In the Communal Bee Andrena Scotica (Hymenoptera, Andrenidae) Rj Paxton, I Fries, Nj Pieniazek, J Tengö

Home , Andrena, Andrena scotica, Nomada marshamella

High incidence of infection of an undescribed microsporidium (Microspora) in the communal bee Andrena scotica (Hymenoptera, Andrenidae) Rj Paxton, I Fries, Nj Pieniazek, J Tengö

To cite this version:

Rj Paxton, I Fries, Nj Pieniazek, J Tengö. High incidence of infection of an undescribed microsporidium (Microspora) in the communal bee Andrena scotica (Hymenoptera, Andrenidae). Apidologie, Springer Verlag, 1997, 28 (3-4), pp.129-141. hal-00891412

HAL Id: hal-00891412 https://hal.archives-ouvertes.fr/hal-00891412 Submitted on 1 Jan 1997

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Original article

High incidence of infection of an undescribed microsporidium (Microspora) in the communal bee Andrena scotica (Hymenoptera, Andrenidae)

RJ Paxton I Fries2 NJ Pieniazek J Tengö

1 Lehrstuhl Entwicklungsphysiologie, Zoologisches Institut der Universität Tübingen, Auf der Morgenstelle 28, D-720 76 Tübingen, Germany; 2 Department of Entomology, Swedish University of Agricultural Sciences, Box 7044, S-750 07 Uppsala, Sweden; 3 Division of Parasitic Diseases, Centers for Disease Control and Prevention, 4770 Buford Highway NE, Mailstop F13, Atlanta, GA 30341, USA; 4 Ecological Research Station of Uppsala University, Ölands Skogsby 6280, S-386 93 Färjestaden, Sweden

(Received 7 March 1997; accepted 25 June 1997)

Summary — We document the abundance and distribution of the spores of an undescribed species of microsporidium within its host, the communal bee Andrena scotica, and relationships between this parasite and its host. Only the host’s adipose tissue (fat bodies) appeared infected, with up to 118 x 106 spores per bee. All hosts at one field site were infected. High spore load within hosts appeared to curtail female reproductive activity. However, we were unable to discern the microsporidium’s effects on male fecundity in that some males with a high spore load did undertake mate searching activity. Longevities of naturally infected A scotica males and females in a flight cage were apparently unaf- fected by the microsporidium. Spores were found in A scotica from other field sites but not in other species of bees, suggesting the microsporidium has a high degree of host specificity. Andrena scotica / communal bee / infection / Microspora / microsporidium

INTRODUCTION though for many their host specificity is little known. Their effects on hosts vary from Microspora is a large group of obligate intra- the benign to highly pathogenic, and they cellular parasites that appear to affect all may have significant effects on the popula- major animal groups. They are particularly tion dynamics of their hosts (Briano et al, widespread and abundant in arthropods, 1995)..

