Emerging evidence implicates alterations in the RAB small GTPases and their associated regulatory proteins and effectors in multiple human diseases including cancer. We have recently shown that RAB25, located at chromosome 1q22, is amplified at the DNA level and overexpressed at the RNA level in ovarian and breast cancer. These changes correlated with a worsened outcome in both diseases. In addition, enforced expression of RAB25 in both breast and ovarian cancer cells decreased apoptosis and increased proliferation and aggressiveness in vivo, potentially explaining the worsened prognosis. A better understanding of genetic alterations as well as the physiologic and pathophysiologic roles of RAB GTPases may open new opportunities for therapeutic intervention and better outcomes.

Cancer is caused by abnormalities in DNA sequence, copy number, rearrangements, or expression. The accumulation of multiple changes in critical genes within a single cell is required to escape from normal controls on cell growth and proliferation, allowing development into a clinically evident tumor (1). Understanding the molecular basis of tumorigenesis is the key to better diagnosis, prognosis, and treatment for cancer. Large-scale profiling of gene expression and genomic alterations has revealed multiple differences between normal and malignant cells, specific genetic and cellular changes that occur at each stage of tumor progression, and aberrations that can distinguish cancers of different origins and metastatic potential. This genome-wide information will facilitate the identification of new diagnostic markers, prognostic and predictive information, and novel molecular targets. This, in turn, will lead to the development, validation, and implementation of new, less toxic, and more effective drugs.

Array comparative genomic hybridization (2) provides a robust, sensitive, and high-resolution approach to the identification of regions of DNA copy number increase and decrease in tumors. These copy number aberrations are selected during tumorigenesis, contributing to the behavior of tumors, and indeed can be predictive of patient outcome (3). Multiple chromosomal amplifications implicated in the pathophysiology of ovarian and breast carcinomas have been detected (4, 5). The identification of the candidate genes driving the development of the DNA copy number aberrations in cancer has progressed at a slow rate. However, new technologies are likely to increase the pace. We, for example, have extensively characterized an amplicon present in ovarian cancer on chromosome 3q, finding that the PIK3CA gene expressing the catalytic α subunit of phosphoinositide-3-kinase is amplified in ovarian cancer and is a critical regulator of different functions in ovarian cancer (6). This, along with other developments, has contributed to the phosphoinositide-3-kinase pathway being identified as a prime candidate for the development of novel therapies.

Array comparative genomic hybridization revealed a frequent increase in a 1.1 mb region at chromosome 1q22 in 54% and 47% of advanced serous epithelial ovarian cancers and breast cancers, respectively (7), suggesting the potential involvement of a gene within this region in breast and ovarian cancers. A total of 34 open reading frames, including 22 known genes and 12 hypothetical proteins, reside in this region. Using the powerful resource of publicly available gene expression data sets, generously made available by colleagues (8), we were able to eliminate 18 genes as candidates as they did not show frequent changes in gene expression. The remaining 16 candidates were studied by real-time quantitative PCR. Although mRNA levels of several of the genes in this region were modestly elevated in a fraction of ovarian cancers as compared with normal ovarian surface epithelial cells or benign ovarian tumors, RAB25 mRNA levels were significantly increased in 89% (P < 0.001) of the ovarian cancers analyzed. In an independent ovarian cancer data set (9), RAB25 levels were increased in 80% of ovarian cancer samples compared with normal ovarian epithelium. Similarly, RAB25 was overexpressed in 67% of the tumors from breast cancer patients when compared with normal breast tissue (10). Linear regression analysis showed a direct relationship between DNA copy number and RNA expression; however, the increase in RNA levels occurs with a greater frequency than the increase in DNA copy number, suggesting that additional mechanisms could lead to deregulation of RAB25 mRNA levels.

Ovarian cancer patients with elevated 1q22 either failed to enter a disease-free state following surgery and chemotherapy or exhibited short durations of disease-free survival, suggesting that a gene in this region regulates the behavior of ovarian cancer. Indeed, ovarian cancer patients with high RAB25 mRNA levels (>2-fold increase compared with normal controls) showed a decrease in disease-free interval and overall survival. Similarly, Kaplan-Meier survival analysis of the 109 patients with breast cancer, for whom follow-up data was available in the data set at the Stanford web site (http://genome-www.stanford.edu/breast_cancer/robustness/data.shtml), revealed a correlation between high RAB25 expression and a decrease in both overall survival (P < 0.02) and disease-free survival (P < 0.01). Together, these data suggest that RAB25 contributes to the aggressiveness of breast and ovarian cancers. Mutations in RAB25 were not detected in ovarian cancer; however, whether RAB25 is mutated in other tumors remains to be determined. Indeed, the PIK3CA gene, which is amplified in serous epithelial ovarian cancer (6), is frequently mutated in endometrioid ovarian cancers, brain, lung, and breast tumors (11, 12).

