(ATN) Hypericum: Common Herb Shows Antiretroviral Activity
AIDS TREATMENT NEWS No. 063 - August 26, 1988
John S. James

A chemical in a common plant used in herbal medicine (St. John's wort), and previously tested in humans as an antidepressant, has been found to strongly inhibit retroviral infections in animal and laboratory tests. Researchers writing in the July 1988 Proceedings of the National Academy of Sciences (abbreviated PNAS) suggested that hypericin, an active ingredient found in plants of the Hypericum family, commonly known as St. John's wort, might become a useful AIDS therapy.

The fact that the drug can be given by mouth, has already been used in humans, and is found in a widely available plant customarily used as a medicinal herb and sold at little cost, suggests that this treatment possibility might also be developed through the herbal and alternative-treatment traditions, avoiding the years of delay built into mainstream pharmaceutical development. But the herb must be used carefully, because large doses have poisoned grazing animals when they fed on it, and because many questions about its possible use as an AIDS treatment remain unanswered.

No human trials for HIV have yet been conducted. We have heard through guerilla clinic sources that several persons have started using the herb for this purpose, but they started only a few days ago, so it is too early to see any results.

Antiviral Tests of Hypericin

The July PNAS article (Meruelo and others, 1988) gave a detailed account of tests of hypericin against two retroviral diseases of mice. (A closely related variant, "pseudohypericin", is also found in the herbs and has similar antiviral effects; this article will use "hypericin" to refer to both chemicals.)

Most of their research did not use HIV, presumably because this virus does not cause diseases in animals, preventing direct animal tests of HIV treatments. Also, there is a serious shortage of laboratory facilities set up to work with HIV. But researchers are increasingly interested in testing possible AIDS treatments in the "animal models" which are available -- animal diseases caused by retroviruses, such as feline leukemia in cats, or the mouse retroviral diseases used here.

Meruelo and colleagues did mention that preliminary labora- tory work showed that pseudohypericin could reduce the spread of HIV.

Most of the testing reported in the July PNAS article used the Friend leukemia virus (FV) in mice. A single small dose of hypericin, given up to one day after the injection of the virus, completely prevented the rapid development of disease and death. Long-term survival, for at least the 150-day period of the experiment, was 44 percent for the treated animals, while all the untreated animals died before the 25th day.

No virus could be found in the infected but treated mice. Also, there was no enlargement of the spleen in treated mice ten days after infection, while the spleen was several times the nor- mal size in untreated mice at that time.

The researchers found no toxicity or side effects from the small treatment dose -- even though they ran a panel of 25 blood tests to look for toxicity.

In this study, the hypericin was given by injection -- a more precise way to administer a drug than orally, and more con- venient in animal tests. But oral administration also proved effective.

Hypericin and AZT

The exact mechanism of antiviral action of hypericin is unknown. But Meruelo and colleagues found that it had no effect against reverse transcriptase, meaning that it works by a different mechanism than AZT. This finding suggests that hypericin might work well in combination with AZT.

An earlier published study, cited by Meruelo and colleagues, showed that AZT alone protected mice from FV infection. But the AZT had to be given repeatedly, at doses highly toxic to the animals. Hypericin suppressed the virus with a single small dose, without toxicity.

What about treatment after disease has already developed? Meruelo and colleagues administered the drug early -- up to one day after exposure to the virus. But their paper mentioned unpublished, apparently preliminary work in which "combinations of hypericin/pseudohypericin with (AZT) have been found remark- ably effective in curing mice from FV-induced leukemia at concentrations and frequencies of administration in which each of the two drugs separately was ineffective".

Human Therapeutic Experience: Medical and Scientific Papers

No human study of antiviral effects of hypericin has been published. But there is safety information from at least four decades of testing Hypericum extracts in humans, as an antidepressant and antibiotic. Some of this research suggests that there may be antibiotics or antivirals other than hypericin in the plant.

Meruelo and others cited a 1949 article on antidepressant use in humans. Several more recent articles have been published, mostly in German and Russian; see references below for a partial list. Medical Hypericum extracts may be available in Germany, the Soviet Union, or other countries, but we do not have details at this time. (West Germany has nearly twice as many approved drugs as the United States.)

Herbal Use

We examined several herb books with information about Hypericum. The books were: John Lust, The Herb Book, Bantam paper- back, 1974; Mrs. M. Grieve, A Modern Herbal, Hafner Publishing, New York, 1967; Michael Weiner, Weiner's Herbal: Guide to Herbal Medicine, Stein and Day, 1980; and Paul Schauenberg and Ferdinand Paris, Guide to Medicinal Plants, Keats Publishing, Inc., New Canaan, Connecticut, 1977.

Different plant species have been studied. All the herb books, and most of the scientific papers on hypericin which we found by a computer search, used Hypericum perforatum. But Meruelo and colleagues extracted hypericin from Hypericum tri- quetrifolium Turra -- a different species in the same family. The readily available Hypericum perforatum is also known to contain hypericin.

