stage breast cancer [
4]. Since then,
ACTN4 has been reported to be associated with the progression and metastasis of many types of cancer, including breast [
5], colorectal [
6], pancreatic [
7], lung [
8-
10], brain [
11], bladder [
12,
13], and ovarian cancers [
14-
16] and salivary gland carcinoma [
17]. In addition,
ACTN4 was found to be highly expressed in a poor survival group of patients with non-small cell lung cancer, suggesting that
ACTN4 is a significant prognostic predictor in this cohort of patients [
8]. A novel alternative splice variant RNA of
ACTN4 has been suggested to be a candidate diagnostic marker of human small cell lung cancer [
9] and a prognostic factor for patients with high-grade neuroendocrine pulmonary tumors [
10]. However, in contrast to the aforementioned function of
ACTN4 as a positive regulator of tumorigenesis or cancer metastasis, several studies have suggested that
ACTN4 may function like a tumor suppressor to suppress malignant behaviors of cancer cells [
18,
19]. Therefore, the association of
ACTN4 expression with tumorigenicity and cancer metastasis needs to be further investigated in the clinic.
In order to define the critical signaling pathways and genes that contribute to the brain metastasis of lung cancer, we used the RNA-Seq approach to investigate the expression profiles of three types of tissues (primary lung cancer, adjacent benign lung tissue, and metastatic brain tissue) from one patient. Subsequently, a series of bioinformatic analyses were performed with the RNA-Seq data to identify differentially expressed genes among these three types of tissues and to discover the critical pathways and genes responsible for the brain metastasis of lung cancer.
Methods
Study subject
A 47-year-old female patient was found to have a space-occupying lesion in her lung in January 2009. A diagnosis of poorly/moderately differentiated adenocarcinoma with a primary tumor–lymph node–distant metastasis stage of T2N1Mx in the right middle lobe of the lung was made in January 2010. Resection of the lower part of the right middle lobe was conducted and followed by chemotherapy. The metastatic tumor at the right frontal lobe of the brain was found and resected in December 2010. Pathological analyses showed that the tumor was a poorly/moderately differentiated adenocarcinoma that had metastasized from the lung. The adjacent benign lung tissue (N16), the original lung cancer (T16), and the metastatic brain tumor (T30) were collected. This study was approved by the institutional review board of China-Japan Union Hospital of Jilin University and conducted in accordance with the ethical guidelines of the Declaration of Helsinki. The patient had signed a consent form before the study. In addition, we collected lung cancer and para-tumor tissues from ten patients for the confirmation study.
RNA isolation and RNA-seq library preparation
Total RNA was isolated from the tissues using a Trizol reagent (Invitrogen, Carlsbad, CA, USA). The RNA quality was assessed using a Bioanalyzer 2200 (Agilent, Santa Clara, CA, USA), and the samples were stored at −80°C until use. The RNA integrity numbers (RIN) of these RNA samples were more than 8.0 and were appropriate for cDNA library construction.
cDNA library construction and sequencing
The TruSeqTM RNA Sample Preparation Kit (Illumina, Inc.) was used to construct the cDNA libraries for these RNA samples, according to the manufacturer’s instructions. Briefly, oligo(dT) magnetic beads were applied to purify mRNA using 10 μg of total RNA, and the purified mRNA was subsequently fragmented into sizes of 200–500 bp using divalent cations at 94°C for 5 min. Reverse transcription (RT) of the first-strand cDNA from the RNA fragments was performed using SuperScript II reverse transcriptase and random primers. The second strand cDNA synthesis was performed using DNA polymerase I and RNase H. The synthesized cDNA fragments were then end-repaired by adding a single “A” base ligated with indexed adapters. These end-repaired cDNA fragments were purified and enriched by the polymerase chain reaction (PCR). The final cDNA libraries were generated by size selection through 2% agarose gel electrophoresis and quantified by a Bioanalyzer 2200. The tagged cDNA libraries were pooled in an equal ratio and loaded in a single lane of the Illumina HiSeqTM 2000 for paired-end sequencing.
qRT-PCR
The endogenous control β-actin was used as a control for RT-PCR amplification measurement of ACTN4 expression. RT-PCR primers were β-actin (5′-CTGGAACGGTGAAGGTGACA-3′ and 5′-AAGGGACTTCCTGTAACAATGCA-3′) and ACTN4 (5′-ACAAGCCCAACCTGGAC-3′ and 5′-GGTGCGGGCAATGGTG-3′). The cDNA was generated using a High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA) and oligo(dT) primers, according to the manufacturer’s instructions. qPCR amplification was performed with the following conditions: 2 min at 50°C, 10 min at 95°C, and 50 cycles of 15 s at 95°C, and 1 min at 60°C. The conditions for the melting curve analysis were 1 min at 90°C, 30 s at 55°C, and 30 s at 95°C.
