HERITABILITY OF STING CHARACTERS IN AFRICANIZED HONEYBEES

Abstract

In this paper we proposed to estimate the heritability of seven morphological characters that compose the sting apparatus of the Africanized honeybee workers. An experimental design to estimate genetic parameters was based on the method developed by Oldroyd and Moran(9). This method was modified to eliminate within-colony environmental effects associated with the additive genetic variance. The estimated h<img SRC="http:/img/fbpe/jvat/v3n2/image1800.gif"> values ranged from 0.17 ± 0.11 (maximum width of bulb of sting stylet and height of the valve of right lancet) to 0.74 ± 0.30 (length of the lancet).

Apis mellifera; honeybees; heritability; sting

Original paper

HERITABILITY OF STING CHARACTERS IN AFRICANIZED HONEYBEES

A. MELO , O. MALASPINA, J. A. F. DINIZ-FILHO

1 Department of Biology/CEIS/CEVAP, Institute of Biosciences of Rio Claro, UNESP, state of São Paulo, Brazil, 2 Department of General Biology, Institute of Biological Sciences, UFG, Goiânia, State of Goiás, Brazil.

ABSTRACT: In this paper we proposed to estimate the heritability of seven morphological characters that compose the sting apparatus of the Africanized honeybee workers. An experimental design to estimate genetic parameters was based on the method developed by Oldroyd and Moran(9). This method was modified to eliminate within-colony environmental effects associated with the additive genetic variance. The estimated h values ranged from 0.17 ± 0.11 (maximum width of bulb of sting stylet and height of the valve of right lancet) to 0.74 ± 0.30 (length of the lancet).

KEY WORDS: Apis mellifera; honeybees; heritability; sting.

INTRODUCTION

In Aculeata Hymenoptera, the ovipositor of females is modified into the sting which is used for hunting and defence. Many recent reviews suggest that the evolution of this characteristic is essential to explain why eusocial behavior arises repeatedly in this group, since large colonies which are very attractive for predators must be carefully defended against vertebrate predators(13,14).

This characteristic (the stinging behavior) is then largely associated with social insects such as bees, wasps and ants. In Brazil, this association was also strongly linked with the introduction of African bees in 1956, since there have been many accidents (some of them lethal). The denomination 'killer bees' was quickly given to these bees, especially by the media in the United States.

It usually is accepted that Africanized bees are composed of distinct hybrid local populations, varying in time and space, due to distinct levels of racial admixture between African (Apis mellifera scutellata) and European (A .m. ligustica and A. m. mellifera) races of honeybees, with a predominance of the African component(5,8). Although differences in the stinging behavior are largely determined by environmental factors, especially temperature and moisture(2), at least a small part of variation in this characteristic should be due to distinct levels of racial admixture. In relation to sting characters, a previous genetic analysis was performed by Stort and Chaud-Netto(15), comparing Africanized and European bees. They related this characteristic to defensive behavior and suggested a polygenic inheritance mode for sting length.

In this paper, we prolonged previous analyses and estimated the additive genetic component for many characteristics which compose the sting apparatus in Africanized honeybees.

MATERIAL AND METHODS

For this study seven morphological characters of sting apparatus were measured in Africanized honeybees workers after dissection of terminalia under stereoscopic microscope. The characters measured followed by a short identification code were: 1- sting length (SL); 2- length of sting stylet (LSS); 3- length of the bulb of sting stylet (LBS); 4- maximum width of bulb of sting stylet (MWB); 5- height of the valve of right lancet (HVL); 6- length of the lancet (LL); 7- number of tooth barb of sting (NTB).

An experimental design to estimate genetic parameters was based on the method developed by Oldroyd and Moran(9), and was described in detail elsewhere(6). In short, combs of 16 distinct colonies were transferred to 4 nursing colonies one day after oviposition of the queens to eliminate within-colony environmental effects (such as nursing patterns) which could be confused with additive genetic variance. Newborn workers were then collected for analysis from each comb which were isolated after pupation by a small metal grid.

A two-level nested analysis of variance (ANOVA)(12) was then used to partition variance components, and the intraclass correlation (t) derived from variation among combs (within nursing-colonies) was used to estimate heritabilities (h ), which is the proportion of additive genetic variance in relation to total (phenotypic) variance. The h is given by h = t/r where r is the average relatedness within colonies, given by r = (1/2z) + 0.25 where z is the number of (unrelated) drones that inseminated the queens. This number is not exactly determined for each experiment, but previous studies with genetic markers(1,7) performed with Africanized honeybees indicated that an average of 17 drones inseminated each queen. So, workers in a colony have the same mother, but may have distinct fathers, and the average relatedness is then related to z, or the probability that a random sample of workers is entirely composed of super sisters (r = 0.75) or half sisters (r = 0.25). Assuming that z = 17 drones, then the relatedness r is equal to 0.2794. Also, with elevated values of z (from 10 to 20, for example), changes in h are small in such a way that errors in exactly estimating this parameter have minor consequences to analyze additive genetic variance(4).