It is hard to summarize the traditional medical uses of this herb (commonly called St. John's wort), because different herb books list different uses. Antidepressant and related applications predominate.

Most but not all of the herb books warn readers about the potential toxicity of the plant.

Toxicity

All sources we have seen referred to only one form of toxicity of St. John's wort: that excessive doses makes the skin abnormally sensitive to light. Such "photosensitizing" is a side effect of various drugs.

Meruelo and colleagues cited reports that in livestock exposed to very high doses and intense light, the reaction can be severe or even fatal. But in their test of hypericin with over 800 mice, they found no serious toxicity. And toxicity has not prevented human use of herbal extracts as an antidepressant.

Weiner's Herbal recommends against internal use because the plant contains hypericin, and light-skinned persons can suffer dermatitis, burning, and blistering of the skin if exposed to sunlight after using the herb -- depending on the amount ingested and the amount of sunlight.

John Lust's The Herb Book warns that St. John's wort has poisoned livestock and can make the skin sensitive to light. But it does include the herb in a number of combinations for medicinal teas.

It would seem prudent for anyone trying St. John's wort to stay out of the sun. Sunscreens have been prescribed for patients taking other photosensitizing drugs.

One well-regarded herbalist, reached shortly before press time, told us that despite earlier concerns about St. John's wort, the toxicity seen in grazing animals has not been a problem in humans. He believed that the difference was due to the very different stomachs of ruminants. We have been unable to find any published reports of cases of human toxicity; however we have not obtained a number of papers on the herb, so such reports might exist.

Our main concern is that herbal preparations can vary greatly in the amount of active ingredients they contain. Standardized preparations and careful medical monitoring will be needed to determine safe and effective doses (if any) for persons with HIV infection. Fortunately the chemical properties of hypericin suggest that herbal extracts should be easy to prepare, test, and standardize.

Chemical Properties of Hypericin

The Merck Index gives the following information:

* Hypericin decomposes at 320 degrees C. (Therefore it will not be harmed by the heat of boiling water if prepared as a tea.)

* It dissolves in water, providing the water is alkaline.

* Hypericin solutions in water are red below pH 11.5; above pH 11.5 they are green with red fluorescence. These properties should make it easy to test whether an extract contains hypericin, and how much. The Merck Index cites articles which published absorption and fluorescence spectra, which laboratories might use for precise tests.

* The Merck Index also noted that "very small quantities appear to have a tonic and tranquilizing action on the human organism".

Interview with Dr. Meruelo

We called Daniel Meruelo, Ph.D., the principal author of the PNAS article. He is concerned that people have started using St. John's wort or commercial extracts, and hopes they will wait for more scientific evidence. He added that it is unfair to people to develop this potential treatment in an improper way; that people could be made hopeful or desperate to get the substance before it is clear that it could be helpful. The result could be hype, distress, profiteering, and potentially great harm.

We expressed our concern that even if hypericin does work, it would take years to become available as a pharmaceutical; why not try the herb in the meantime? Dr. Meruelo replied that his team had found only minute amounts of hypericin in the commercial preparations they have tested. He doubted that it would be possible to get enough of the chemical from the herb. We asked how that could be a problem, if it is possible to get a toxic dose by taking too much of the herb; no one would want to use more than that. He answered that the hypericin did not seem to be responsible for the toxicity, as his mice showed little or no evidence of toxicity when given the purified chemical.

St. John's wort contains at least 50 chemicals, not only hypericin. Dr. Meruelo commented that earlier published studies which reported no toxicity from medical herbal preparations had not published blood tests to confirm the lack of toxicity.

Dr. Meruelo also said that the work was proceeding very rapidly; more would be learned in the next several months, and if all went well clinical trials might start within a year. Right now his team cannot obtain enough purified hypericin to do toxicity tests in dogs or higher animals; they get relatively little by extraction from St. John's wort, or by known means of chemical synthesis. They are now working on a better method for producing the chemical synthetically; that work is just beginning, however.

Dr. Meruelo also emphasized that everybody involved, at NIH or elsewhere, has been very cooperative with this project; every- one recognizes its urgency. The team, so far financed only by the researchers' institutions (New York University, and the Weizmann Institute of Science in Israel), has been working very hard for almost two years. Now that they have data they hope to get more funding, and to interest a pharmaceutical company.

We asked Dr. Meruelo why he and his colleagues decided to try hypericin in the first place. He replied that there was no reason to believe that there would be an anti-retroviral effect. But earlier papers had reported possible effectiveness of St. John's wort extracts against other viruses, such as herpes simplex and influenza (see references below); since retroviruses are a more urgent problem than herpes, and since one of the team members is a professor of organic chemistry at the Weizmann Institute and had an interest in the plant, they decided to try it.

