Journal of Chemical Ecology, Vol. 19, No. I, 1993

MATURATION OF TERGAL GLAND ALKENE PROFILES IN EUROPEAN HONEY BEE QUEENS, Apis mellifera L.

R O Y - K E I T H S M I T H , 1'4'* M A R L A S P I V A K , 2

O R L E Y R. T A Y L O R , JR . , 3 C L A Y T O N B E N N E T T , 1

and M I C H E L L E L. S M I T H 4

1Southern College of Technology 1100 South Marietta Parkway

Marietta, Georgia 30060

2Department of Entomology University of Minnesota

St. Paul Minnesota 55108

3Department of Entomology University of Kansas

Lawrence, Kansas 66045

4Apichemieal Consultants 9401 Grace Lake Drive

Douglasville, Georgia 30135-1758

(Received July 6, 1992; accepted September 8, 1992)

Abstract--ln a series of husbandry and stop-time chemical experiments with honey bee queens, the production of tergal gland alkenes was found to be stimulated by natural mating and not by instrumental insemination. Carbon dioxide, physical manipulation of the sting chamber and vagina, presence of sperm in the spermatheca, egg production, and chemicals transferred via drone semen are demonstrated to not initiate the synthesis of the tergal gland alkenes. The compounds probably do not function as sex pheromones. However, the circumstances and timing of the initiation of production of the tergal gland alkenes strongly suggests a communication role for the compounds within the hive.

Key Words--Hydrocarbons, honey bees, insects, Hymenoptera, mating, nat- ural mating, instrumental insemination, communication, pheromones, exo- crine glands, Apidae, gas chromatography, chemical communication

*To whom correspondence should be addressed at: Apichemical Consultants, 9401 Grace Lake Drive, Douglasville, Georgia 30135-1758.

133

0098-0331/93/0100-0133507.00/0 �9 1993 Plenum Publishing Corporation

134 SMITH ET AL.

INTRODUCTION

Researchers have long recognized the ability of the queen honey bee to regulate hive functions through her glandular exudates; however, in many cases the exact chemicals involved have yet to be elucidated. Evidence that suggests that queens chemically mark rival queen cells for destruction has been published (Caron and Greve, 1979). Kin recognition is believed to be based partly upon cuticular hydrocarbons and other compounds (Page et al., 1991). It has been suggested that Varroa mites chemically cloak themselves in the hive with cuticular hydro- carbons (Nation et al., 1991). Secretions from the tergal gland, located on the dorsal surface of the abdomen, are highly attractive to worker bees in the court surrounding the queen, much more so than the head, the source of the queen pheromone (De Hazan et al., 1989). These secretions are believed to mediate hive cohesiveness. Queens are also thought to produce a close-range sex attrac- tant from the tergal glands. The active compound has been proposed to be decyl decanoate (Espelie et al., 1990).

In 1987, we described a series of mono-and diunsaturated straight-chain hydrocarbons that were found in hexane extracts from queen honey bees, but not workers. The monounsaturated alkenes formed a homologous series of odd- chain compounds from Ca3 to C37 and higher, all unsaturated 15 carbons from one end of the chain (Smith, 1987). Further studies of these compounds (Smith and Taylor, 1990) found a concentration gradient of the compounds on the cuticle of the queen centered on the dorsal part of the abdomen, suggesting that their point of origin was the tergal glands. Other information suggested that their production was controlled independently from the synthesis of the other cuticular alkenes. Recently we have studied the alkene patterns of queens to distinguish Africanized from European queens (Smith and Taylor, 1988; Smith et al., 1992). Part of the last paper described experiments examining the matu- ration of alkene profiles in queens. These same experiments provided prelimi- nary information about the development of the tergal gland alkene patterns, which we decided to explore in greater detail.

METHODS AND MATERIALS

Queens used in this study were obtained from stocks at the Carl Hayden Bee Research Laboratory, Tucson, Arizona, or the apiary of R.-K. Smith, Doug- lasville, Georgia. The Smith stock was originally obtained from the Rossman Apiaries, Moultrie, Georgia. The Mackensen technique (Laidlaw, 1977) was used for all instrumental insemination procedures. Four breeding and rearing experiments provided the data for this study. They were conducted as follows and are summarized in Table 1.