The role of RAB25 in the aggressiveness of ovarian and breast cancers was assessed by altering RAB25 levels in cell lines. Elevating the expression of RAB25 in ovarian and breast cancer cell lines increased anchorage-independent colony-forming activity, cell proliferation, and cell survival under multiple stress conditions including serum starvation, anchorage deprival, UV irradiation, and chemotherapy treatment (paclitaxel). Decreasing RAB25 expression by small interfering RNA decreased cell proliferation and markedly increased the sensitivity to apoptosis of both breast and ovarian cancer cells. In murine xenograft assays, RAB25 facilitated tumor formation and increased the growth rate of ovarian cancer cells. However, whereas RAB25 overexpression was sufficient to make cancer cells more aggressive, in contrast to RAS (13), RAB25 overexpression could not cooperate with SV40 large T antigen and hTERT to drive tumor formation by immortalized human cells. Thus, RAB25 seems to play a role in tumor progression and aggressiveness rather than tumor initiation, a concept compatible with the increases in RAB25 levels at later stages in ovarian cancer.

Analysis of the antiapoptotic effects of RAB25 revealed Bcl-2 and phosphoinositide-3-kinase pathways that are known to regulate cell survival (14, 15). We found that enforced RAB25 expression in ovarian cancer cells not only decreases levels of the BAX and BAK proteins, two key effectors of mammalian apoptotic signaling cascade (16), but that RAB25 also increases AKT phosphorylation, an indication that the phosphoinositide-3-kinase pathway is active. Down-regulating RAB25 reversed the decrease in BAX and BAK levels and the increase in AKT activity, supporting the potential involvement of the Bcl-2 and phosphoinositide-3-kinase pathways in mediating RAB25 action.

Information arising from global DNA, RNA, and protein analysis will not only aid in unraveling the initiation, development, and progression of genetic diseases such as cancer, but will also facilitate the identification of new biomarkers and molecular targets. This will rapidly open a new chapter in drug discovery, development, and patient care. Using this approach, we are able to identify the driver of an important chromosomal amplification at 1q22 in breast and ovarian cancer. The amplification at 1q22 had been previously shown to be present in breast cancers (17) but had not been noted in ovarian cancer. However, in breast cancer, the region of genomic aberration is large, encompassing much of 1q. In ovarian cancer, the regional amplification was quite narrow and thus, missed by chromosomal comparative genomic hybridization. The narrow region of amplification, as well as the resolution of array comparative genomic hybridization, facilitated the identification of RAB25 as a driver of the aberration at 1q22. Indeed, since the publication of our paper, we have done higher resolution array comparative genomic hybridization and have further delimited the region of aberration in ovarian cancer, demonstrating a narrow region of amplification at 1q22. The mechanisms contributing to the different structures of the 1q amplicon in breast and ovarian cancer warrants further investigation to determine whether this reflects different mechanisms leading to genomic instability or the selection of additional genes in breast cancer. In addition to breast and ovarian cancers, Wilms' tumors also show an amplification at 1q22 (18); however, whether this involves RAB25 remains to be determined. Nevertheless, increased copy number at chromosome 1q22 is associated with a high relapse rate in invasive ductal breast carcinomas (17) and Wilms' tumors (18). In addition, Kudoh et al. (19) have reported that a significantly higher ratio of genetic changes, characterized by increased copy number of chromosome 1q21-q22 and 13q12-q14, are present in cisplatin-resistant than in cisplatin-sensitive ovarian tumors. Together, these data suggest that a gene(s) located at 1q22 controls tumor progression, aggressiveness, and potentially, chemosensitivity. Our studies revealed RAB25 as a major player in the development of the 1q22 amplicon and in tumor behavior.

