A key characteristic shared by all living cells is their ability to communicate with other cells and with their environment. This process, accomplished by intricate cell-cell and cell-signaling interaction networks, is critical during development, homeostasis, and disease pathogenesis. In recent years, learning about cell signaling has provided a better comprehension of cancer biology. At the same time, details about signaling in malignant cells have been critical for understanding cellular physiology.
“One of our interests is to understand the detailed molecular mechanisms related to how dysregulation of the Akt signaling pathway contributes to different types of cancers,” says Pengda Liu, Ph.D., assistant professor of biochemistry and biophysics at the University of North Carolina–Chapel Hill. A major focus in Dr. Liu’s laboratory is the identification of therapeutic targets that could modulate signaling in the PI3K-Akt pathway.
This pathway is hyperactivated in virtually all the solid tumors that have been described. “But the mechanisms of activation are different from tumor to tumor,” notes Dr. Liu.
The PI3K-Akt pathway integrates extracellular signals to regulate critical cellular processes. Dysregulation in this pathway has been linked to changes in oncogenes and tumor-suppressor genes. Studies in recent years have also revealed that this pathway is involved in regulating the epigenome.
Some of the challenges in identifying compounds that modulate the PI3K-Akt pathway are that its regulation is complex, multiple regulatory mechanisms are involved, and many of them cross-talk with other signaling pathways. Another challenge stems from the fact that even though very tumor-specific Akt activators have been well-characterized as potential drug targets, specific inhibitors and possible combined treatment options have yet to be developed.
“We would like to determine whether we could find cancer-specific upstream regulators of Akt that are not present or functional in normal cells,” informs Dr. Liu.
In Dr. Liu’s laboratory, investigators are using multidisciplinary experimental approaches to explore signaling events mediated by protein modifications and protein-protein interactions that are dysregulated under pathological conditions in the PI3K-Akt pathway. The investigators hope to identify enzymatic inhibitors and discover antibodies that could be used as therapeutic tools.
“By better understanding the interplay between key cancer signaling events, we would like to connect cancer signaling, metabolism, and epigenetics,” states Dr. Liu. “Our ultimate goal is to identify and develop new targeted therapies to combat cancer.”
Cancer research and therapeutic development is increasingly focused on tumor heterogeneity, or cell-to-cell variability in genetic, epigenetic, morphological, and phenotypic profiles. Historically, one of the main sources of cell-to-cell variability has been protein expression variation.
“Protein expression variation certainly is part of heterogeneity, but our data argue that molecular processes themselves also contribute to heterogeneity,” says Jay. T. Groves, Ph.D., professor of chemistry at the University of California, Berkeley. “These processes are inherently stochastic.”
The take-home message from his team’s work, Dr. Groves indicates, is that certain molecular processes fluctuate between long-lived functional states. These fluctuations, he adds, likely “contribute to cell-to-cell heterogeneous behaviors.”
In a proof-of-concept study, Dr. Groves and colleagues examined the activation of H-Ras, a small guanosine triphosphatase (GTPase), by Son of Sevenless (SOS), a guanine nucleotide exchange factor (GEF).1 Dr. Groves’ team used a single-molecule enzymatic assay to assess the activation of membrane-bound GTPases by GEFs. This approach revealed that SOS undergoes stochastic fluctuations between distinct, long-lived functional states. These states may last more than 100 seconds, reaching time scales characteristic of receptor signaling processes.
Historically, studies have examined cellular populations, and parameters have been expressed in the form of mean values, “averaging out” fluctuations between different activity states. Hidden within ensemble averages, these fluctuations eluded capture.
“We use our membrane microarray technology to observe H-Ras activation by SOS molecules, assessing large numbers of molecules while still tracking each molecule individually,” details Dr. Groves. “In this way, it becomes experimentally practical to map out the whole distribution. We don’t have to rely on average measurements—that will be very important.”
In studies of live cells, certain kinds of variability in protein characteristics tend to remain obscure. For example, proteins in different subcellular locations may have different post-translational modifications. Also, cellular distributions of post-translational modifications may be heterogeneous. Both these sorts of variability may be lost during purification. “One of the technical challenges is how to study molecules that are unpurified, in their native state,” advises Dr. Groves.
Other proteins have been relatively resistant to structure-function analyses because of challenges in reconstituting them in vitro. SOS is one such protein. It has a disordered C-terminal proline-rich domain that makes purification and X-ray crystallography difficult.
Several studies have linked truncations in the proline-rich domain of SOS to human cancers.2,3,4 To interrogate the structure-function relationship and dynamics of full-length native SOS, Dr. Groves and colleagues developed a membrane microarray assay that uses single SOS molecules from raw cell lysates.
“In our study, the cell lysates still originate from many cells,” notes Dr. Groves. “One of the challenges will be to do an experiment on the cytoplasm from a single cell, so that we get to see what one cell is doing rather than looking at averages.”
Despite relying on averages, the study revealed that the proline-rich domain suppresses SOS by reducing the kinetic rate of activation. The process occurred only on the membrane surface, in an activity that was independent of the N-terminal domain of the molecule. “As technologies get better,” predicts Dr. Groves, “we will see more of these large-scale single-molecule studies.”
A new approach to fighting prostate cancer combines the actions of steric-blocking splice-switching oligonucleotides (SSOs) and a novel nanotechnology-based approach for targeted delivery of DNA to tumor cells. “In our nanotechnology-based approach, we use specific SSOs that change their conformation under acidic conditions,” says Alexander V. Kazansky, Ph.D., associate professor in health and biomedical science at The University of Texas, Rio Grande Valley.
