Bee Breeding
Peter Loring Borst
Danby NY 

Since the discovery of the genetic basis of breeding, great expectations have been placed on the future of bee breeding. Looking at predictions for the future -- from the future -- is always interesting and sometimes instructive. People tend to take past successes as an indication of where we can go. In the propagation of bees, there were great successes prior to the Twentieth Century. 

The honey bee was carried to the New World, where it spread to every country. Then the process was nearly replicated as beekeepers replaced the irascible northern black bee (A. m. mellifera) with the more congenial Italian bee (A. m. ligustica). Naturally it was thought that further “upgrades” would be possible. However, the search for the best bees from around the world was brought to a halt in the United States when the borders were close to importation in 1922, as a move to prevent the entry of the tracheal mite from Europe, thought to be the cause of the Isle of Wight Disease which wiped out thousands of colonies. 

This did not prevent the importation of the African bee (A. m. scutellata) into Brazil, which resulted in most of the Americas being recolonized by another type of bee, one which made the beekeeping with the ornery black bees of the past seem like the good old days. In the story of the African bees’ unexpected and successful conquering of the New World, there are many lessons to be learned. The chief among these are to be careful what you strive for, because the consequences are not always predictable. 

Use of a Pedigree System

Early breeders were undaunted by the import ban, as most of the best strains of bees had already been brought in at one time or another by then, and the new science of Genetics was promising great advances. Prof. George A. Coleman of Berkeley, California writing in the June 1921 issue of the American Bee Journal, laid out the plans. Invoking Gregor Mendel, he described the process of hybridization and the successes of animal breeders in general. He goes on to apply these principles to bee breeding:

> Use of a Pedigree System. The only rational method of determining which individuals are best for breeders is one in which a careful record is kept of the ancestry for at least four generations, and of the behavior of the individual queen for at least two generations.

And yet, he immediately recounts its shortcomings!

> The author has been using this system and the results have been surprising in many ways, disappointing in some, perhaps, but finally resulting in a few strains of Italians which have given a good account of themselves. 

Finally, he points to the future, when things will all work out:

> From these strains, the author proposes to select tested queens and breeders, and to give them as wide a distribution as possible, in order that they may be tried out under varied conditions. We will thus be able to eliminate queens which are not strictly up to standard. (Coleman 1921)

Instrumental Insemination of Queens 

The most significant advance in bee breeding in the Twentieth Century was the mastery of instrumental insemination of queen honey bees. William C. Roberts of the USDA at Baton Rouge, tells of its adoption. By 1946, he says, it proved so useful that natural matings were abandoned in the breeding program. By controlling the heredity, they found that traits such as disease resistance could be intensified. They also found that honey bee breeding was very different from raising livestock. He states:

> It soon became evident that we could not follow the systems of matings with honey bees that were successfully used by other animal breeders. It was discovered that there existed in honey bees a series of lethal allelic genes which in homozygous condition resulted in this low vitality brood. 

In other words, when crosses are made from lineages that are too closely related, they inherit the same sex alleles, which causes the eggs to become males instead of females. The nurse bees recognize them as such and remove the eggs, leaving the spotty brood pattern characteristic of inbred bees. The only way to restore viability to such a lineage is to cross it with another that is sufficiently unrelated, providing the needed variation between the chromosomes of the parents which are combined in the offspring when the eggs are fertilized. 

Undeterred, Roberts began raising hybrid bees in earnest by crossing lines that were inbred but not too closely related to each other, in the manner that enabled great increases in production in corn. Yet the results were not as expected. Quoting Roberts:

> The beekeepers who received these queens reported to us on the performance of these hybrids. We were at first surprised to learn that not all beekeepers considered our hybrids superior to other stocks.

But still, there is always the future!

> Rapid progress in beekeeping economy is expected within a few years. A major portion of this will be the direct result of controlled bee breeding and hybridization based upon the techniques and knowledge developed in our lifetime.

