--============_-1379316406==_============
Content-Type: text/plain; charset="us-ascii"
 
Dear fellow beekeepers,
 
Judging recent postings to this list there seems to be some interest about
genetic control in honeybees to fight varroa.
 
In 1994 the magazine "Bee World" published an interesting article about
this subject written by Ralph Buechler, a German scientist. IBRA gave me
the permission to republish this article for the benefit of this discussion
group. You will find it as an attachment to this and the next message.
 
Happy reading
 
Hans
 
e-mail: [log in to unmask]
CompuServe: 100045,2556
Fax: ++41 1 633 10 77
 
 
________________________________________________________________________
 
IBRA is the International Bee Research Association, a non-profit
organization formed in 1949 and devoted to advancing apicultural education
and science worldwide.
 
One advantage of joining this association is receiving their journal "Bee
World". It contains timely  and well written articles about all aspects of
beekeeping.
 
They can be reached by the following means:
 
IBRA
18 North Road
Cardiff CF1 3DY
England
 
Tel: ++44 1222 37 24 09
Fax: ++44 1222 66 55 22
e-mail: [log in to unmask]
www: http://www.cf.ac.uk/ibra/index.html
 
I have been a happy member of this association for a number of years now
and can highly reommend joining them.
 
--============_-1379316406==_============
Content-Type: text/plain; name="Varroa_tolerance_(Part_1)_TXT"; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment; filename="Varroa_tolerance_(Part_1)_TXT"
 
Bee World 75(2): 54-70 (1994)            IBRA 1994
 
 
VARROA TOLERANCE IN HONEY BEES-OCCURRENCE,
CHARACTERS  AND BREEDING
 
RALPH B=DCCHLER
 
Hessische Landesanstalt fur Tierzucht, Abt. f=FCr Bienenzucht, Erlenstrasse =
9,
35274 Kirchhain, Germany
 
 
Introduction
 
Although effective treatments against Varroa jacobsoni are now available,
varroa disease still has a damaging effect on beekeeping with Apis
mellifera. Mite infestation constitutes a continuous risk, which may lead
to weakness or loss of colonies as soon as failure of control and treatment
regimes employed occurs. Regular treatments are time consuming and costly;
the repeated use of drugs causes residue problems in the hive products
which may have a negative influence on sales prospects.
 
The options for controlling varroa by use of colony management or
biotechnical measures alone are limited. For a long-term solution to the
varroa problem priority should be given to research on the genetic
improvement of colony defence mechanisms. This need has been increasingly
realized in recent years and it is, therefore,a fascinating task to review
the scientific findings in this field and to evaluate future opportunities.
 
 
Varroa tolerance in natural populations
 
=46or a discussion of colony defence mechanisms against V. jacobsoni it is
sensible to first evaluate the situation of the Asian honey bee, Apis
cerana, the original host. Serious damage to these bees has never been
reported, and treatment against varroa is not needed in beekeeping with A.
cerana. According to different reports, infestation levels range between
zero and several hundred mites. Rath and Drescher(43) observed a maximum
infestation of 798 mites for one A. cerana colony in southern Thailand.
 
Koeniger et al.(25) were the first to report infertility of V. jacobsoni in
A. cerana brood samples from Sri Lanka. Several authors later confirmed
this observation. Successful reproduction of mites in worker brood was
found only in exceptional cases or under experimental
conditions'(17,42,58). Peng et al.(40) observed an intensive self-,
nestmate- and group-cleaning behaviour (hygienic behaviour) as a reaction
by A. cerana workers to mite infestation: within two hours, after an
artificial infestation, 99.6% of themites were removed by the bees and
partly killed. In a similarly designed investigation B=FCchler et al.(15)
confirmed the more effective grooming behaviour of A. cerana compared to A.
mellifera, although the absolute success rate was lower than reported by
Peng et al.
 
In another study Peng et al.(41) discovered that A. cerana workers emptied
varroa infested A. mellifera brood cells introduced into their colonies,
but this was dependent on the rate of infestation of the cells. The
brood-removal behaviour initiated by varroa was systematically investigated
by Rath and Drescher(43). They found 98.8% of artificially infested brood
cells were emptied within six days. Rosenkranz et al.(51) demonstrated that
the source of mites has an important influence on the removal response.
Brood cells with mites in A. cerana colonies are regularly recapped after
the mite has been removed, while cells with mites from A. mellifera
colonies are usually completely emptied.
 