Experiment 1. One-day-old sister larvae from a line called GL at the Tucson

MATURATION OF TERGAL GLAND ALKENES 135

TABLE 1. NUMBERS AND TREATMENT OF BRED EUROPEAN QUEENS OF KNOWN AGE

POSTEMERGENCE

Age II ~ Naturally mated ~

Hours Days Virgin + Sperm c - Sperm d Saline c No eggs Eggs

o o 2 ~

12 0.5 4

24 1 4

36 1.5 7

48 2 12

60 2.5 4

72 3 6

84 3.5 4

96 4 8

120 5 7 144 6 4

168 7 4

180 7.5

192 8 3

216 9 4

240 10 9 264 11 3

288 12 4

312 13

336 14 3 372 15.5 4

384 16

408 17 2

432 18

480 20 2 552 23 5 + 8 n

720 30 840 35 5

1008 42

2 2

2 2

2 g 2 2 3

2 g 2 2 4

2 g 2 1 2

1" 1" 2 1 2 I g -b 2* 1" 1 e + 1 3

3

2 i

2 i

"II at 7 days in all cases except those marked with an asterisk. hAllowed to fly free at day seven; the lack or presence of eggs in the hive is indicated by "No

eggs" or "Eggs , " respectively. CSperm found in spermatheca after dissection. aNo sperm in spermatheca after dissection. elnsect saline used in place of sperm during the II. fSwarm cell queens raised naturally by bees in a hive, captured as they emerged from the cells. g LQ line of queens. hTreated with carbon dioxide at 7 and 10 days. Four were allowed to lay drone eggs while the other four were confined in queen cages and prevented from laying eggs.

iOther Q lines from CHBRC USDA-ARS.

136 SMITH ET AL.

lab were grafted into queen cups, then the cups placed into queen rearing col- onies for rearing and capping. The capped cells were transferred to a 34,35~ 75-85 % relative humidity incubator until the queens emerged. All queens that emerged during a 12-hr shift were assigned the age of 12 hr postemergence. The queens were transferred to holding cages in a queen bank. Queens were removed from the holding cages at intervals of time and terminated by freezing.

Experiment 2. A second group of 1-day-old sister larvae from the GL line were grafted and reared and the newly emerged queens stored in a queen bank as in experiment 1, then divided into two groups. The first group (N = 8) was treated with carbon dioxide at seven days postemergence and again at 10 days, then four were returned to the queen bank with the control group, and the other four treated queens were confined and allowed to lay drone eggs in mating nucs. The treated and control groups were terminated at 23 days postemergence by freezing.

Experiment 3. A third group of 1-day-old sister larvae from the GL line were grafted and reared as in experiments 1 and 2. The newly emerged queens were stored in a queen bank and divided into a control group and four treatment groups. Two treatment groups were subjected to the instrumental insemination technique at day 7 with one group (N = 10) receiving drone semen and the other group (N = 8) insect saline. Both groups were exposed to carbon dioxide again at day 10. The second set of treatment groups was subjected to the instru- mental insemination technique at day 13, with one group (N = 5) receiving semen and the other (N = 3) saline. Samples of the treatment groups and controls were terminated at intervals of time after treatment by freezing. This experiment was repeated with queens reared from a different stock called LQ.

Experiment 4. A group of 1-day-old sister larvae from the GL line were grafted and reared as in the above experiments. The newly emerged queens were transferred to mating nucs and confined to them with queen excluders. The queens were divided into two groups: one treatment (N = 26) and the other control (N = 13). The queen excluders were removed from the treatment nucs at day 7 to allow free mating. By day 9, two days after the queens were allowed free flight, all the queens sampled had sperm in the spermatheca, indicating the queens had mated between days 7 and 9. Samples from the treatment and control groups were removed at time intervals and terminated by freezing.