In the American Bee Journal (1969) an article by Roberts and Ward Stanger describes the queen and package bee industry in great detail. It is a story of remarkable enterprise and efficiency where hundreds of thousands of queens, packages and nucs are shipped from California and the southeastern US to northern buyers, mostly during the month of April.  They uncover very little in the finely tuned system that could benefit from improvement. The one suggestion they made was that natural mating with nucs could be replaced entirely by instrumental insemination. 

W. C. Roberts continued working on stock improvement into the 1970s. In the December 1973 issue of the ABJ, the project at the Honey Bee Stock Center is described at length. The center was maintaining 18 inbred lines and experimenting with hybrids using these distinct lines. They assert that there are probably no pure lines existing in the US, and they categorize their types on the basis of color, behavior, or original breeder. For example, line H was started with Hastings stock, characterized as very gentle, almost “timid,” very black, and excessive propolizers. They stated their long range goal as “to develop superior stock that will benefit American agriculture.”

Colony Collapse and the Loss of Diversity

All of this work on breeding has led us to the type of bees present in the US today. When the phenomenon called Colony Collapse was first publicized, one of the predominant theories that arose was the idea that the honey bee in the United States may no longer contain sufficient genetic diversity, leading to a loss of general vigor. Much was made of the narrow selection that had been going on for many decades, with the sources of breeding stock becoming fewer. Quoting Ben Oldroyd, researcher from Australia:

> Some scientists have suggested that because Varroa has seriously reduced the number of feral honey bees, commercial bees are more likely to mate with close relatives than they were in the past, potentially leading to reduced genetic diversity within colonies. (Oldroyd 2007)

But when I asked him if he thought that was really the source of the problem, he said no, -- he just included the reference in the article for the sake of completeness. 

Be that as it may, the country was combed for bee stock and the genetic diversity of it was analyzed. The most definitive conclusions were arrived at by Debbie Delaney and her colleagues. They sampled 178 colonies from 21 commercial apiaries in the West and 185 colonies from 22 Southeastern queen breeders. They contend that most of the one million queens raised in 2005 were derived from less than 500 breeder queens. They compared genetic diversity among populations collected over a period of ten years and concluded that there is sufficient variety in the honey bees of the United States. Genetic material from Australian packages, the Russian bee program as well as the influx of African bees all have more than counteracted the apparent narrowing of the genetic basis of the country’s bee population. However, they go on to state:

> The maintenance of adequate genetic diversity in U.S. commercial honey bee populations will probably depend on the future inflow of new alleles (Delaney 2009)

This project is already well under way. In the February 2012 issue of the American Bee Journal Steve Sheppard describes efforts to obtain new stock to rejuvenate the three primary subspecies of honey bee already present in the US, namely the Italian, Carniolan and Caucasian types. Among other places, the team visited the Republic of Georgia’s Caucasus Mountains. Sheppard writes that this importation of genetic material will increase diversity in the existing populations and provide the basis for the selection of valuable traits. He concludes:

> It seems reasonable to expect that we could make significant and sustainable improvements to our own managed pollinator population by more fully utilizing the resources of the honey bee genome. However, it will take a concerted effort by bee breeders, queen producers and researchers involved in selection programs to improve the genetic stocks of honey bees available in the US. (Sheppard 2012)


Back to the Future

More than one hundred years have passed since Albert J. Cook wrote the following words in his Manual of the Apiary:

> It occurs to me that in this matter of careful selection and improvement of our bees by breeding, rests our greatest opportunity to advance the art of bee-keeping. As will be patent to all, by the above process we exercise a care in breeding which is not surpassed by the best breeders of horses and cattle, and which no wise apiarist will ever neglect. (Cook 1880)

But comparing bees to beef, consider this: a hundred years ago a typical milk cow produced about 3000 lbs per year, a yield which hadn’t quite doubled by 1950. The average yield today is more than 20,000 lbs; Israel reports averaging 30,000 pounds! This is a tenfold increase in production. Meanwhile, the honey bee industry has simply not experienced anything like these improvements. Perhaps there is something intrinsically wrong with conventional approaches to bee breeding. Time to look at genetics in the 21st century. 