In some regions a high level of tolerance to varroa was noted in A.
mellifera colonies. Ruttner et al.(55) reported on colonies in Uruguay
which could resist varroa. infestation without any treatment; this was
mainly due to a 70-90% infertility of the mites in worker brood cells.
Similar observations were reported by Ritter and De Jong(46), Camazine(16)
and Engels et al.(19) for Africanized colonies in Brazil. Ritter and
DeJong(46) also observed 53% of the mites in A. m. ligustica control
colonies to be infertile, while Camazine(16) found only 25% of the mites to
be without offspring in con  trol colonies of European origin.
 
In 1984, De Jong et al.(18) pointed out the strong effect climate can exert
upon the development of varroa infestations. In a comparison of three
different climatic regions in Brazil, and with Africanized bees and hybrids
from Africanized and=20A. m. Iigustica bees, Moretto et al (32) found that
the effect due to climatic conditions dominated over the genetic effects on
the varroa infestation. In the temperate climate of Argentina, Marcangeli
et al.(30) estimated that, depending on season, between 28 and 44% of the
mites in A. m. ligustica colonies were infertile.
 
In Tunisia, A. m. intermissa is able to resist high infestations without
any treatment. Ritter(45) reported the comparatively high number of
infertile mites (20-50%) as one reason for the increased tolerance of these
bees. In a current report Ritter and Boecking(47) mention intensive
grooming and brood-removal behaviour as other important tolerance
characteristics of the Tunisian bees.
 
 
Tolerance to V. jacobsoni in experimental populations
 
=46or several years Buchler(7) compared different European strains under
standardized conditions. Starting with a uniform mite infestation,
differences of up to seven fold could be seen for mite infestations of
honey bees of different origins after a test period lasting between 1 and
1.5 years. A correlation between mite infestation and the attractiveness of
the brood, and the duration of the capped stage (sealed brood) was found.
No noticeable differences were seen in the fertility of the mites; colony
performance was independent of its susceptibility to the mite. A
simultaneous selection for both performance and tolerance to varroa should
therefore be possible.
 
Otten(37) tested A. mellifera, A. m. carnica and A. m. Iigustica colonies
over periods lasting from 9 to 18 weeks. In three repetitions, the A.
mellifera colonies showed the highest levels of infestation, up to three
times higher than the other races. The capped stage lasted about six hours
longer for the A. mellifera colonies. The number of infertile mites
produced depended on seasonal effects, and there were some differences in
the timing for the three races.
 
=46uchs and Bienefeld(21) used small bee-units to investigate the reaction o=
f
bees to high infestation levels. For bees of eight different origins, which
included various European races and also A. m. capensis, no significant
differences were found in the infestation level at colony failure.
Dependence of the survival period on the infestation level was similar for
all races tested.
 
In another test series, which started with a low initial level of mite
infestation, Fuchs(20) found significant correlations between development
of the infestation and the following characteristics: attractivenes of
brood to mites, fertility of the mites and duration of the capped stage.
There was also a positive correlation between brood attractiveness and the
fertility of the mites.
 
At the Austrian bee institute at Lunz an isolated apiary has been managed
without the use of any treatment against varroa since 1986(53). Colony
losses are compensated for each year by bringing in new colonies which are
requeened by daughters of the surviving colonies. Over the years the colony
losses have declined. The rate of infertile mites in worker brood samples
has increased to between 50 and 60%, which is extremely high for European
conditions. The fact that similar results have been obtained in unselected
control colonies indicated that a change in the mite population has
occurred rather than an increase in tolerance of the colonies.
 
Kulincevic et al.(27) chose a stock of A. m. carnica as the parental
colonies for a bidirectional genetic selection on the characteristic
'percentage of worker brood cells infested by fertile mites' bred from
queens which had survived heavy colony losses in Yugoslavia. In the
following years four filial generations were selected for the same
characteristic: high brood infestation in the susceptible line and low
brood infestation in the resistant line. A significant difference was noted
for the selection criterion in all generations and also for the natural
mite mortality in the third and fourth generations. The results indicate
that there is a considerable genetic additive compgnent for the
characteristic which allows for a classical selection.
 