RAB GTPases play a master role in regulating intercellular vesicle trafficking in both exo- and endocytic pathways. Approximately 60 human RAB genes are encoded by the human genome. However, the complexity of the RAB protein family might be even greater as there is evidence that alternative splicing of RAB mRNAs results in the production of functionally distinct isoforms (20). Most RAB proteins are constitutively expressed in all mammalian cells, although several RAB proteins have been shown to be differentially expressed in epithelia and neurons, perhaps fulfilling specialized transport functions of these polarized cells. RAB protein function is regulated at multiple levels including protein expression, localization, activation, binding partners, and recruitment of specific downstream effectors. Recent studies have shown multiple links between RAB GTPase dysfunction and associated regulatory proteins in human diseases. For instance, Griscelli syndrome type 2, caused by mutation of the RAB27a gene (21), is a rare autosomal recessive disorder that results in pigmentary dilution of the skin and hair, the presence of large clumps of pigment in hair shafts, and an accumulation of melanosomes in melanocytes. Most Griscelli syndrome type 2 patients also develop an uncontrolled T lymphocyte and macrophage activation syndrome, known as hemophagocytic syndrome, leading to death in the absence of bone marrow transplantation. Altered function of multiple RAB regulatory proteins, such as RAB escort protein, RAB geranylgeranyl transferase, and RAB GDP dissociated inhibitor, causes choroideremia (retinal degeneration), Hermansky-Pudlak syndrome (a type of albinism which includes a bleeding tendency and lung disease), and X-linked nonspecific mental retardation, respectively (20).

Dysregulation of RAB gene expression may be a generalized component of human tumors. In addition to RAB25, many RAB proteins exhibit elevated expression at the RNA level. RNA microarray analyses in ovarian cancer show that ∼45% to 50% of known RAB or RAB-associated genes display increases in mRNA expression. Moreover, these increases are frequent, occurring in over two-thirds of tumor samples. Up-regulation of RAB5a and RAB7 occurs in thyroid-associated adenomas (22). We have shown an increase in RAB25 gene expression in ovarian and breast cancers. Increased RAB25 levels have also been noted in prostate cancer (23), transitional cell carcinoma of the bladder (24), and invasive breast tumor cells (25), suggesting the pathologic role of RAB proteins in the development or progression of tumors in multiple epithelial lineages.

It is likely that additional evidence will accumulate implicating RAB dysfunction in cancer and other diseases, as RAB proteins have crucial roles in vesicle trafficking, signal transduction, and receptor recycling which in turn regulate normal cellular activity (Fig. 1). For example, a recent study has provided direct evidence that multiple signaling molecules including AKT, extracellular signal-regulated kinase 1/2, and p38 mitogen-activated protein kinase associate on endocytic vesicles and mediate their function by facilitating translocation of the vesicles to the nucleus, implicating RAB GTPases in conducting cell survival signals (26). In addition, RAB7 has been shown to mediate endocytosis and degradation of glucose and amino acid transporters in growth factor–deprived conditions, which in turn control cellular nutrient uptake and sensitivity to cell death (27). Based on the prevalence of altered RAB expression and function in many diseases, it is of interest to consider the therapeutic potential of controlling RAB function or RAB-regulated pathways through pharmacologic or genetic modulation.

In conclusion, the RAB GTPase family may be involved in a variety of disease settings including cancer progression. As the mechanisms of their actions are identified, RAB GTPases and the signaling and trafficking pathways in which they operate may present novel targets for therapeutic invention.

Grant support: Odyssey Program of the Houston Endowment Scientific Achievement award from M.D. Anderson Cancer Center (K.W. Cheng). NIH Specialized Programs of Research Excellence grants P50-CA83639, PPG-PO1 CA64602 (G.B. Mills and J.W. Gray), P50-CA58207 (J.W. Gray) and P30 grant CA16672-28. U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research [Contract DE-AC03 765F00098 (J.W. Gray)].

