IN the 1950s no one thought of the honeybee as particularly adapted to answering genetic questions that were difficult in other organisms. But it has some distinct advantages. For one thing, unfertilized eggs develop into haploid males, whereas fertilized eggs become females, a property it shares with other hymenoptera. But in addition, a queen bee lays some 1000 eggs per day and the drone cells in the comb are distinguishable from those developing into workers. This makes it simple and unambiguous to distinguish between 100% dominant lethals and sperm inactivation and to have sufficient numbers for quantitative studies, much more difficult in Habrobracon (Heidenthal 1945). Finally, the queen bee mates for life and can retain sperm for several years. In the 1950s there was considerable discussion of whether there was any repair of mutations in irradiated sperm. The Drosophila data were not very extensive, since the sperm could be stored only for a week or two (Abrahamson and Telfer 1956). In contrast, in the honeybee sperm a possible repair process could be studied over years.

Fifty years ago I presented research in Genetics entitled “The dosage response curve for radiation induced dominant lethal mutations in the honeybee” (Lee 1958). Having managed my own apiary of 60 colonies during high school and undergraduate years, I realized in graduate school that the honeybee offered unique characteristics for answering specific questions concerning dominant lethals, questions very difficult to answer in other organisms. This research based on my Ph.D. dissertation, at the University of Wisconsin, took advantage of the unique properties of the honeybee. Two significant results emerged. First, as the dosage of gamma radiation was increased, 100% dominant lethals occurred before sperm was inactivated. Second, irradiated mature sperm stored in the spermathecae of the queen did not show repair of mutational damage during a period of 1 year.

In contrast to the advantages just mentioned, there are disadvantages in using the honeybee for studying dominant lethals: (1) the large number of chromosomes (n = 16) and their small size makes this material unsuitable for cytogenetic analysis by standard techniques; (2) each queen must be artificially inseminated, a delicate technique, if matings are controlled; and (3) there is only one reproductive female per colony.

My data on the dosage response curve included the effect of fractionating the dosage and the stability of dominant lethals in sperm stored in spermathecae of queens for 1 year.

In these experiments, I exposed drones to gamma radiation from 500 to 23,000 R and artificially inseminated the pooled semen from simultaneously irradiated drones into virgin queens. The procedure included collecting and counting the eggs from inseminated queens, placing the eggs in colonies manipulated to ensure that a high percentage of the eggs hatched, and later determining the percentage that hatched. The small variance in these experiments among queens receiving similarly treated sperm suggested that no common lethal alleles affected the results. Such lethals are indeed common, because of their association with the large number of multiple alleles in the sex-determining system.

In the results, the number of nonhatching eggs laid in worker cells increased with increasing gamma dosage until >99% of the eggs failed to hatch at a dose of 10,900 R, a dose that approximates 100% dominant lethals with <1% of the eggs hatching. A plateau of <1% eggs hatching occurs from 10,900 to 23,000 R.

At 34,000, 60,000, and 86,000 R, all progeny developed into drones showing parthenogenic development. Since the spermatheca of each queen on examination contained a normal number of spermatozoa, the increase in parthenogenic development at ≥34,000 R must be attributed to sperm inactivation.

Defining a dominant lethal as a modification in one of the gametes that results in the death of the zygote, I concluded that the decrease in viability of eggs laid in worker cells compared to that in controls was due to dominant lethals because (1) honeybee eggs normally hatch whether or not fertilized and (2) there was a severalfold difference between the dose giving 100% nonhatching eggs and the smallest sperm inactivation dose. I found nearly all dominant lethals caused death at the egg stage since there was no appreciable reduction in the viability of larvae and pupae in the progenies from irradiated males.

My data on the dosage response curve included the effect of fractionating the dosage. Fractionating the dose with a 1-hr interval between two 20-min radiation periods and comparing this with a 40-min continuous radiation produced no significant fractionation effect. Variations in the time of exposure also did not affect the shape of the dosage response curve.

In determining the stability of dominant lethal mutations I found the effect of time over 1 year to be insignificant. Because of the long life of the queen, I was able to inseminate queens with irradiated semen, determine the percentage of inviable eggs after they began laying, and 1 year later repeat the same test for inviable eggs on the same females. I concluded there is no significant evidence for recovery from induced dominant lethals in the sperm during the period of 1 year.

Several analyses of variance concerning the dosage response curve showed the curve relating radiation-induced dominant lethals to dosage is mixed, consisting of both a linear and an upward nonlinear multihit component.

After irradiating Drosophila males in the same apparatus used for irradiating drones, I found no significant difference between the average number of dominant lethals per sperm induced in honeybees and Drosophila spermatozoa.

Fifty years ago I stated these conclusions in my summary: (1) That to inactivate honeybee sperm the dosage of gamma radiation must be severalfold higher than the nearly 100% dominant lethal dosage; (2) nearly all induced dominant lethals caused death in the egg stage; (3) after 1 year of storage in the spermatheca of the queen, the proportion of dominant lethals in irradiated sperm did not change; and (4) there was no significant fractionation effect, the percentage of dominant lethals being the same when 2000 R were given in a continuous dose as when given in two equal fractions separated by 1 hr.

The curve relating dominant lethals to dosage shows a highly significant departure from linearity (after correction for natural mortality and saturation), but approaches linearity at low dosages, results quantitatively in agreement with those of Drosophila and Habrobracon. This is consistent with the hypothesis that dominant lethals are due primarily to single-chromosome breaks at low dosages and multiple-break phenomena at higher dosages.

Since the time of this article, there has been a great deal of development of honeybee genetics. As an example, see an earlier Perspectives (Page et al. 2002).

For the past 40 years my research as principal investigator with peer-reviewed federal grants has predominantly involved studying recessive mutations using Drosophila. However, currently with a colleague of mine in the Department of Nuclear Science, Louisiana State University, I am studying dominant lethals induced by different parts of the ionizing radiation spectrum on various stages of embryo development in Drosophila. We have recently submitted a manuscript for publication. Perhaps the old adage, “What goes around, comes around,” could be applied to my 50 years of research.

• Copyright © 2008 by the Genetics Society of America