A Bonus From Foulbrood

by E. C. HOLST

FOR AMERICAN foulbrood, a bacterial disease of honeybee larvae, and by far the most serious plague of beekeeping, no proved cure is known. Reasoning that a treatment or cure may be discovered through an increased knowledge of all the factors involved in the disease, we investigated the physiology of the causal organism, Bacillus larvae. As is often the case, this pure research yielded much information concerning an antibiotic that is active against several species of bacteria, as well as certain enzymes produced by the bacillus. The information has had immediate practical value, though the original physiological problem has not been solved.

The nutritional requirements of B. larvae were first studied. From findings gained through this work, it was possible to design a superior diagnostic culture medium for this organism that is cheap and easily prepared. With earlier media, an inoculation of at least 50,000 bacterial spores was required before growth occurred, but this one gives growth with only one or two spores. This is of particular advantage in determining whether a honey contains bacteria infectious to bees or is safe for feeding honeybee colonies. B. larvae is pathogenic only to bees, fortunately not to man or animals.

The new medium was satisfactory for routine diagnostic work, but it was unsatisfactory as a stock medium and for the production of the enzymes or the antibiotic to be discussed, because the bacillus on the medium does not complete its life cycle with the formation of spores. A water extract of honeybee larvae added to the medium, though, was found to induce spore formation.

Since larvae are difficult to obtain in quantity, and can only be obtained during a limited season, we tried to find a substitute for the extract. We tried all the known vitamins as replacements, but without success. Asparagus pulp, added to the medium, however, was found to stimulate B. larvae to sporulate. The chemical nature of the substance is being studied further.

We discovered also that during the spore-formation stage of the life cycle of B. larvae an enzyme was produced that rapidly liquefied the casein, or curd, in milk. The enzyme was quite powerful and acted even at 90 C. (194 F.). It was found to be uniformly present in larvae dead of American foulbrood (scales), which contain huge numbers of B. larvae spores. When such scales are dropped into a small amount of diluted whole or reconstituted milk at ordinary temperatures, the milk becomes clear and transparent, usually within 15 minutes. Because no field test for American foulbrood existed, the use of the reaction as the basis for such a test was probed. The test proved specific for American foulbrood; no other disease or material found in the hive gave a "false positive." The crily exceptions were American foulbrood scales that had been fumigated with formaldehyde or paradichlorobenzene. In such cases the reaction was either slower or completely negative. Sulfa drugs did not interfere.

For many years scientists have examined larvae dead of American foul-brood and have found B. larvae, almost without exception, present in pure culture. That is quite unusual, for when animal tissue dies, it is usually attacked at once by a large number and variety of bacteria. The data indicated that B. larvae might be suppressing such secondary invaders in the dead-honeybee larvae by producing an antibiotic.

A few preliminary experiments showed that B. larvae did indeed produce an antibiotic active against a wide range of bacterial species. Plates of nutrient agar were seeded with a soil suspension or market milk. Scales were placed on the surface on the medium, and the plates were incubated. After 24 hours, zones free from bacterial growth surrounded the scales, while large numbers of colonies had grown outside the zones. An antibiotic had obviously diffused out from the scales, preventing bacterial development. Sporulated laboratory cultures also produce the agent.

The results were so encouraging that similar experiments were made in which pure cultures of bacteria were used. Some of these organisms were harmless, but we observed an effect against bacteria that cause boils, typhoid, undulant fever, abortion in cattle, and human and bovine tuberculosis, but not avian tuberculosis. Such action against disease-producing bacteria in artificial culture does not, of itself, mean that the material can be used in animals affected by the organisms. To determine this requires extensive animal experimentation.

A serial dilution test was used for estimating the potency of solutions prepared from American foulbrood scales. The minimum amount of antibiotic required to inhibit growth of a test organism in one milliliter of nutrient broth is considered as one unit. Since Bacillus subtilis proved to be extremely sensitive, it was used as the test organism. By using the serial dilution method, it was shown that different species of bacteria vary greatly in their susceptibility to the antibiotic. In one instance where an aqueous scale extract was used, B. subtilis was inhibited by 1 unit per milliliter, Staphylococcus aureus by 10 units, but Escherichia coli was not inhibited by 100 units. On the other hand, whereas E. colt cells are dissolved by a strong scale extract, S. aureus cells are not.

As a first step in determining whether the antibiotic could be used therapeutically, water extracts of scales were tested by abdominal injection into mice. Massive doses (400 units per gram of the animal's body weight) proved fatal. Doses of 200 units produced acute toxemia for approximately 48 hours, while doses of 50 units produced symptoms of 2 to 3 hours' duration.

Attempts to purify the agent and reduce its toxicity by the means commonly used in the antibiotic work were unsuccessful, because it is insoluble in the usual organic solvents. Several hundred methods of purification and detoxification have been tried, of which three or four appear promising. After a method has been tried, the product is routinely tested for potency by serial dilution assay, and tested for toxicity by intraperitoneal injections into mice. If the mice show no symptoms, the material is similarly tested in guinea pigs. Frequently lots of material show no symptoms in mice, but prove very toxic to guinea pigs, since the guinea pig is an extremely sensitive animal.

It appears that the antigram positive and antigram negative fractions of the antibiotic may be separate factors. By some of the purification methods, both factors come through, while in others the anticoli factor is destroyed or left behind; it has never been separated by itself.

Because some lots of the antibiotic with encouragingly low toxicity and high potency have been produced, the work is being continued in the hope that toxicity can be further reduced if not completely eliminated. Some of these products have seemed very close to the threshold of therapeutic safety.

THE AUTHOR

E. C. Holst is an associate bacteriologist in the Division of Bee Culture, Bureau of Entomology and Plant Quarantine. A graduate of the University of Wisconsin, Dr. Holst has done extensive research on the southern pine beetle.