* Correspondence and reprints Tel: (49) 7071-2975342; fax: (49) 7071-296950; e-mail: [email protected] Notwithstanding their apparent preva- 1989). It is univoltine, imagines flying from May lence in arthropods, only three species of to early July at our principal study site, Törn- bottens on the Baltic island of microsporidia are currently known from the Stugby, Öland, SE Sweden (16°34’ E, 56°29’ N, 35 m asl). At over 20000 described species of bees this site, all nests of A scotica have been 1974). Nosema perma- (Apoidea) (Michener, apis nently marked with plastic and metal tags. Adult and Nosema ceranae are of the parasites female bees are communal, that is, two or more, ventricular epithelial cells of Apis mellifera and rarely up to several hundred, females share a and Apis cerana, respectively (Fries et al, common nest entrance (Paxton et al, 1996). 1996). The former can have a severely debil- Beneath the nest entrance, each female constructs itating effect upon its host in temperate cli- her own brood cells in the soil, mass provisions mates (Fries, 1993). Nosema bombi, origi- them with pollen and nectar, and in each she lays one Larvae consume the stored food and nally confused with N apis (eg, Showers et egg. then metamorphose into adults before entering al, 1967, but see and Vette, 1993; Eijnde winter diapause in their natal cells (Paxton et al, McIvor and Malone, is an infection of 1995), 1997). Offspring emerge above ground in the apparently numerous species of Bombus following spring through their mother’s nest across the world (MacFarlane et al, 1995). It entrance; sons remain above ground in search of parasitises cells of host Malpighian tubules, mates whilst mated daughters construct and pro- and possibly other tissue types too (Eijinde vision their own cells in fossorial nests. Imag- ines that and in one do not sur- and Vette, 1993), though its effects on its emerge fly year vive to the hosts are seemingly benign (Fisher and following year. Pomeroy, 1989). In a recent study of Euro- pean bumble bees, Shykoff and Schmid- Collection of material Hempel (1991b) found a low prevalence of N bombi. However there is in general little from the host bee, A scotica data on the abundance and distribution of microsporidian parasites in natural popula- A random sample of host males and females was collected from Törnbottens Stugby in 1995 as tions of or on their effects on hosts. hosts, they first emerged in May and June from their In this paper we report on a high inci- natal nests (for details, see Paxton and Tengö, These A scotica were dence of the spores of a currently unde- 1996). ’emerging’ weighed scribed microsporidium within the haemo- on an electronic benchtop balance to ± 0.1 mg to provide an estimate of their size (Rust, 1991). coel of imagines of a primitively social (communal) bee Andrena scotica Perkins Additional, reproductively active males were 1916 (= Andrena jacobi Perkins 1921) in collected at random from Törnbottens Stugby in 1995 as undertook their characteristic mate SE Sweden. We document the distribution they searching flights (Tengö, 1979) over bushes and and abundance of these microsporidian low trees (2-10 m height), and additional, repro- and some spores within and among hosts ductively active females were collected at random relationships between host and microsporid- in the same year as they returned to nests at Törn- ium. A full description of the microsporid- bottens Stugby with pollen provisions. The head ium will be given by Fries et al in a future widths of these imagines, termed ’field’ bees, were measured under a dissection paper. microscope using an eyepiece graticule to ± 0.03 mm (16 x magnification) in lieu of weight as an estimate of body size because weights of field bees vary con- AND MATERIALS METHODS siderably depending upon the quantity of nectar and pollen that they carry. The host, Andrena scotica ’Overwintering’ adults were excavated from their underground natal cells at Törnbottens The host, A scotica, is a fossorial bee, common Stugby in the winter of 1995-96 and their head and widespread in northern Europe (Westrich, widths were also measured. Adipose tissue (fat bodies), brain, Dufour’s carefully examined for the presence of gland, Malpighian tubules, ovarioles, small intes- microsporidian spores. In 1996, A scotica female tine, thoracic muscle and the ventriculus of ’field’ bees were also collected from flowers at infected imagine hosts were carefully examined Kalmar on the Swedish mainland, 14 km west using phase contrast light microscopy (400 x of Törnbottens Stugby. Fat bodies of some of magnification) to locate microsporidian spores. these bees were carefully examined for Spores were removed from host tissue, fixed and microsporidia whilst other Kalmar ’field’ bees stained for transmission electron microscopy to were homogenised to quantify the numbers of aid in identification (Fries et al, unpublished microsporidian spores they contained, as data). described above. To determine the incidence of infection of A small number of bees of other Andrena ’emerging’ host bees at Törnbottens Stugby, species was collected from Törnbottens Stugby imagines were dissected under insect saline (0.9% and Kalmar in 1996 and examined for NaCl) and scored for the presence or apparent microsporidian spores. In addition, imagines of absence of microsporidian spores in their adi- a cleptoparasitic bee, Nomada marshamella pose tissue. For other ’emerging’ hosts, and all (Kirby, 1802) (Hymenoptera, Anthophoridae), ’field’ and ’overwintering’ bees collected from whose larvae consume the offspring egg or young Törnbottens Stugby, we quantified the number of larva in an A scotica cell and then consume the spores per bee by homogenising imagines in I food stores of that cell, were collected as they or 10 mL of insect saline and counting first emerged in spring 1995 from the natal nests microsporidian spores in the homogenate using of their host A scotica at Törnbottens Stugby. a Neubauer Improved haemocytometer (depth They were also carefully examined for the pres- 0.1 mm, see Cantwell, 1970). ence of microsporidian spores.