In addition, a number of queens of known mating histories and ages were examined and included as data in the study. All queens were terminated by freezing and preserved by air drying, then the hydrocarbons extracted and assayed by already described methods (Smith et al., 1992). Gas chromatographic anal- yses were performed on an Hewlett-Packard 5890A FIDGC instrument with a 50-m Ultra-2 column as described previously (Smith et al., 1992). The data were processed through a Macintosh SE 40/40 (courtesy of Apichemical Con- sultants) or a Macintosh SE/30 40/80 (Southern College of Technology) with

M A T U R A T I O N OF T E R G A L G L A N D A L K E N E S 137

the software program StatView SE +~ (Abacus Concepts, 1984 Bonita Ave, Berkeley, California 94704).

The presence of the TG alkenes was expressed as a decimal fraction of the total extractable alkenes. For queens not expressing the TG alkenes the mean of the TG/total alkene ratio was 0.053 (+__0.020 SD, N = 88) and an observed range of 0.018-0.135. For queens displaying the TG alkenes a mean of the TG/ total alkene ratio was 0.500 (+0.188 SD, N = 27) and an observed range of 0.195-0.780.

RESULTS

Experiment 1. I f tergal gland alkenes were produced and present on the queen by the onset of normal mating flights, the expected results of experiment 1 would be substantial development of the TG alkenes by five days (120 hr) postemergence. The results indicated that there was no observed development of the TG alkenes before day 8 (192 hr) postemergence in any sample. The compounds were observed sporadically in virgins ages 8-17 days (192-408 hr), and then with increasing frequency, until by day 35 (840 hr) all virgins exhibited TG alkenes. These data are illustrated in Figure 1.

Experiment 2. Carbon dioxide is commonly used for queen anesthesia dur-

.8-

.6.

FIG.

.5.

.4.

F- .3.

.2.

A

A

A

A A A

A

] .A

A 0

0 100 2 0 0 3 0 0 " 4 0 0 " 5 0 0 " 6 0 0 " 7 0 0 " 8 0 0 9 0 0

Age Hours

1. Ratio of TG alkenes to total alkenes found in extracts from virgin queens of known age postemergence.

138 SMITH ET AL.

ing the instrumental insemination process and acts as a trigger for initiation of egg laying by inseminated queens (Laidlaw, 1977). Experiment 2 was performed to determine if carbon dioxide was acting also as a trigger for production of the TG alkenes in queens. The carbon dioxide-treated queens were divided into two groups of four each to verify that the treated queens had received sufficient gas to initiate egg laying. The fraction of TG alkenes in the total alkene extract per queen at 23 days (552 hr) postemergence is presented in Table 2. The carbon dioxide-treated queens and the untreated queens were not shown to be different by the Mann-Whitney U test (P > 0.9999). It is noteworthy that two of the eight carbon dioxide-treated queens exhibited essentially no TG alkenes, which would weigh heavily against the gas acting as a trigger for production of the compounds.

Experiment 3. The objective of experiment 3 was to determine if there was some chemical agent in drone semen that initiated the production of the TG alkenes or if physical manipulation of the queen's sting chamber or vagina was a sufficient trigger. The amount of TG alkenes produced was expressed as a fraction of the total extracted alkenes and plotted against time, then compared with control virgins. The results are presented in Figure 2. The two early pro- ducers of TG alkenes, ages 192 and 216 hr were both treated with insect saline. The older queen (336 hr) exhibiting the TG alkenes had her spermatheca only partially filled with semen. These data qualitatively indicate that neither the instrumental insemination procedure nor physical stimulation of the vagina is a sufficient stimulus for production of TG alkenes.

