"In an article in the journal Anticancer Drugs, the growth inhibitory effects of limonene and other monoterpenes (including perillyl alcohol) on pancreatic carcinoma cells carrying a K-Ras mutation were examined. Limonene caused an approximately 50% growth reduction."


A compound in the fragrant oil found in orange peels has now joined a growing list of substances under study for their ability to stop cancer before it starts.

Called limonene, the ringed, 10-carbon orange oil compound can reduce and prevent human breast cancer tumors in mice, says Michael N. Gould of the University of Wisconsin-Madison. A similar oil compound - originally derived from lavender plants - has proved five to 10 times as potent as limonene and seems less irritating to the digestive system, he reported last week in Orlando, Fla., at the annual meeting of the American Association for Cancer Research.

At that meeting, several researchers described their progress in finding and assessing substances like limonene. Their studies reflect an increased interest in and emphasis on "chemoprevention" - the use of specific chemical agents, not just better diets - to stop cancer before it has a chance to take hold.

"We're showing we can develop drugs to prevent the development of cancer," says Michael B. Sporn of the National Cancer Institute in Bethesda, Md.

In their work, Gould and his colleagues fed mice a diet containing 2 percent perillyl alcohol, the lavender compound. Tumors disappeared in 60 percent of the animals and shrank in an additional 20 percent, they report. Mice with these tumors, even after treatment for cancer, typically develop one or more new tumors within a few months, Gould says. But he and his colleagues observed almost no additional tumors in mice on the perillyl alcohol diet. These results parallel earlier findings in mice on diets with 5 percent limonene, he adds.

In mice, enzymes break the compounds down into metabolites that exert several effects, Gould explains. First they stimulate the production of enzymes that help break down carcinogens.

Then these substances sidetrack proteins produced as the result of cancer-stimulating genes called oncogenes. To cause rapid cell growth, these proteins must acquire a carbon chain that enables them to move to a new home along the inner edge of the cell membrane. The new chemopreventive agents block the addition of the carbon chain, Gould says.

The metabolites also slow the construction of a particular enzyme in the cancer cells' mitochondria, thereby reducing the amount of energy the cells produce. In addition, they increase the levels of a substance called transforming growth factor-beta, which inhibits the growth of breast cancer cells.

In studies of six people who took limonene, Gould found that humans also produce these metabolites, which presumably will have similar effects. But he is just beginning to assess whether limonene fights cancer in people.

Tests in humans have begun, however, for another plant-derived chemopreventive compound, this one from soybeans (SN: 3/28/87, p.206). In February, Ann R. Kennedy and her colleagues at the University of Pennsylvania School of Medicine in Philadelphia began administering this substance, called Bowman-Birk inhibitor, to 24 people with precancerous patches in their mouths.

The study participants gargle and swallow mouthwash containing the inhibitor, consuming an amount equivalent to that in the typical Japanese diet, says Kennedy She expects to finish this trial by fall.

Because so few precancerous spots actually turn into malignant oral tumors, and because this transformation can take many years, Kennedy also is seeking faster ways to determine the effectiveness of the inhibitor. Like other scientists, she hopes to find intermediate biochemical indications of impending cancer that she can monitor during these chemoprevention trials.

To thwart cervical cancer, researchers at the Arizona Cancer Center in Tucson have developed a different way of delivering chemopreventive drugs: via vaginal sponges similar to those used for contraception. The sponges release a synthetic relative of vitamin A. The drug signals skin and other kinds of epithelial cells to stop dividing and become specialized.

About 300 women who had developed precancerous cells, or dysplasia, in the cervix used these sponges for four days at the beginning of the study and again for four days at three and six months. Half of the women were given sponges containing the drug, and half were not. Examinations of the women's cervical tissue over 15 months, sometimes longer, revealed that precancerous spots disappeared in significantly more women with moderate dysplasia who took the drug than in similar women who did not. However, no such significant difference appeared in cases of more severe dysplasia, says Frank L. Meyskens Jr., now at the University of California, Irvine.


By E. Pennisi


Share this with your friends


A phase I trial of limonene in pancreatic and colorectal cancer is planned to start this week at Charing Cross Hospital in London, UK. Prof Charles Coombes' group will enrol up to 30 patients who are refractory to current best treatment in an oral dose-escalating regimen.

Limonene, a simple monoterpene, occurs naturally in many plant oils, including extracts from orange peel and other citrus fruits, and in mint and in dill, caraway, and celery seed oils. Its anti-tumour effect with low toxicity was first reported in rodent models in 1984 by Michael Gould's group at the University of Wisconsin's cancer centre in Madison, Wisconsin.

In rodent models, limonene and derivatives inhibit tumour formation and cause tumour regression. Two mechanisms of action are being actively studied. First, in cell culture, limonene and perillyl alcohol (an extract from lavender oil) inhibit the post-translational isoprenylation of small G proteins (21-26kDa), which include members of the ras group of signal-transduction proteins that are implicated in carcinogenesis. Disruption of isoprenylation prevents cellular localisation and hence protein function.

The second is an intriguing mechanism in a rat mammary tumour model that Gould and co-workers, together with Randy Jirtle from Duke University, have been investigating.(n1) Transforming-growth-factor beta1 (TGF-beta1) is a potent mitoinhibitor, but is normally latent. This compound can be activated by binding to the mannose-6-phosphate/insulin-like-growth-factor-II (M6P/IGF-II) receptor, which also degrades IGF-II, a potent mitogen. Cells from tumours in regression after treatment with limonene had substantially increased concentrations of TGF-beta1 and of M6P/IGF-II receptor. Tumours that did not respond to limonene did not overexpress M6P/IGF-II.

The gene for this receptor is maternally imprinted and hence hemizygous. Modification of the one expressed allele of this gene could prevent terpene-induced up-regulation. Drugs that interact with the M6P/IGF-II receptor may represent a novel class of anticancer agent. One possibility, says Jirtle, is the de-repression of the allele inactivated during imprinting.

(n1) Jirtle RL, Haag JD, Ariazi EA, Gould MN. Increased mannose 6-phosphate/insulin-like growth-factor II receptor and transforming growth factor beta1 levels during monoterpene-induced regression of mammary tumours. Cancer Res 1993; 53: 3849-52.


By David McNamee