Longevity of infected host bees Statistical analysis

To document the relationship between the abun- A log-linear model (logit analysis, Norusis, 1990) dance of the in a host and host microsporidium was used to compare (i) differences among host A scotica were collected as longevity, imagines sex, (ii) differences among host nest of origin, from their natal nests at Törnbot- they emerged and (iii) the interaction between host sex and tens in 1995, and Stugby weighed individually nest of origin in the proportions of A scotica marked on their thoraxes. Then were held in they emerging with or apparently without spores in a m x 1 m x 1 m) in nylon netting flight cage (1 1995 at Törnbottens Stugby. The relationship of a room illuminated and with an ad by daylight host size at emergence (ordinate) upon date of libitum of solution and water. supply honey Upon emergence (abscissa) was analysed by linear death, the date was recorded and immediately and differences among bees with or the number of in the indi- regression, microsporidian spores apparently without spores in (i) their weight at vidual bee was as described ’cage’ quantified emergence and (ii) their date of emergence were above. tested by ANOVA. Data on the number of spores per individual A scotica were square root transformed to conform in A scotica at Microsporidian spores to the assumptions of ANOVA. Then ANOVA other sites and in other species of bees was used to compare variation in the mean number of spores per individual bee at different peri- To ascertain whether the microsporidium was ods in adult life (at ’emergence’, from the ’field’ specific to A scotica at Törnbottens Stugby, the (site Törnbottens Stugby) when reproductively species was sampled from three additional sites. active, at death within a flight ’cage’, ’overwin- In 1995, A scotica individuals were collected tering’, and from the ’field’ [site Kalmar] when from flowers at Abbantorp and Tävelsrum, 12 km reproductively active) for both males and north and 3 km south-west of Törnbottens females. Where there was significant hetero- Stugby, respectively, and both on the island of geneity amongst groups, means were compared Öland, the former site probably well beyond the a posteriori using Fisher’s protected least sig- typical flight range of A scotica. These bees were nificant difference.

Linear regression was used to examine the as brain, Dufour’s gland, Malpighian relationship of the number of spores per imago tubules, ovarioles, small intestine, thoracic host (ordinate) upon the size of the host muscle and ventriculus despite careful measured as or head width, (abscissa), weight does not for bees in each period of their adult life. Spear- inspection. However, this preclude the existence of of the man rank correlation was used to investigate the vegetative stages relationship between imago longevity in a flight microsporidium in one or more of these tis- cage and number of spores at death. Statistical sue types. analyses were performed with the computer pack- ages SPSS (Norusis, 1990) and STATVIEW (Abacus Concepts, 1995). Prevalence of microsporidian spores in A scotica RESULTS There was a high incidence of infection of A scotica the Of the 87 Transmission electron microscopic investi- by microsporidium. male and 56 female A scotica that were col- gation of infected A scotica tissue revealed lected at first emergence from their natal spores with an internal anatomy compris- nests at Törnbottens Stugby in 1995 (six ing a complex cell wall and a coiled polar nests with n ≥ 10 bees examined) and whose filament typical of microsporidian spores fat bodies were examined dissection and (Larsson, 1986, 1). Their by fig unique 87.4% contained exospore structure and molecular charac- light microscopy, clearly numerous microsporidian spores (table I), teristics indicate that they represent a new there no difference between the species (Fries et al, unpublished data). All being appar- spores found in infected A scotica and reported here were of a similar size and shape, suggesting they belonged to the same species of microsporidium.

Location of microsporidian spores within an imago A scotica

Microsporidian spores were primarily located within the adipose tissue of the gasters of both male and female A scotica imagines. Where infection was particularly heavy, the normally pearly white fat bodies that cover the terga and sterna of the host bee appeared grey and became amorphous masses, basically sacs of microsporidian spores (fig 2) that were often ruptured, fill- ing the haemocoel with spores. In such hosts, spores could also be found in the haemolymph of the head and mesosoma, presumably having been transported there from the haemocoel in haemolymph. Spores of the microsporidium were not detected in other tissue of the host bee such ent incidence of infection of male and F = 1.334, df = 1 and 109, ns; females, female host bees (G = 0.062, df = 1, ns). F = 0.231, df = 1 and 67, ns). Hence infected There was significant variation across nests adults of either sex were no heavier than in the proportion of bees that emerged appar- apparently uninfected ones at emergence ently containing spores (G = 14.387, df = 5, (table II), and they could not be distin- P < 0.05), internest variation that was con- guished visually from apparently uninfected sistent across both males and females (3- hosts in external morphology nor in way interaction G = 4.330, df = 5, ns). How- behaviour. ever, infected bees from every nest emerged For male bees, there was no difference (table I). Therefore the microsporidium in the date of emergence of those with or seems to have been distributed within widely apparently without spores (ANOVA its host at Törnbottens population Stugby. F = 2.110, df = 1 and 109, ns). For females, A scotica is sexually size dimorphic; however, infected hosts emerged signifi- imago males are only 40% the size (weight) cantly earlier than apparently uninfected of females (Paxton and Tengö, 1996). There- ones (ANOVA F = 8.301, df = 1 and 67, fore males and females are henceforth anal- P < 0.01) but only by an average of 5 days ysed separately. (table II). Weights of males or females at emergence did not vary significantly across the period of sampling at Törnbottens Stugby Exact counts of microsporidian spores (ANOVA of regression: males, F = 1.733, in A scotica df = 1 and 109, ns; females, F = 2.548, df =1 and 67, ns), a pattern approximately sup- Microsporidian spores were found in all ported by analogous data from the previous bees at Törnbottens Stugby (n = 39 male A spring. Further, for either sex, there was no scotica and n = 41 female A scotica, relationship between the weight of emerging table III) that were homogenised for quan- adults and whether they did or apparently tification of the number of spores per bee. did not contain spores (ANOVA: males, Though approximately 12% of emerging A scotica apparently did not contain spores than those at other periods of adult life, on visual inspection of squashed tissue, the namely at ’emergence’ or held from emer- more thorough observations provided by gence to death in a flight ’cage’ or ’over- homogenisation of hosts suggest that every wintering’ (table III). A scotica imago emerging at Törnbottens Stugby in 1995 contained some spores.