RESULTS AND DISCUSSION

The genetic parameters estimated for the seven morphological characters of the sting are shown in Table 1. The F-values for differences among nursing colonies were not significant at 5% level (except for number of tooth barb of sting), indicating that within-colony effects are not important to determine sting morphology.

of seven characters. TABLE 1. Values of the estimated h of seven characters.

The estimated h values ranged from 0.17 ± 0.11 (MWB and HVL) to 0.74 ± 0.30 (LL). These values are relatively high and within the limits usually established for morphological traits(3,10).

The existence of a reasonably large additive genetic variance for sting morphology opens some possibilities for discussion. In the first instance, it supports previous conclusions by Stort and Chaud-Netto(15) and can provide an adaptive basis to explain differences between African and European honeybees, especially if there is indeed a genetic correlation between sting morphology and defensive behavior. Differences in life-history strategies of these races which surely have an adaptive basis for living in temperate and tropical climates can then produce genetically correlated responses in sting morphology. Also, sting itself could have been primarily selected in these two environments, but this seems to be a less parsimonious hypothesis. On the other hand, differences in sting may be simply an allometric response to changes in worker body size which have also been attributed to adaptations to climatic conditions(11). To test all these hypotheses, a large and detailed study of morphological and behavioral genetic integration should be performed within and between races of honeybees.

For practical purposes, additive genetic variance in sting morphology implies that breeding programs could be performed to reduce sting, decreasing danger in managing practices because sting would be less effective in injecting venom. This possibility, however, should not be completely useful, because some apiculturists in Brazil, paradoxically, do not want to reduce the defensive behavior in their colonies because this is a way to avoid theft in regions of extensive apicultural activities.

ACKNOWLEDGMENTS

Authors thank Sergio A. Pascon for help in experimental work and FAPESP and CNPq for the several grants given to the authors during the development of this work.

REFERENCES

01 ADAMS JE., ROTHMAN ED., KERR WE., PAULINO-SIMÕES ZL. Estimation of the number of sex alleles and queen matings from diploid male frequences in a population of Apis mellifera. Genetics, 1977, 86, 583-96.

02 BRANDEBURGO MAM., GONÇALVES LS., KERR WE. Effects of Brazilian climatic conditions upon the aggressiveness of africanized colonies of honey bees. In: JAISSON P. Ed. Social insects in the tropics. Paris: Universite Paris, 1982: 255-80.

03 COLLINS AM. Quantitative genetics. In: RINDERER TE. Bee genetics and breeding. New York: Academic Press., 1986: 283-304.

04 DINIZ-FILHO JA., BUENO OC., CHAUD-NETTO J., MALASPINA O. Heritability of the number of ovarioles in honey bee workers (Apis mellifera). Rev. Bras. Genet., 1993, 16, 917-21.

05 DINIZ-FILHO JA., MALASPINA O. Evolution and population structure of africanized honey bees in Brazil: Evidence from spatial analysis of morphometric data. Evolution, 1995, 49, 1172-9.

06 DINIZ-FILHO JA., MALASPINA O., CAVALHERI DH. Genetic and within colony environmental components of variation in the hamuli number of africanized honey bees (Apis mellifera L.). Rev. Bras. Genet., 1994, 17, 255-8.

07 LOBO JA. Modelos para análise de frequências gênicas e mistura racial emApis mellifera. Ribeirão Preto: Universidade de S. Paulo, Faculdade de Medicina, 1991. 82 p. [Tese-Doutorado].

08 LOBO JA., DEL LAMA MA., MESTRINER MA. Population differentiation and racial admixture in the Africanized honeybee (Apis mellifera). Evolution, 1989, 43, 794-802.

09 OLDROYD B., MORAN C. Heritability of worker characters in the honeybee (Apis mellifera). Aust. J. Biol. Sci., 1983, 36, 323-32.

10 OLDROYD B., RINDERER T., BUCO S. Heritability of morphological characters used to distinguish European and Africanized honeybees. Theor. Appl. Genet., 1991, 82, 499-504.

11 RUTTNER F. Biogeography and taxonomy of honeybee. New York: Springer-Verlag, 1988: 284p.

12 SOKAL RR., ROHLF FJ. Biometry. 3.ed. New York: W.H.& Freeman, 1995: 859p.

13 STARR CK. Enabling mechanisms in the origin of sociality in the Hymenoptera - The sting is the thing. Ann. Entomol. Soc. Am., 1985, 78, 836-9.

14 STARR CK. In Reply, is the sting the thing? Ann. Entomol. Soc. Am., 1989, 82, 6-8.

15 STORT AC., CHAUD-NETTO J. Study of the size of sting and comparison with the aggressive behavior in Africanized and Italian bees. Ciênc. Cult., 1978, 30, 332-5.

Received 01 March 1996

Accepted 01 April 1996

CORRESPONDENCE TO:

O. MALASPINA - Departamento de Biologia, Instituto de Biociências, Caixa Postal 199, CEP 13.506-900, Rio Claro, São Paulo, Brasil.

Publication Dates

• Publication in this collection
  08 Jan 1999
• Date of issue
  1997

History

• Accepted
  01 Apr 1996
• Received
  01 Mar 1996