1
Hanahan D, Weinberg RA. The hallmarks of cancer.
Cell
2002
;
100
:
57
–70.
2
Pinkel D, Segraves R, Sudar D, et al. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays.
Nat Genet
1998
;
20
:
207
–11.
3
Saretzki G, Hoffmann U, Rohlke P, et al. Identification of allelic losses in benign, borderline, and invasive epithelial ovarian tumors and correlation with clinical outcome.
Cancer
1997
;
80
:
1241
–9.
4
Gray JW, Suzuki S, Kuo WL, et al. Specific keynote: genome copy number abnormalities in ovarian cancer.
Gynecol Oncol
2003
;
88
:
S16
–21.
5
Albertson DG. Profiling breast cancer by array CGH.
Breast Cancer Res Treat
2003
;
78
:
289
–98.
6
Shayesteh L, Lu Y, Kuo WL, et al. PI3KCA is implicated as an oncogene in ovarian cancer.
Nat Genet
1999
;
21
:
99
–102.
7
Cheng KW, Lahad JP, Kuo WL, et al. The RAB25 small GTPase determines aggressiveness of ovarian and breast cancers.
Nat Med
2004
;
10
:
1251
–6.
8
Perou CM. Show me the data!
Nat Genet
2001
;
29
:
373
.
9
Schaner ME, Ross DT, Ciaravino G, et al. Gene expression patterns in ovarian carcinomas.
Mol Biol Cell
2003
;
14
:
4376
–86.
10
Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets.
Proc Natl Acad Sci U S A
2003
;
100
:
8418
–23.
11
Campbell IG, Russell SE, Choong DY, et al. Mutation of the PIK3CA gene in ovarian and breast cancer.
Cancer Res
2004
;
64
:
7678
–81.
12
Samuels Y, Velculescu VE. Oncogenic mutations of PIK3CA in human cancers.
Cell Cycle
2004
;
3
:
1221
–4.
13
Liu J, Yang G, Thompson-Lanza JA, et al. A genetically defined model for human ovarian cancer.
Cancer Res
2004
;
64
:
1655
–63.
14
Gross A, McDonnell JM, Korsmeyer SJ. BCL-2 family members and the mitochondria in apoptosis.
Genes Dev
1999
;
13
:
1899
–1911.
15
Mills GB, Lu Y, Fang X, et al. Role of abnormalities of PTEN and the phosphatidylinositol 3′ kinase pathway in breast and ovarian tumorigenesis, prognosis and therapy.
Semin Oncol
2001
;
28
:
125
–41.
16
Wei MC, Zong WX, Cheng EH, et al. Proapoptotic BAX and BAK: A requisite gateway to mitochondrial dysfunction and death.
Science
2001
;
292
:
727
–30.
17
Zudaire I, Odero MD, Caballero C, et al. Genomic imbalances detected by comparative genomic hybridization are prognostic markers in invasive ductal breast carcinomas.
Histopathology
2002
;
40
:
547
–55.
18
Lu YJ, Hing S, Williams R, Pinkerton R, Shipley J, Pritchard-Jones K. UK Children's Cancer Study Group Wilms' tumor group. Chromosome 1q expression profiling and relapse in Wilms' tumour.
Lancet
2002
;
9330
:
385
–6.
19
Kudoh K, Takano M, Koshikawa T, et al. Gains of 1q21-q22 and 13q12-q14 are potential indicators for resistance to cisplatin-based chemotherapy in ovarian cancer patients.
Clin Cancer Res
1999
;
5
:
2526
–31.
20
Stein MP, Dong J, Wandinger-Ness A. Rab proteins and endocytic trafficking: potential targets for therapeutic intervention.
Adv Drug Deliv Rev
2003
;
14;55
:
1421
–37.
21
Menasche G, Pastural E, Feldmann J, et al. Mutations in RAB27A cause Griscelli syndrome associated with haemophagocytic syndrome.
Nat Genet
2000
;
25
:
173
–6.
22
Calvo A, Xiao N, Kang J, et al. Alterations in gene expression profiles during prostate cancer progression: functional correlations to tumorigenicity and down-regulation of selenoprotein-P in mouse and human tumors.
Cancer Res
2002
;
62
:
5325
–35.
23
Croizet-Berger K, Daumerie C, Couvreur M, Courtoy PJ, van den Hove MF. The endocytic catalysts, Rab5a and Rab7, are tandem regulators of thyroid hormone production.
Proc Natl Acad Sci U S A
2002
;
99
:
8277
–82.
24
Mor O, Nativ O, Stein A, et al. Molecular analysis of transitional cell carcinoma using cDNA microarray.
Oncogene
2003
;
22
:
7702
–10.
25
Wang W, Wyckoff JB, Frohlich VC, et al. Single cell behavior in metastatic primary mammary tumors correlated with gene expression patterns revealed by molecular profiling.
Cancer Res
2002
;
62
:
6278
–88.
26
Delcroix JD, Valletta JS, Wu C, Hunt SJ, Kowal AS, Mobley WC. NGF signaling in sensory neurons: evidence that early endosomes carry NGF retrograde signals.
Neuron
2003
;
39
:
69
–84.
27
Edinger AL, Cinalli RM, Thompson CB. Rab7 prevents growth factor-independent survival by inhibiting cell-autonomous nutrient transporter expression.
Dev Cell
2003 Oct
;
5
:
571
–82.