Between 16 x 103 and 118 × 106 spores The relationship between A scotica size were found per bee (table III). There was and the number of spores they contain no difference in the number of spores per bee for A scotica males collected at the four Within any one group of male or female periods in the imago’s life (ANOVA F = bees collected at a different period of imago 1.414, df = 3 and 35, ns, table III). life, there was little or no relationship For A scotica females, in contrast, there between host size and the number of was heterogeneity in the number of spores microsporidian spores they contained; they contained at different periods in their regression analyses of the number of spores adult life (ANOVA F = 27.031, df = 4 and per bee upon host size were universally non- 63, P < 0.001, table III). Reproductively significant (P > 0.05, except for ’cage’ active ’field’ bees contained far fewer spores females, see below). Longevity of infected A scotica numbers of spores per female host upon host in a flight cage size: rs = -0.239, n = 11, ns).

For A scotica adults that were held in a flight cage immediately after emergence in 1995, The occurrence of microsporidian there was no relationship among either spores at other field sites males or females between longevity and and in other species of bees numbers of microsporidian spores they contained at death rank correlation: (Spearman A high incidence of spores of the males, = -0.235, n = 9, ns; females, = rs rs microsporidium was detected in A scotica -0.198, n = 11, ns, fig 3). imagines from sites other than Törnbottens For female A scotica held in the flight Stugby, both on the island of Öland and on cage, there was a significant positive rela- the nearby Swedish mainland (table IV). tionship of the number of spores per host at Indeed, the absolute number of spores in A host death upon host size (weight at emer- scotica ’field’ bees at Kalmar was no dif- gence; ANOVA of regression, F = 8.741, ferent to that of ’field’ bees collected at df = 1 and 9, P < 0.05). However, female Törnbottens Stugby (table III). Microsporid- host longevity in the flight cage was still ian spores were neither detected in other independent of the number of microsporid- species of Andrena collected from Törn- ian spores it contained after partitioning out bottens Stugby and Kalmar nor were they the effect of host size (Spearman rank cor- detected in N marshamella at Törnbottens relation of female host longevity in the cage Stugby, the common cleptoparasite of A versus residuals from the regression of the scotica at this site (table IV). DISCUSSION The microsporidium was clearly widespread among its host, A scotica, at Distribution and abundance several sites in SE Sweden, and it occurred of microsporidan spores in all imagines of both sexes at Törnbottens within and among hosts Stugby when examined carefully. There- fore it can potentially exert a great effect on Fat bodies are a common site of infection the population dynamics of A scotica. We of microsporidian parasites of arthropods could not, however, detect the microsporid- (Raina et al, 1995). For A scotica too, the ium in other bee species, including the host’s fat body appeared to be the only tissue type common cleptoparasitic bee, N mar- that was infected with microsporidia, shamella. Parasites and parasitoids of other although heavily infected hosts also con- hymenopteran species are known to become tained large numbers of spores in their infected with their hosts’ microsporidia haemocoels, suspended in haemolymph. (Blunck, 1954; Briano et al, 1996) and may Relationships between A scotica size and potentially act as conduits for the horizontal microsporidian spore load were either weak transfer of parasites among host species or non-existent. Where a positive relation- (Werren et al, 1995). That N marshamella ship did exist, namely between the number did not apparently contain microsporidian of spores and the size of female ’cage’ bees, spores, though most likely ingesting the there was over a 70-fold range in the num- microsporidium when it consumed either