Dr. Raymond C. Bushland and Dr. Edward F. Knipling were awarded the World Food Prize for the development of the technique of sterilizing insects as a means of controlling their breeding. In 1954, the Sterile Insect Technique was used to completely eradicate screwworms from the 176-square-mile island of Curaçao. This marked the beginning of the era of genetic engineering applied to insects. The next step would be genetic modification:

> In 1982, an efficient method was reported to integrate exogenous DNA into the genome of an insect so that it is stably inherited as a transgene in subsequent generations. This process of genetic transformation was first demonstrated in the fruitfly Drosophila melanogaster. Hopes were raised that transgenic approaches could also be applied to other insect species. (Wimmer 2012)

Reprogramming the Gene Machine

Essentially, what they did was to reprogram cells by inserting a bit of DNA. DNA is a very long molecule which serves as the instruction code for living cells. By altering a few lines of this code, the cell’s program can be effectively changed. The process of evolution is based upon the continual modification of this code, which sometimes leads to improvements in an organism’s chance of survival. The new genetic instructions are passed on to the offspring and this improved family line may become the dominant type, if they outcompete the organisms still stuck on the old plan. 

There are several methods of incorporating new genetic instructions into the genome of an organism. The DNA can be inserted into the egg and become incorporated as the cell divides and replicates. Or the eggs may be soaked in a solution containing the material that the researcher wishes to add to the eggs. Regardless of the method, the process is hit or miss. Only a few eggs will develop into adults which contain the modified genomes; many die or are simply born unchanged. 

Initially very simple changes were made, such as inserting a gene that causes the cells to produce fluorescent proteins. Maybe you have seen the multi-colored fish that glow when UV light is shined on them. This may seem a bit pointless, until you realize the implications of such a feat. In the first place we can see that technique has worked. Secondly, such a gene can accompany another section of code that will actually have a particular usefulness, and the glowing protein will show that the segment has in fact been successfully incorporated. 

The discovery and development of green fluorescent protein (GFP) earned Martin Chalfie, Osamu Shimomura, and Roger Y. Tsien the Nobel Prize in Chemistry in 2008.  More than ten years ago, bee breeder Sue Cobey and her colleagues were able to introduce GFP into bee eggs by piggybacking the genetic material on drone sperm. However, while the fluorescence could be observed in developing bees it wasn’t incorporated into the genome, so it wouldn’t be inherited. 

In addition to very great technical obstacles to the genetic modification of bees, there are also serious legal and ethical issues. The consequence of bee breeding gone awry has already been played out by the African bee experiment. Further, there are good reasons for wanting to preserve the genetic integrity of such an important species as Apis mellifera, which has evolved over many millions of years. However, there is still much we can do with the information that is being discovered in the genomes of living organisms. 

Genetic Markers for Behavior

> Selective breeding for Varroa-tolerant honey bees is a promising approach in the fight against V. destructor. Here, we described the development of a SNP assay containing 44,000 SNPs partially preselected for Varroa tolerance. This assay will be a valuable tool in the identification of QTL regions for the trait ‘Varroa-specific defence behaviour’ and should also facilitate the future discovery of QTL for other traits. The identification of QTL is the first step in the dissection of a trait and can lead to the identification of trait-influencing genes. Moreover, the assay is highly relevant for honey bee breeding because it constitutes a useful resource to improve genetic evaluation in the honey bee. (Spötter, et al. 2012)


Genetics has moved away from talking about genes. The term “gene” was coined long before anybody even knew what a gene was. When DNA was discovered, people were pretty sure they would find “genes” for everything. Some behaviors seem to have a genetic basis, but others are obviously learned. In the honey bee, behavior is presumably mostly genetic, because we haven’t seen any evidence of bees teaching each other anything, other than the location of food sources.  But instead of finding a gene for hygienic behavior, we have found regions called QTLs (qualitative trait locus) that seem to correlate to specific behaviors. You can think of this as a sort of genetic fingerprint. If one can take a DNA sample from a queen and determine what behaviors she will pass on to the colony she produces, you have a short cut to breeding. Instead of raising hundreds of colonies and picking out one or two to breed from, you could pre-select the breeders based on a simple DNA test.