Experiment 4. The objective of experiment 4 was to monitor the production

TABLE 2. RATIO OF TG ALKENES TO TOTAL ALKENES EXTRACTED FROM 552-HR

POSTEMERGENCE VIRGIN QUEENS, UNTREATED OR EXPOSED TO CARBON DIOXIDE

Treatment TG/total alkenes

None 0.040 None 0.447 None 0.511 None 0.594 None 0.080 Carbon dioxide 0.033 Carbon dioxide 0.378 Carbon dioxide 0.479 Carbon dioxide 0.681 Carbon dioxide 0.228 Carbon dioxide 0.470 Carbon dioxide 0.033 Carbon dioxide 0.641

M A T U R A T I O N O F T E R G A L G L A N D A L K E N E S 1 3 9

, 7 '

. 6 '

. 5 '

~ . 4

<

~ . 3 '

.2

.1 �84

O O

w a e o e 175 200 225 250 275 300 325

Age Hours

350 375

FIG. 2. TG/total alkenes ratio for queens subjected to the instrumental insemination procedure with either semen or insect saline. Insemination performed at either 7 days or 13 days. Age in hours postemergence. The two queens at 192 and 216 hr exhibiting high ratio of TG/total alkenes are saline treated. The one queen at 336 hr with a high TG/ total alkene ratio is the GL queen with the spermatheca partially filled with semen.

of TG alkenes over time for a group of naturally mated queens and to compare the results against a sister group of queens that remained virgins, yet were allowed to roam free within the hive. Queen excluders served to isolate the queens until they were released on day 7 for mating. The results are plotted as the ratio of TG alkenes in the total extracted alkenes from each queen against the time of termination and are presented in Figure 3. Ignoring any time vari- ation, the Mann-Whitney U test indicated a significant difference between the naturally mated queens and both the virgin queen controls and the instrumentally inseminated queens from the previous experiment (P = 0.0001). The experiment also served to eliminate conclusively the sting glands as the source of the TG alkenes as these were removed from all the mated queens to check for the presence of sperm in the spermatheca. The TG alkenes were still found in large quantities in all of the free mated queens.

D I S C U S S I O N

The first experiment was conducted with the twofold purpose of: (1) deter- mining the minimum age for identification and/or certification of queen honey bees via extracted unsaturated hydrocarbons, and (2) gaining some insight on

1 4 0 S M I T H ET AL.

.8 §

.7

,I, . 6 ,I[, ,tl, ,I,

A A ,II, ,II,

,I, . 5

.4.

+

. 3 §

A . 2 4,

.1.

200 2;0 3o0

q. ,4.

§ A

4 . § ,I. 4 . .I.

4. .41. ,,I. q.

A

A

350 400 450 500

FIG. 3. Ratio of TG alkenes to total alkenes extracted from virgin (A), instrumentally inseminated with drone semen (Q)) and naturally mated (+) queens vs. age in hours postemergence.

the development of the TG alkenes with a possible sex pheromone function in mind. The first objective has been reported (Smith et al., 1992). Queens are ready to mate naturally by days 5-7 postemergenee, requiring that any chemicals participating in the mating process be present by that time. The absence of significant amounts of the TG alkenes within this time window indicates that the compounds lack any sex pheromone function. We extended the observation time of the first experiment and found that virgin queens do not have a particular point in time for initiation of syntheses of the TG alkenes, but that by day 35 postemergence, all seem to exhibit the compounds.

We had examined a very large number of queens (over 700) of European, Africanized, and pure African origin, most of them mated naturally, and had observed the presence of the TG alkenes in all but a very few of the queens. Many of the European queens in those we had examined were known to be less than 20 days old. This suggested that the TG alkenes may be linked to the mated or egg laying status of the queen.

Experiments 2 and 3 probed for specific stimuli for the production of the TG alkenes. Based on the results of these tests, exposure to carbon dioxide, physical manipulation of the sting chamber or vagina, the presence of sperm in the spermatheca, transferred chemicals from the drone to the queen via drone semen, and the production or laying of eggs were eliminated as appropriate

MATURATION OF TERGAL GLAND ALKENES 141

triggers. Experiment 4 verifies that natural mating of queens is a suitable stim- ulus for the production of the TG alkenes.