ber of spores per bee for a 1.4 fold range in its host A scotica’s egg, larva or food pro- the size of the host. Variation in the level visions at Törnbottens Stugby, suggests a of infection of a host cannot be accounted high degree of host specificity by the for by the size of the host alone. microsporidium. Costs of microsporidian infection infected A scotia females may correspond to A scotica to the faster physiological ageing observed in honey bees (A mellifera) infected by N Some microsporidia are pathogenic and have apis (Wang and Moeller, 1970). Studies been associated with profound changes in incorporating the experimental manipula- host behaviour and physiology that are detri- tion of levels of infection need to be under- mental to host fitness or even cause host taken in tandem with our correlational obser- death Raina et whilst others (eg, al, 1995) vations to evaluate more fully the costs of have a chronic or benign effect on their host microsporidian infection to A scotica. (eg, Fisher and Pomeroy, 1989). Using nat- infected results from our urally hosts, cage and mode of transmission that A scotica’s Biology experiment suggest of the microsporidium microsporidium had no apparent effect on its host’s longevity in the laboratory. However, That the level of microsporidian infection natural variation among hosts in their sus- did not vary for male A scotica sampled or tolerance to the ceptibility microsporid- across adult life from metamorphosis to ium was not controlled; experimental manip- reproductive activity in the field implies ulation of load would be needed to spore that, once spores develop within a male host demonstrate whether or not the microsporid- larva or pupa, they remain within its adi- ium host reduces longevity. pose tissue without further multiplication. The effects of the microsporidium on Within individual A scotica females, it male A scotica reproductive activity were seems that microsporidian spores follow not immediately obvious; the number of similar dynamics to those in male hosts, spores of some mate searching ’field’ males building up to variable and sometimes high was no different from that of males sam- numbers in overwintering adults but not pled before or at emergence. This is sur- increasing thereafter. prising given that up to 42 x 106 spores were Infection of hosts thus seems to occur at found and that a male’s per male, adipose the egg or larval stage, and is seemingly tissue may appear to have been completely highly efficient, all hosts at Törnbottens exhausted owing to infection. Sperm com- Stugby being infected. Transovarial trans- plements of males may have been affected mission, a common vertical mode of trans- by the microsporidium, though they were mission, is frequent in many invertebrate not measured. microsporidia (eg, Raina et al, 1995). But For female A scotica, reproductively location of microsporidian spores in the fat active individuals (’field’ bees) contained bodies of A scotica adults provides no clue fewer spores than those sampled before or at to the microsporidium’s mechanism or mode emergence, suggesting that females emerg- of transmission. ing with a high spore load did not provision Vertically transmitted microsporidian a nest. Thus high spore load may exert a parasites of other arthropods have been cost to females through reduced fecundity. shown to act as sex ratio distorters of their In other female of arthropods, high rates hosts (Dunn et al, 1995), biasing the sex microsporidian infection of adipose tissue ratio towards female hosts and theoretically have been associated with loss of fecundity enhancing the spread of the microsporid- due to oocyte resorption (Raina et al, 1995). ium in host populations (Hurst, 1993; Dunn It is more difficult to interpret the costs, et al, 1995). Other vertically transmitted if any, of the earlier emergence of A scotica cytoplasmic parasites are also thought or females whose adipose tissue was obviously known to increase their transmission by bias- infected with spores. Earlier emergence of ing the sex ratio of their hosts towards