At the Annual Summit on Cell Therapy and Molecular Medicine, a conference held in September 2017 in Chicago, Dr. Kazansky described how alternative splicing could be used to inhibit the expression of STAT5B, a proto-oncogene involved in prostate cancer progression, and simultaneously increase the expression of its naturally truncated isoform, STAT5?B, which acts as a tumor suppressor gene. The truncated STAT5?B isoform results from the insertion of an alternatively spliced exon that introduces a premature stop codon in STAT5B.
STAT proteins are cytoplasmic transcription factors critical for cellular processes, and mammalian cells contain at least seven types of STAT protein. One of the challenges in targeting STAT proteins is that existing approaches indiscriminately suppress both their proto-oncogene and tumor suppressor gene functions.
The approach used by Dr. Kazansky and colleagues is based on the finding that under acidic conditions, such as the ones encountered in cancer cells, a water-soluble peptide forms an alpha-helix across membrane bilayers.
“We generated a conjugate of our switching oligonucleotide, which was connected to the C-terminus of a peptide molecule,” details Dr. Kazansky. “This allows the peptide to work as a nanosyringe.”
Once the conjugate is inside the cell, it encounters a reducing environment, which breaks the disulfide bond that conjugates the insertion peptide to the switching oligonucleotide. Breaking this bond releases the oligonucleotide and allows it to perform its action.
“In mice that carried xenografts, we deployed oligonucleotide conjugates that were labeled with a near-infrared dye,” states Dr. Kazansky, “and we demonstrated that the conjugates were delivered to the cancer.”
“In our studies, which combined transcriptome and whole-genome sequence analyses, we validated several fusion genes that occur in prostate cancer,” says Jianhua Luo, M.D., Ph.D., professor of pathology and UPMC Endowed Chair of Molecular Carcinogenesis at the University of Pittsburgh Medical Center. This work with fusion genes, Dr. Luo suggests, could illuminate the molecular and genomic changes associated with aggressive behavior in prostate cancer.
Previously, Dr. Luo and colleagues compared whole-genome DNA methylation patterns in prostate cancer cells, benign cells adjacent to the tumor, and cells from healthy prostate tissue. The investigators revealed that CpG island methylation did not clearly correlate with RNA expression and could not explain, by itself, the suppression of gene expression.
Subsequent efforts in Dr. Luo’s laboratory combined whole-genome and transcriptome sequencing, and identified several fusion transcripts that were present in cancer cells. Eight of the fusion constructs that were found were validated using reverse transcription polymerase chain reaction (RT–PCR), Sanger sequencing, and fluorescence in situ hybridization. Most patients harboring one of these fusion proteins showed a higher risk of recurrences, metastases, or prostate cancer-specific death after surgery.
One of these fusion constructs, MAN2A1-FER, is a fusion between MAN2A1, which encodes a Golgi enzyme involved in converting mannose to complex N-glycan molecules for the glycosylation of membrane proteins, and FER, an oncogene that encodes a tyrosine kinase. The fusion construct harbors a deletion of the 441 C-terminal amino acids from MAN2A1 and of the 571 N-terminal amino acids from FER. Dr. Luo and colleagues identified the same fusion protein in a cancer cell line and in six different human cancers, including ovarian cancer, liver cancer, and glioblastoma multiforme.
Co-immunostaining revealed that MAN2A1-FER localizes to the Golgi apparatus. To test whether this fusion protein retained its tyrosine kinase activity, Dr. Luo and colleagues expressed and purified the construct from bacterial cells for in vitro enzymatic studies.
“We found that the fusion construct had a four-fold higher kinase activity,” reports Dr. Luo. This finding suggested that loss of the SH2 domain from FER, which is not part of the fusion protein, could increase its tyrosine kinase activity.
Subsequent analyses revealed that the fusion construct phosphorylates the N-terminus of EGFR. “We identified the phosphylation site as tyrosine 88, which is not supposed to be phosphorylated or contribute to any kind of kinase activity, because that domain is exposed extracellularly,” notes Dr. Luo.
This tyrosine is situated at the interface between domains I and II of EGFR. “Phosphorylation of tyrosine 88 is essential for the MAN2A1-FER-mediated activation and for dimerization of EGFR,” adds Dr. Luo.
Mutagenesis of this tyrosine residue into alanine, which eliminates the phosphorylation site, abrogated EGFR activation in cell lines containing the fusion protein. This is the first example of a constitutively active tyrosine kinase that is translocated to the Golgi apparatus and leads to the oncogenic activation of the EGFR pathways.
“This tyrosine residue is present only in cancer cells, and it is exposed,” Dr. Luo points out. “Eventually it could become a valuable cellular target.”
1. L. Iversen et al., “Ras Activation by SOS: Allosteric Regulation by Altered Fluctuation Dynamics,” Science 345(6192), 50–54 (July 4, 2014).
2. J.M. Rojas et al., “Mammalian Son of Sevenless Guanine Nucleotide Exchange Factors,” Genes Cancer 2(3), 298–305 (March 2011).
3. S.M. Christensen et al., “One-Way Membrane Trafficking of SOS in Receptor-Triggered Ras Activation,” Nat. Struct. Mol. Biol. 23, 838 (2016).
4. J.J. Gao et al., “Integrative Analysis of Complex Cancer Genomics and Clinical Profiles Using the cBioPortal,” Sci. Signal. 6, doi:10.1126/scisignal.2004088 (2013).