Science Editor

Given that most cancer victims ultimately are killed by their metastases, identifying ways to target the travels of cancer is an area of research with some urgency.

And frankly, given that many tumor therapies now target all dividing cells, anything that can preferentially target cancer cells likely is to represent an improvement over current approaches. Research described in the September 2004 issue of Cancer Cell might offer the possibility to do both.

The research described in the paper "Cell surface expression of the stress response chaperone GRP78 enables tumor targeting by circulating ligands" was done by scientists from the University of Texas M.D. Anderson Cancer Center in Houston and the University of Sao Paolo, Brazil, and began when the protein GRP78 was discovered on the cell surface of metastatic prostate tumors.

Inside the cell, GRP78 is involved in the cellular-stress response. Specifically, it is "a chaperone protein," said Renata Pasqualini, professor of medicine and cancer biology at the cancer center. "It interacts with other proteins and helps them fold into shape. It was widely known, but only based on its function inside the cells. It was not even really considered for targeting."

After the researchers discovered that GRP78 also exists on the cell surface, they designed peptide ligands that were predicted to bind to the receptor. For those experiments the authors used phage vector constructs to present the targeting peptides to their receptors. Two short peptide sequences were chosen for further study and first evaluated for their ability to bind to purified GRP78, as well as in cell cultures of human DU-145 prostate cancer cells. The peptides showed strong and specific binding under both circumstances.

Change Of Address For Corkscrew Of Death

The scientists next tested the peptide's ability to enter the cell and deliver a therapeutically active substance to cancer cells. In one experiment, the targeting peptides again were expressed as part of a phage vector construct and incubated with DU-145 cells. The researchers then investigated whether the vector construct could be found inside the DU-145 cells.

In another experiment, the targeting peptides were fused to another peptide sequence that can induce cell death.

"Basically, it's this little corkscrew that disrupts the mitochondrial membrane," Wadih Arap, also professor of medicine and cancer biology at the M.D. Anderson Cancer Center, told BioWorld Today. That pro-apoptotic peptide (which has an amino acid sequence, a repeat of KLAKLAK, even sounds vaguely sinister) has been used in previous studies to disrupt other tumors and also to disrupt the blood flow to fat tissue in mouse models of obesity.

"We changed the address, but the death-inducing moiety is common," Arap said. The scientists found that the chimeric protein made of targeting and pro-apoptotic peptides was able to induce suicide in cultured prostate cancer cells.

With respect to the targeting capabilities of GRP78, there's good news and bad news. The first bit of good news is that "we can see very clearly, using immunohistochemistry methods, that [GRP78] is present in both primary and metastatic tissue," in the bone marrow of prostate cancer patients, Pasqualini said.

Another encouraging finding about GRP78 is that the scientists achieved better targeting results in vivo than in vitro. In drug discovery, the opposite often is the case, with compounds that look promising in a test tube unable to live up to expectations in the more complex environment of the whole animal.

Here, when the scientists tested cells in culture, every type of cell they studied, whether tumor cell or not, expressed surface GRP78 and could be induced to undergo apoptosis when exposed to the fused targeting and KLAKLAK peptides.

However, when the fused peptides were used to inject mice with either prostrate or breast tumors, a quite different (and much more encouraging) picture emerged. There, the fused peptides clearly homed in on the tumor tissue. After four weeks of weekly injections, mice that were given the chimeric protein had significantly smaller tumors than with those that received either no treatment or KLAKLAK peptides without targeting peptide attached.

The researchers attribute that difference to the fact that GRP78 is involved in the stress response. That is part of what makes it a good potential target, since tumor tissue often is oxygen- and glucose-deprived and thus expresses high levels of stress-response proteins. On the other hand, even the best of cell culture conditions are "not exactly physiological," Arap notes, and so cells in culture also tend to express high levels of stress-response proteins.

While it is present in high amounts there, GRP78 is not completely specific to tumor cells. A complete characterization of the protein is ongoing and its expression by cell type has not been comprehensively cataloged, but it is clearly not unique to tumors.

Arap notes that the targeting system will not abolish toxicity to normal cells. "Do I worry about toxicity?" he asked. "Sure, I worry. But this is better than worrying about toxicity and not having a targeting system in the first place.

"Most of the differences between cancer cells and normal cells are subtle," he said. "But even a small difference in targeting would have a great impact in the clinic."

Pasqualini, Arap and their colleagues intend to take their findings to the clinic; they currently are working on reproducing their experiments with GMP-grade reagents, which are necessary for the data that are used to support an investigational new drug application filing. They also are doing preclinical studies on dosing necessary for an IND. Arap said the plan is to discuss with the FDA "within the next several months" to see whether they will allow an IND.