females (Hurst, 1993). In this light, it is hôte-parasite. Les abeilles hôtes ont été interesting to note a female biased sex ratio récoltées lors de leur émergence en un lieu, within A scotica at Törnbottens Stugby, TS, dans le sud-est de la Suède, disséquées though one which may be partly accounted et examinées au microscope à contraste de for by inbreeding (Paxton and Tengö, 1996; phase pour déceler la présence et la locali- Paxton et al, 1996). sation des spores de microsporidies. Le tissu infecté a été examiné en élec- Parasites have been hypothesized to pro- microcopie à transmission Des abeilles vide an important selective force shaping tronique (TEM). hôtes ont été récoltées à divers the genetic structure of social groups (Sher- également moments de leur vie adulte : man et al, 1988; Shykoff and Schmid- pendant dans leurs cellules Hempel, 1991a) and to impose a cost to l’«hivernage» natales, à l’« « au » au cours de group living (Keymer and Read, 1991). We émergence », champ did not collect data explicitly to examine leur activité de reproduction et à leur mort the relationships between the number of A dans une « cage » de vol. Pour chacun de ces stades un a été fait et le scotica females sharing a communal nest, homogénat microsporidian parasitism and the socio- nombre de spores compté à l’aide d’un genetic organization of communal nests. hémocytomètre. Des abeilles hôtes ont été Andrena scotica exhibits high relative intran- récoltées à l’émergence et maintenues en est genetic variability (low nestmate relat- cage de vol jusqu’à leur mort pour détermi- edness, Paxton et al, 1996), a condition ner la relation entre la charge en spores et la which would appear to mitigate the effects longévité. On a également recherché les of parasitism (Shykoff and Schmid-Hempel, spores de la microsporidie dans des spéci- 1991 a). However the high rate of infection mens d’A scotica provenant d’autres sites of A scotica from all nests at Törnbottens et provenant d’autres espèces d’abeilles (y Stugby suggests that microsporidian para- compris Nomada marshamella, l’abeille sitism may have little bearing on A scotica’s coucou d’A scotica à TS). La TEM laisse sociogenetic organization. penser que le parasite est une espèce actuel- lement non décrite de Microspora (fig 1). La microsporidie est limitée au tissu adi- ACKNOWLEDGMENTS peux (fig 2) et à l’hémolymphe d’A scotica. Il semble qu’elle épuise le tissu adipeux des We thank N Gyllenstrand for assistance in the individus fortement infectés. À TS, tous les field and laboratory, R Larsson for his generous nids hôtes renfermaient quelques abeilles help and discussion on the identity of émergentes qui étaient infestées (tableau I). and an referee for microsporidia, anonymous help- Les femelles hôtes infestées ont ful comments on the RJP émergé manuscript. gratefully tôt celles acknowledges the support of the European Union légèrement plus que qui appa- (HCM-program), the Wenner-Gren foundation remment ne l’étaient pas, mais on n’a pas and the DFG and JT that of the Swedish NFR. noté de différence de poids (tableau II). Le nombre de spores dans les homogénats d’A scotica du site TS (jusqu’à 118 x 106 par Résumé — Forte incidence de l’infection individu) suggère que les abeilles étaient par une microsporidie non décrite toutes infectées (tableau III). Ces données (Microspora) chez l’abeille Andrena sco- suggèrent aussi que l’activité de reproduc- tica (Hymenoptera, Andrenidae). Nous tion des mâles n’est pas influencée par étudions l’abondance et la distribution des l’infection microsporidienne, alors que les spores d’une espèce non décrite de Micro- femelles fortement infectées semblent avoir spora au sein de son hôte, l’abeille mono- été moins susceptibles d’approvisionner un voltine Andrena scotica, et les relations nid. Il n’y a pas de relation entre la longévité des hôtes mâles et femelles en cage de vol et und bis zu ihrem Tod in Flugkäfigen gehal- la charge en spores (fig 3). La microsporidie ten. Weiterhin wurde Andrena scotica von semble donc avoir un effet négatif sur la anderen Freilandvorkommen, sowie