These experiments establish the presence or absence of the TG alkenes as a quantifiable chemical difference between the virgin and naturally mated state of a queen. Instrumentally inseminated queens are no different than virgins by this test. It has long been recognized within both the research community and the beekeeping industry that instrumentally inseminated (II) queens are not as productive as naturally mated queens. Observable differences are problems with initial introduction and acceptance of the II queens, rapid replacement of the introduced II queen by a queen raised from her eggs and decreased brood pro- duction by II queens (Harbo and Szabo, 1984; S. Cobey and J. Thomas, personal communications). These problems make utilization of II queens for maintenance of European stocks in areas subject to Africanization quite difficult. We suggest that the TG alkenes may play a pivotal role in the care and acceptance of the queen and her eggs by workers in the hive.

Acknowledgments--We thank the many bee breeders who have contributed queens to this study over the last five years. Drs. A1 Dietz and Murray S. Blum, University of Georgia; Drs. Gerald Loper and Gloria Hoffmann, USDA, ARS Carl Hayden Bee Research Laboratory, Tucson, Arizona; S. Cobey, Ohio State University; and J. Thomas, Texas Department of Agriculture, are thanked for their discussions and advice. Dr. James L. Nation, University of Florida, and Dr. John Harbo, USDA-ARS, Baton Rouge, Louisiana, are thanked for their presubmission reviews of the manuscript. Support for this project was provided by Southern College of Technology, Marietta, Georgia, and Apichemical Consultants, Douglasville, Georgia.

REFERENCES

CARON, D.M., and GREVE, C.W. 1979. Destruction of queen cells placed in queenright Apis mellifera colonies. Ann. Entomol. Soc. Am. 72(3):405-407.

DE HAZAN, M.Y., LENSKY, Y., and CASSIER, P. 1989. Effects of queen honeybee (Apis mellifera L.) ageing on her attractiveness to workers. Comp. Biochem. Physiol. 93A(4):777-783.

ESPELIE, K.E., BUTZ, V.M., and DIETZ, A. 1990. Decyl decanoate: A major component of the tergite glands of honeybee queens (Apis mellifera L.). J. Apic. Res. 29(1): 15-19.

HARBO, J., and SZABO, T. 1984. A comparison of instrumentally inseminated and naturally mated queens. J. Apic. Res. 23(1):31-36.

LAIDLAW, H.H., JR. 1977. Instrumental Insemination of Honey Bee Queens. Dadant & Sons, Hamilton, Illinois.

NATION, J.L., SANFORD, M.T., and MILNE, K. 1991. Comparison of cuticular hydrocarbons from Varroa mites and honey bees. Am. Bee J. 131(12):778-779.

PAGE, R.E., JR., METCALF, R.A., METCALF, R.L., ERICKSON, E.H., JR., and LAMPMAN, R.L. 1991. Extractable hydrocarbons and kin recognition in honeybee (Apis meUifera L.). J. Chem. Ecol. 17(4):745-756.

SMITH, R.-K. 1987. Identification of Africanization in honey bees based on extracted hydrocarbons assay. Presented at the Second International Conference on Africanized Honey Bees and Bee Mites, Ohio State University, Columbus, Ohio, March 30-April 1.

142 SMtTH ET AL.

SMITH, R.-K., ROSSMAN, F.R., YORK, H.F. JR., and TAYLOR, O.R. JR., 1988. Stock certification of European Queens. A m Bee J. 128:676-678.

SMITH, R.-K., and TAYLOR, O.R. JR. 1988. Determination of Africanization in Honey Bee Queens. Am. Bee J. 128(12):808-809.

SM~Trl, R.-K., and TAYLOR, O.R., JR. 1990. Unsaturated extracted hydrocarbon caste differences between European queen and worker honey bees, Apis mellifera L. (Hymenoptera: Apidae). J. Kans. Entomol. Soc. 63(3):369-374.

SMITH, R.-K., SPIVAK, M., TAYLOR, O.R., JR., BENNETT, C., and SMITH, M.L. 1992. Chemotax- onomy of honey bees (Apis mellifera L.) Part 3: Identification of Africanization in honey bee queens (Hymenoptera: Apidae). Bee Sci. 2(2):93-105.