However, there is one fly in the ointment: the fact that DNA does not contain what an organism is. The genetic component is critical, but it is only a building plan and does not dictate what the structure must be used for. The perfect example of this is the obvious difference between a queen, a worker and a drone. These all contain genomic DNA from their mother, but it is expressed in an entirely different manner for each. We know that a person is not only what is in their DNA, they are a product of this as well as all that they learn from their parents and the world in general, including what each individual discovers about life. The honey bee colony is a society and so we must realize that each hive will develop differently, regardless of how similar their genetic makeup might be.  Ryszard Maleszka writes:

> We already know nearly the entire DNA sequence of the Apis  genome and most of its neural proteome, but the predicted behaviors of the organism do not emerge from this knowledge.  Most types of behaviors depend on interplay between environmental factors and multiple genes operating in highly interconnected, frequently overlapping networks. The discovery of epigenomic mechanisms in Apis brings a fresh perspective to behavioral studies in this organism. (Maleszka 2012)

And taking it one step further, Timothy Linksvayer suggests:

> The evolution and development of complex phenotypes in social insect colonies can only be fully understood within an expanded mechanistic framework of Developmental Evolution. (Linksvayer 2012)

Backing into the Future

As we can see by this overview, many predictions have been made for the future of bee breeding. The future itself seems to be the place where our hopes lodge, where the branches of our dreams bear fruit. Just this casual glance at prior generations’ predictions makes it clear that we are a restless and dissatisfied species. We want to cut out the dead wood, fertilize the soil and get bigger and better apples for our efforts. But to look back is often to end up thinking that we never knew how good we had it. Not all change is for the better, after all. Sometimes the wise move is to hunker down and enjoy what you already have. The ancient Greeks realized that what we clearly see is yesterday and today, laid out as the long path we have taken. Tomorrow, lies behind us, out of view. In a very real sense, we are unknowingly backing into the future.


References

Coleman. GA. 1921. The Principles of Scientific Breeding, as Applied to the Breeding of Queen Bees. American Bee Journal 61(6): 225-27

Cook, AJ. 1880. Manual of the Apiary

Delaney, et al. 2009. Genetic Characterization of Commercial Honey Bee (Hymenoptera: Apidae) Populations in the United States by Using Mitochondrial and Microsatellite Markers Ann. Entomol. Soc. Am. 102(4): 666-673 

Linksvayer TA, Fewell JH, Gadau J, Laubichler MD. 2011. Developmental evolution in social insects: regulatory networks from genes to societies. J. Exp. Zool. (Mol. Dev. Evol.) 318:159–169.

Maleszka. 2012. Elucidating the Path from Genotype to Behavior in Honey Bees: Insights from Epigenomics. In: Honeybee Neurobiology and Behavior, edited by Galizia, Eisenhardt, and Giurfa

Oldroyd, BP. 2007. What’s killing American honey bees? PLoS Biol 5(6): e168. doi:10.1371/journal.pbio.0050168

Roberts, WC. 1969.  The Development of Hybrid Bee Breeding in the U.S. The 21st International Apicultural Congress 23:226-229

Roberts, WC & Ward Stanger. 1969. Survey of the Package Bee and Queen Industry. American Bee Journal 

Sheppard, WS. 2012. Honey Bee Genetic Diversity and Breeding – Towards the Reintroduction of European Germplasm. American Bee Journal 

Spötter, et al. 2012. Development of a 44K SNP assay focussing on the analysis of a varroa-specific defence behaviour in honey bees.  Molecular Ecology Resources 12, 323–332

Wimmer, EA. 2003. Applications of Insect Transgenesis. Nature Reviews Genetics 4, 225-232

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html