weitere fécondité de l’hôte femelle et n’influencer ni Bienenarten auf einen Befall hin untersucht la fécondité de l’hôte mâle ni la longévité (darunter auch Nomada marshamella, die des deux sexes. Des spores de la microspo- Kukuksbiene von A scotica, im Freiland TS). ridie ont été décelées chez des individus d’A Nach den Ergebnissen der TEM — Unter- scotica d’autres sites dans le sud est de la suchung ist anzunehmen, daß es sich um eine Suède mais pas chez d’autres espèces bislang nicht beschriebene Art von Micros- d’abeilles (tableau IV). La microsporidie pora handelt (Fig 1). Der Befall war auf das serait donc largement répandue chez A sco- Fettgewebe (Fig 2) und die Hämolymphe tica mais serait hautement spécifique quant von A scotica beschränkt, bei hochbefalle- à son hôte. nen Tieren wurde offensichtlich das Fettge- webe aufgezehrt. Alle Wirtsnester am Unter- Andrena scotica / infection / parasite / suchungsort TS enthielten einige befallene microsporidie / Microspora Tiere (Tabelle I). Befallene weibliche Wirts- tiere schlüpften etwas eher als offensichtlich unbefallene Tiere, ihr Gewicht unterschied sich nicht Nach Zusammenfassung — Gehäuftes Auftreten allerdings (Tabelle II). Spo- an den indivi- von Infektionen bei der kommunal nisten- renzählungen Homogenaten den Biene Andrena scotica (Hymenoptera, dueller A scotica vom Untersuchungsort TS Andrenidae) durch ein bisher nicht wiesen alle Tiere von dort einen Befall auf, es wurden bis zu 118 × 106 beschriebenes Microsporidium (Micros- Sporen pro Tier (Tabelle III). Die Daten nahe, pora). Wir beschreiben die Häufigkeit von gefunden legen Sporen einer bislang unbeschriebenen Art daß die reproduktive Aktivität der männli- chen Tiere durch den Befall nicht beein- von Microspora und deren Verteilung zwischen seinen Wirten, der einjährigen, in trächtigt war, während stark befallene Weib- chen ein Nest Gemeinschaft nistenden Biene Andrena sco- weniger häufig versorgen konnten. Die Lebensdauer in tica, sowie die zwischen die- Flugkäfigen Beziehungen stand weder bei den Weibchen noch bei den sem Parasiten und seinem Wirt. Die Bienen wurden beim Schlupf im Freiland im südöst- Männchen mit dem Mikrosporidienbefall im lichen Schweden gesammelt (TS). Die Tiere Zusammenhang (Fig 3) Das Mikrosporidium wurden präpariert und mit Phasenkontrast- hat damit einen negativen Einfluss auf die mikroskopie auf Befall und Ort des Befalls Fruchtbarkeit der weiblichen Wirte, keinen durch Mikrosporidiensporen hin untersucht. Einfluss auf die der männlichen Wirte, und Befallenes Gewebe wurde mit Transmissi- keinerlei Einfluss auf die Lebensdauer beider onselektronenmikroskopie (TEM) weiter Geschlechter. Die Microsporidiensporen wurden auch an anderen Sammelorten in Südost- untersucht. Die Wirtsbienen wurden zu ver- schiedenen Zeiten ihres Erwachsenenlebens schweden gefunden, aber nicht in den ande- gesammelt: während des Überwinterns in ren untersuchten Bienenarten (Tabelle IV). ihren Brutzellen, beim Schlupf, während ihrer Dieses Ergebnis spricht dafür, daß dieses Reproduktion im Freien und nach ihrem Tod Mikrosporidium zwar weitverbreitet ist, dabei aber einen hohen Grad an auf- in einem Flugkäfig. Die Anzahl von Sporen Wirtsspezifität in Homogenaten der Bienen wurden in einem weist. Hämozytometer bestimmt. Um den Zusam- menhang zwischen Sporengehalt und Andrena scotica / kommunale Bienen / Lebensdauer bestimmen zu können, wurden Infektionen / Microspora / Mikrospori- einige Wirtstiere beim Schlupf gesammelt dium REFERENCES McIvor CA, Malone LA (1995) Nosema bombi, a microsporidian pathogen of the bumble bee Bombus terrestris New Zealand J Zool 22, 25-31 Abacus Concepts (1995) STATVIEW 4.5. Abacus Con- (L). cepts Inc, Berkeley, USA Michener, CD (1974) The Social Behavior of the Bees. Becnel J (1994) Life cycles and host-parasite rela- A Comparative Study. Belknap Press of Harvard tionships of Microsporidia in culicine mosquitoes. University Press, Cambridge, MA, USA Folia Parasitol 41, 91-96 Norusis MJ (1990) SPSS Advanced Statistics 4.0 guide. Blunck H (1954) Mikrosporidien bei Pieris brassicae SPSS