Question to be Answered: Can an individual bee's behavior be predicted by examining a profile of gene expression in its brain?

Worker bees performing nursing activities (comb maintenance and broodcare) and foraging. Photos by Zach Huang.

Individuals Performing this Study:
Dr. Charles Whitfield, Anne-Marie Cziko, and Dr. Gene Robinson, Entomology Department, University of Illinois at Urbana-Champaign

Introduction:
All natural behaviors involve the interaction of an individual with its environment. Scientists have recently been able to show that several forms of behavior are influenced by the activity of specific genes (Whitfield, et. al., 2003). Dr. Gene Robinson, whose research is part of a federally funded project to sequence the honey bee genome, has long been interested in the mechanisms involved in honey bee division of labor as a model to understand the relationships between genes, brain and behavior (Barlow, J., 2003).

Honeybee colonies typically consist of a single queen, thousands of female workers and a small number of males (drones). The queen's activities center around egg laying to insure the vast numbers of individuals required to maintain a hive. Workers, sterile females, provide virtually all of the efforts required to maintain function within a hive. Drones, the male bees within a colony, can be distinguished from workers and queens by their large size, rectangular abdomens, large conspicuous eyes, and noisy flight. All drones lack a sting, and have more eye facets than a worker (6,000-7,000 vs. 3,000-5,000).

Worker (left), drone (middle) and mature queen (right). Photo by Zach Huang.

This study focused on behavioral changes that occur within the female workers over the duration of their lifespan. Newly emerged workers (nurse bees) clean cells and care for the young for their first two to three weeks of life. Shortly thereafter they shift their behaviors to foraging for nectar and pollen for the rest of their six week lifespan (Fox, 2003). As the behavioral transition occurs the bees experience changes in brain structure and brain chemistry (Barlow, J., 2003).

The normally encountered sequence of development from nurse to forager is influenced by occurrences in the colony. For example, if a colony is short of foragers some of the nurse bees will mature more quickly (Fox, 2003) becoming precocious foragers. Experimental removal of foragers has resulted in nurse bees starting to perform foraging activities as young as five days of age. Other experiments have shown that the normal nurse-forager transition is reversible. Experimental removal of nurse bees will result in some foragers reverting to nursing (Whitfield, 2005).

DNA and Experiment Technology Background:
Information encoded in DNA is found in the chromosomes, which are housed in the nucleus of the body's cells. Gene information is converted to messenger RNA, which leaves the cell nucleus and moves to the cytoplasm. Once in the cell's cytoplasm, messenger RNA information is translated into proteins. The amount of RNA produced by a gene is variable and is influenced by the needs of the organism at a particular time.

It has become clear that that thousands of genes and their products (i.e. RNA and proteins) in a given living organism function in a complicated and orchestrated way. Early methods in molecular biology generally focused on single genes. In the past several years a new technology, DNA microarray technology, has attracted tremendous interest. This technology allows monitoring thousands of genes (Barlow, J., 2003) or an entire genome on a single chip so that researchers can have a better picture of the interactions among all of those genes simultaneously [DNA Microarray (Genome Chip) Web Site, 2002]. Specifically, microarray technology allows researchers to get a broad view of gene activity by generating simultaneous measurements of messenger RNA, which directly reflect levels of protein production. The microarry is essentially a plate on which chemicals react with active genetic products, glowing luminescently when exposed to certain lights. Genes are said to be upregulated when they produce more RNA and downregulated when they produce less RNA. Robinson, Whitfield, and Cziko analyzed 5,500 different active honeybee genes in their study. They created their own microarray for the study. In their experiment, yellow luminescence indicated upregulation of a gene, whereas blue luminescence indicated downregulation. Though the microarrays allow for comparison of gene product, they don't give any information regarding actual amounts of gene product (Whitfield, 2005).

Microarray data (also called a "gene expression profile") indicating genes that are upregulated (yellow) and downregulated (blue).