The DEGseq algorithm was applied to filter the differentially expressed genes with a fold change > 2, P < 0.5, and false discovery rate (FDR) < 0.05 [
20]. Gene ontology (GO) analysis was performed according to the GO annotations from NCBI (
http://www.ncbi.nlm.nih.gov/), UniProt (
http://www.uniprot.org/), and Gene Ontology (
http://www.geneontology.org/). The pathway analyses were performed to determine the significant pathways associated with the differentially expressed genes according to the KEGG database. Fisher’s exact test, P values, and FDRs were applied in the GO and pathway analyses, according to a previous study [
21]. Series cluster analysis was performed to classify the differentially expressed genes in eight clusters based on the reads per kb per million reads (RPKM) change tendency of genes in these three types of tissues (N16, T16, and T30), according to a previous study [
22]. For example, the genes with the following expression pattern were classified into Cluster 1: expression in N16 > expression in T16 = expression in T30. The Path-Act-Network analyses were performed to reveal the interactive network among the pathways with enriched differentially expressed genes based on the KEGG database, including the metabolism, membrane transport, signal transduction, and cell cycle pathways [
23]. Cytoscape software was used to generate the graphical representations of the pathways [
24]. Gene-Act-Net analyses were conducted to reveal the network of the differentially expressed genes based on the interactions among the genes, proteins, and compounds included in the KEGG database.
Discussion
The expression profiles of primary lung cancer, adjacent benign lung tissues, and brain metastatic tumor tissues from a single patient were explored using the RNA-Seq technique. A series of bioinformatic analyses revealed gene functions and signaling pathways essential for lung cancer development and brain metastasis. Among these significant genes and pathways, ACTN4, encoding a nonmuscle actin cytoskeleton protein, and the pathway involved in regulation of actin cytoskeleton appeared to play a pivotal role in the process of lung cancer metastasis to the brain.
Overall, the quality of the RNA-Seq data and read mapping in our current study met the requirements for the bioinformatic analyses. However, the total reads and the mapping rate for the adjacent benign tissue were not as good as those of the primary lung cancer and the metastatic brain tumor tissues, probably resulting from the limited amount of the benign tissue and the total RNA available for the RNA-Seq approach. Nevertheless, appropriate bioinformatic analyses were performed on the RNA-Seq data from this study.
Differential gene expression analysis based on the RPKM values with upper-quartile normalization revealed that many genes were differentially expressed among these three types of tissues. The RPKM values refer to the reads per kb per million reads, according to a previous study [
25]. These differentially expressed genes were classified into eight clusters according to their changed expression patterns in these three types of tissues. The genes in cluster 4 are likely candidate genes that are indispensable for lung cancer metastasis to the brain because these genes presented a significantly increased expression in the metastatic brain tumors but not in the primary lung cancer tissue or the benign tissue. The genes in cluster 4 involve a variety of cellular functions, including cytoskeleton-dependent intracellular transport, membrane-to-membrane docking, actin filament bundle assembly, cell-cell adhesion, and actin cytoskeleton organization. The pathway analysis showed similar results for the cluster 4 genes. These genes are mostly involved in the signaling pathways associated with extracellular molecular interaction, cellular adhesion, adherens junction, and cytoskeleton organization. Interestingly,
ACTN4, encoding the alpha-actinin-4 protein, is among cluster 4 genes. Alteration of ACTN4 expression was further validated in these three types of tissues. ACTN4 has been shown to play important roles in cytoskeleton organization, cell adhesion, and cell migration. It has been suggested that ACTN4 is indispensable for mononuclear phagocyte response both in inflammation and tumor invasion processes [
26]. A recent study also supported that ACTN4, particularly relying on its C-terminal tail, mediates the cytoskeleton to the adhesion site during cell migration [
27]. Our results demonstrated that these functions of ACTN4 contribute to the process of lung cancer metastasis to the brain.