Inc, Chicago, IL, USA Ento- L, ihren Parasiten und Hyperparasiten. Z Ang Paxton RJ, Tengö J (1996) Intranidal mating, emer- mol 313-333 36, gence, and sex ratio in a communal bee Andrena Briano JA, Patterson RS, Cordo HA (1995) Long-term jacobi Perkins 1921 (Hymenoptera: Andrenidae). J studies of the black imported fire ant (Hymenoptera: Insect Behav 9, 421-440 infected with a Env- Formicidae) microsporidium. Paxton RJ, Thorén PA, Tengö J, Estoup A, Pamilo P iron Entomal 24, 1328-1332 (1996) Mating structure and nestmate relatedness in Briano JA, Patterson RS, Becnel JJ, Cordo HA (1996) a communal bee, Andrena jacobi (Hymenoptera: The black imported fire ant, Solenopsis richteri, Andrenidae), using microsatellites. Mol Ecol 5, infected with Thelohania solenopsae: intracolonial 511-519 of infection and evidence for transo- prevalence Paxton Hedström L, J varial transmission. J Invertebr Pathol 67, 178-179 RJ, Tengö (1997) Fly parasites and other associates of a communal bee, Andrena Standard methods for Cantwell GE (1970) counting scotica (Hymenoptera: Apoidae), on Öland, SE Nosema spores. Am Bee J 110, 222-223 Sweden. Entomol Tidskr 117, 165-178 Hatcher Tofts C Evo- Dunn AM, MJ, Terry RS, (1995) Raina SK, Das S, Rai MM, Khurad AM (1995) Transo- of transmitted lutionary ecology vertically para- varial transmission of Nosema locustae transmission of a sites: transovarial microsporid- (Microsporida: Nosematidae) in the migratory ian sex ratio distorter in Gammarus duebeni. locust Locusta migratoria migratorioides. Para- Parasitology 111, 91-109 sitol Res 81, 38-44 J van den, Vette N Nosema infection in Eijnde (1993) Rust RW in Osmia and bumblebees (1991) Size-weight relationships honeybees (Apis mellifera L) Cresson terrestris L). Proc Entomol lignaria propinqua (Hymenoptera: (Bombus Exper Appl Kansas Entomol Soc 62, 174-178 Netherlands Entomol Soc 4, 205-208 Megachilidae). J Fisher RM, Pomeroy N (1989) Incipient colony manip- Sherman PW, Seeley TD, Reeve HK (1988) Parasites, ulation, Nosema incidence and colony productivity pathogens, and polyandry in social Hymenoptera. of the bumble bee Bombus terrestris (Hymenoptera: Am Nat 131, 602-610 Apidae). J Kansas Entomol Soc 62, 581-589 Showers RE, Jones FA, Moeller FE (1967) Cross-inoc- bee with Fries I (1993) Nosema apis a parasite in the honey bee ulation of the bumble Bombus fervidus colony. Bee World 74, 5-19 the microsporidian Nosema apis from the honey bee. J Econ Entomol 60, 774-777 Fries I, Feng F, Silva A da, Slemenda SB, Pieniazek NJ (1996) Nosema ceranae n sp (Microsporidia, Nose- Shykoff JA, Schmid-Hempel P (1991a) Genetic relat- matidae), morphological and molecular character- edness and eusociality: parasite-mediated selection ization of a microsporidian parasite of the Asian on the genetic composition of groups. Behav Ecol honey bee Apis cerana (Hymenoptera, Apidae). Sociobiol 28, 371-376 Eur J Pratistol 32, 356-365 Shykoff JA, Schmidt-Hempel P (1991b) Incidence and Hurst LD (1993) The incidence, mechanisms and evo- effects of four parasites in natural populations of lution of cytoplasmic sex ratio distorters in ani- bumble bees in Switzerland. Apidologie 22, 117-125 mals. Biol Rev 68, 121-193 Tengö J (1979) Odour-released behaviour in Andrena Keymer AE, Read AF (1991) Behavioural ecology: male bees (Apoidea, Hymenoptera). Zoon 7, 15- the impact of parasitism. In: Parasite Host Associ- 48 ations: Coexistence or Conflict (CA Toft, A The division of labor Aeschlimann, L Bolis, eds), Oxford University Wang DI, Moeller FE (1970) nosema-infected Press, New York, NY, USA and queen attendance behaviour of worker honey bees. J Econ Entomol 63, 1539-1541 Larsson R (1986) Ultrastructure, function, and classi- fication of Microsporidia. Progress in Protistol- Werren JH, Zhang W, Guo LR (1995) Evolution and of Wolbachia: of ogy 1, 325-390 phylogeny reproductive parasites Proc R Soc Lond B 261, 55-71 Macfarlane RP, Lipa JJ, Liu HJ (1995) Bumble bee arthropods. pathogens and internal enemies. Bee World 76, Westrich P (1989) Die Wildbienen Baden Württem- 130-148 bergs. Verlag Eugen Ulmer, Stuttgart