Comment

Many questions remain. But despite Dr. Meruelo's understandable caution, we do believe it is important to investigate whether a useful treatment based on a simple herbal extract could be developed.

Until now the scientific community has believed that hypericin was largely responsible for the toxicity of excessive amounts of St. John's wort. The lack of side effects in Dr. Meruelo's mice casts doubt on this hypothesis, but does not conclusively disprove it. The question is crucial, because if hypericin does cause the toxicity, then it should be possible to get all of it one could use through herbal teas or other preparations, avoiding the need to wait for chemical synthesis and official drug approval.

The fact that the researchers are just beginning to learn how to synthesize hypericin efficiently, and will then start toxicity tests on dogs, and do not yet have a pharmaceutical company involved, suggests that it will take some time before hypericin arrives in the drugstores. Human testing alone usually takes seven to ten years for U. S. drug approval.

Hopefully standardized herbal extracts can be tested in community-based trials, with good medical supervision and scientific control so that we can quickly learn whether or not the treatment is helpful. Any such efforts using the herbs must be distinguished from efforts to develop the purified chemical. We hope both projects will move quickly.

Meanwhile several people have just started using a St. John's wort tea for HIV, outside of a formal trial. While their reports will be anecdotal and not the equivalent of a scientific trial, they will be the first information we will have on human use of the herb for AIDS/HIV.

Anyone considering using St. John's wort for AIDS/HIV should realize that there is no human experience yet with such use, serious question remain on whether enough hypericin could be obtained this way, and there might be toxicity if one is exposed to sunlight or other strong light. For these reasons, we believe that this herb should not yet be considered a routine alternative treatment. Those who try it at this time are pioneers; they should educate themselves thoroughly, work with others, and take all appropriate precautions.

We have so far heard from two people who made a tea from dried St. John's wort; the tea did not have the characteristic red color of hypericin, suggesting that little of the chemical was present. One person reported nausea after drinking a dose of the tea much larger than suggested in the herb books.

Kolesnikova (1986) reports that extracts from St. John's wort leaves and flowers work better as antibacterials than decoc- tions (teas prepared by boiling) -- and also allow better control of the dosage. We have seen only an abstract of the article and do not know how the extracts were made.

Various tinctures and other extracts of St. John's wort are routinely sold in health-food stores; some do have the expected red color. But companies can legally call a product Hypericum (or St. John's wort) extract even if it contains only infinitesimal quantities of the herb. Clearly we need chemical testing to answer serious doubts about whether various preparations contain enough hypericin to be worth trying. Independent non-profit groups such as HIV-positive buyers clubs could easily standardize doses by testing each lot of commercially available extracts and publishing the results -- avoiding the complications of re-mixing and re-bottling a standard product of their own.

Alternatively, physicians may be able to obtain standardized medical extracts from abroad.

In the United States, St. John's wort is usually harvested in July and August. If more of the herb is needed, it will not be necessary to wait another year, because the same plant species is common in Australia, which has its summer during our winter. The old plants, incidentally, contain the most hypericin, while the young plants seem most toxic to livestock (Horseley, 1934).

AIDS Treatment News will continue to cover hypericin and St. John's wort, publishing updates as new information becomes available.

References

Derbentseva NA, and others. Action of tannins from hypericum perforatum L. on the influenza virus. Language: Russian. Mikrobiol ZH, vol. 311 no. 6, pages 768-772, 1972.

Hoffmann J, and Kuhl ED. Therapy of depressive states with hypericin. Language: German. ZFA (Stuttgart), vol. 55 no. 12, pages 776-782, April 30, 1979.

Horseley CH. Investigation into the action of St. John's wort. J Pharmacol Exp Ther, vol. 50, pages 310-318, 1934.

Kiriliuk ZhI. Treatment of suppurative infection with St. John's wort and kalanchoe preparations. Language: Russian. Vestn Khir, vol. 119 no 9, pages 112-116, September 1977.

Kolesnikova AS. Bactericidal and immunocorrective properties of plant extracts. Language: Russian. Zh Mikrobiol Epidemiol Immunobiol no. 3, pages 75-78, March 1986.

Meruelo D, Lavie G, and Lavie D. Therapeutic agents with dramatic antiretroviral activity and little toxicity at effective doses: Aromatic polycyclic diones hypericin and pseudohypericin. Proceedings of the National Academy of Sciences, USA, vol. 85, pages 5230-5234, July 1988.

Muldner H, and Zoller M. Antidepressive effect of a Hypericum extract standardized to an active hypericine complex. Biochemical and clinical studies. Language: German. Arzneimittelforschung, vol. 34 no. 8, pages 918-920, 1984.

Nozaki J, and others. New antiviral agent isolated from Hypericum erectum with activity against herpes simplex, influenza, rabies, hepatitis B virus, etc. Abstract # C84-129876; we do not have the complete reference.