Of note, another alpha-actinin gene,
ACTN1, also appeared in cluster 4 and presented a similar altered expression pattern as
ACTN4 in these three types of tissues. Indeed,
ACTN1 has been reported to be essential for cytoskeleton organization and cell motility in some types of cells [
28]. Foley and Young recently have shown that ACTN4 and ACTN1 form a heterodimer in many types of cells [
29]. However, it also has been shown that
ACTN1 and
ACTN4 contribute to distinct malignant properties of astrocytoma cells and that
ACTN4 may be more important for cell motility and cell adhesion in some cell lines [
11].
In contrast to the functions of the genes in cluster 4, the genes in cluster 1 presented distinct functions such as regulation of reactive oxygen species, response to interlukin-6, and regulation of the activation of JAK2 kinase activity. Accordingly, the pathway analysis results were also distinct for the genes in cluster 1 and cluster 4. These results support our hypothesis that the genes in cluster 1 likely include candidate genes critical for tumorigenicity and that the genes in cluster 4 likely include candidate genes indispensable for metastasis.
The Pathway-Act-Network analysis based on the RNA-Seq data from the three types of tissues suggested that the pathways associated with the regulation of actin cytoskeleton are the pivotal players during lung cancer metastasis to the brain. Our data indicated that alteration of these actin cytoskeleton pathways could contribute to lung cancer metastasis to the brain through interacting with several other pathways involved in cellular processes, such as focal adhesion, adherens junction, and ECM-receptor interaction. The important function of the cytoskeleton in cancer metastasis has been widely recognized [
30]. The results from a study of transendothelial migration of small cell lung cancer cells across human brain microvascular endothelial cells showed that the Rho/ROCK pathway contributes to actin cytoskeleton reorganization [
31]. Consistent with this report,
ras homolog family member C (
RHOC) also appeared among the genes in cluster 4, and its expression was increased in the brain metastatic tumor but not in the primary lung cancer tissue.
ACTN4 is a critical gene related to actin cytoskeleton regulation and has been reported by multiple studies to play an important role in cell adhesion, cell motility, and cancer metastasis [
4,
5,
12,
15,
17,
18]. It also has been suggested that
ACTN4 may interact with
Rho family members to regulate cell motility and cancer metastasis through regulating cytoskeleton organization and focal adhesion [
11,
27,
28,
30,
31]. In short, the bioinformatic analysis data revealed that the pathways involved with actin cytoskeleton regulation were pivotal pathways in the Pathway-Act-Network and that the
ACTN4 gene was one of the key players in the Gene-Act-Network. Our current data are consistent with many previous studies, including a microarray and immunostaining data on
ACTN4 and its association with pathways contributing to lung cancer metastasis [
8,
10]. Our study provided the first RNA-Seq data to support the essential function of the
ACTN4 gene and the relevant cytoskeleton organization pathways in the brain metastasis of lung carcinoma. However, the fact that only one patient’s samples were used is a major limitation of our current study. A future study with more samples will help to confirm and support our current findings.
Acknowledgments
This work was supported in part by grants from the National Natural Science Foundation of China (#81372696), China Postdoctoral Science Foundation (#2013 M541314), Jilin Provincial Science and Technology Department (#20090175 and #20100733), Scientific Research Foundation for Returned Overseas Chinese Scholars, State Education Ministry (#2009-36), Health and Family Planning Commission of Jilin Province (#2010Z068), Scientific Research Foundation for the Returned Overseas Chinese Scholars, Human Resources and Social Security Department of Jilin Province (#2012-2014), Postdoctoral Science Foundation of Jilin Province, and Human Resources and Social Security Department of Jilin Province (2012).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
GHL and YCH carried out the molecular genetic studies, participated in the sequence alignment, and drafted the manuscript. ZS carried out the immunoassays. XYL and JHW participated in the sequence alignment. YFG and LKS participated in the design of the study and performed the statistical analysis. JWS and YJ conceived of the study, participated in its design and coordination, and helped to draft the manuscript. All authors read and approved the final manuscript.