Experiment # 1:
In the first experiment a colony was allowed to undergo typical life cycle development. Scientists measured gene expression in the brains of nurses and foragers exhibiting age-typical behavior. Nurses (5 to 9 days old) and foragers (28 to 32 days old) showed significant differences in brain gene expression for 39% of their genes (Whitfield, et. al., 2003). Experiment #1 made clear that there are indeed statistically significant differences in the profiles of gene expression in the brains of nurses and foragers.

Experiment # 2:
In spite of the fact that Experiment #1 indicated significant differences between profile of gene expression in the brains of nurses and foragers, the researchers had to consider whether the observed gene expression profile differences were due to the differences in activity (nurse vs. forager) or the differences in age (5-9 days vs. 28-32 days). To determine whether differences in activity or differences in age caused the observed differences in gene profile in the brain of the nurses and foragers, the scientists created colonies ("single-cohort colonies") consisting entirely of same-aged bees. They began by creating single-cohort colonies composed entirely of young bees. Because these young colonies lacked foragers some young individuals took on the forager role, becoming "precocious foragers." Later, as the entire worker population of these colonies aged, the lack of production of young, replacement bees causes some individuals to remain working as nurses despite advancing chronological age. Individual brain gene expression profiles were performed for age-matched young nurses and young ("precocious") foragers, as well as age-matched old foragers and old ("overage") nurses.

Gene expression profiles were performed using six individuals from each of the age-matched pairings. Each group was analyzed in conjunction with nurses and foragers from typical colonies, resulting in a total of six sample groups from two colony types, who comprised a total of sixty individuals. Visual inspection of brain gene expression profiles for the 60 bees suggested that most individuals can be readily distinguished as nurse or forager, irrespective of age, genetic source, or colony type (single-cohort or 'typical') (Whitfield, et. al., 2003). To formally test whether gene expression profiles in individual bee brains can predict behavior, each of the sixty gene expression profiles were analyzed using "leave one out cross validation analysis." Each time, a computer would remove a single profile (leaving 59 profiles), determine a criteria for distinguishing between the gene profiles of known nurses and foragers, and then compare that profile to the 59 known profiles. A computer was able to predict role status of the bee for 57 of the 60 gene profiles provided. Experiment #2 made clear that gene expression profiles showed clearly that the brain differences corresponded specifically to the activities of the bee, not the age (Whitfield, 2005).

Microarray data from Whitfield, et. al. experiment showing specific genes that differed in messenger RNA production levels within the experimental groups. Upregulation is indicated by yellow, downregulation is indicated by blue. YN=Young Nurse, YF=Young Forager, OF=Old Forager, ON=Old Nurse

Final Conclusions:
Experimental findings make clear that genes and behavior are more closely related than commonly believed - that nature and nurture are closely entwined (Whitfield, et. al., 2003). A honeybee turns on and off 40 percent of her genes as she matures from being a "nurse" to a forager in her short, busy life (Fox, 2003). According to Robinson, the genome is responding dynamically to changes in the bee's social environment. Genes and behavior go together in honey bees so strongly that an individual bee's occupation can be predicted by knowing a profile of its brain's gene expression. "We have discovered a clear molecular signature in the bee brain that is robustly associated with behavior," said Robinson. "This provides a striking picture of the genome as a dynamic entity, more actively involved in modulating behavior in the adult brain than we previously thought (Barlow, J., 2003)."

References
Barlow, J. (2003). Gene expression tied to social behavior in honey bees. EurekaAlert. Retrieved June 18, 2006 from http://www.eurekalert.org/pub_releases/2003-10/uoia-get100603.php.

DNA Microarray (Genome Chip) - Monitoring the Genome on a Chip. 2002. Retrieved June 24, 2006 from http://www.gene-chips.com/.

Fox, M. (2003). Gene scan tracks a bee as she grows up. PlanetArk World Environment News. Retrieved June 18, 2006 from http://www.planetark.org/avantgo/dailynewsstory.cfm?newsid=22506.

Whitfield, C.W. University of Illinois Bioinformatics Seminar Dealing with Gene Expression Profiles in the Prediction of Bee Behavior, January, 2005.

Whitfield, C.W. et al. (2003). Gene expression profiles in the brain predict behavior in individual honey bees. Science, 302 (5643), 296 - 299.