In=201990, offspring from the resistant Yugoslavian-line were compared with
American carnica- and ligustica-lines in a performance test in Florida,
USA(44). The results from two repetitions at two locations were not
uniform, and the Yugoslavian stock showed no clear superiority in the
tolerance against varroa.
 
Moritz and Mautz(35) tested the infestation development of five A. m.
carnica and five A. m. capensis colonies which were initially infested with
the same number of mites. From the beginning of July until the middle of
October, increasing differences in natural mite mortality, brood
infestation and bee infestation were noted between the two groups. Mite
development was limited in the A. m. capensis colonies and the authors
supposed that a more intensive grooming reaction and a shorter duration of
the capped stage were reasons for this, but mite fertility was not
investigated.
 
A comparison of six A. m. monticola-A. m. ligustica hybrids with a 75%
monticola gene contribution, and four A. m. ligustica control colonies was
mad by Thrybom and Fries(59). Starting with a uniform infestation on 21
May, the infestation level of the monticola hybrids was 28.9% of the
infestation level in the control colonies at the end of August. However,
brood production in the hybrid colonies was significantly lower than in the
control colonies, and this factor has to be considered. The monticola
hybrids had a higher rate of infertile mite production in the worker brood,
and the length of the capped stage may possibly have been shorter although
this was not investigated separately. A further evaluation of monticola
hybrids is recommended in order to increase varroa tolerance in European
honey bees.
 
=46or several years now all colony performance tests carried out at the
Hessian bee institute in Kirchhain, Germany have started with a uniform
initial infestation. Artificial swarms from highly infested colonies are
mixed and divided into samples of uniform weight (fig. 1). One of these
samples, which should be infested with about 100 mites, is introduced into
each test colony at the beginning of the winter period (fig. 2). The
infestation level of these samples varies with in 10-15%(8). After the
following year's honey harvest, the colonies are treated with a contact
acaricide and dead mites are collected to determine the total infestation
level. Differences between the colonies are used to estimate the tolerance
to varroa.
 
As an interim result of the studies, it may be stated that there are great
differences in the susceptibility to mites within and between different
races of the European honey bee, A. mellifera. Differences in
susceptibility to mites are found to be correlated with several
characteristics of the bees (fig. 3).
 
 
Characteristics of honey bees affecting tolerance to varroa
Grooming as a mite defence behaviour
 
The behavioural mechanisms involved in the grooming activity of A. cerana
against V. jacobsoni are described in the investigations of Peng et al.(40)
and Buchler et al.(15). The infested bee tries to remove the mite by
vigorous movements and a well directed wiping with the legs. Occasionally
mites can be observed being caught in a quick movement of the mandibles. If
the infested bee fails to remove the mite it may demand grooming assistance
from its nestmates by performing a specific shaking dance, with fast
lateral movements of the abdomen. The nestmates inspect the whole body,
paying greater attention to the petiolus region and the wing bases. Foreign
particles are picked off with the mandibles and are eventually chewed.
 
In both studies a comparison with A. mellifera in observation hive
experiments showed similar behaviour patterns for both species, but with a
reduced intensity and effectiveness in the European honey bee.
 
In a comparison between Africanized honey bees and A. m. ligustica in
observation hives Moretto et al.(33) found great differences in the
grooming success. Within 30 min after infestation only 5.75% of the mites
were removed by A. m. ligustica bees, but 38.5% were removed by the
Africanized bees.
 
Ruttner and Hanel(54) examined the natural mortality of five A. m. camica
colonies in Austria. From September 1990 to April 1991, on average, 26% of
the mites collected from inserts showedinjuries to the legs but rarely to
the cuticle of the idiosoma. Damaged, but live miteswere found immediately
after emergence from infested brood cells. An anatomical comparison of the
mandibles and their muscles showed no fundamental differences between A.
cerana and A. mellifera. The ability to kill mites using the mandibles,
which is well proved for A. cerana, can therefore also be assumed for A.
mellifera.
 
Moosbeckhofer(31) reported a significant negative correlation between the
number of damaged mites found in the period August/September, and the
infestation of brood and bee samples and the total infestation found in a
field test with 111 colonies. These findings are supported by an
investigation by Buchler(11) on 21 colonies over a one-year period. The
average number of damaged mites found was dependent on the timing of the
control and ranged between 10% in March and 64% in June. The repeatability
of colony effects on the degree of damage was highest in May/June (w =3D
0.48). There were significant differences between the colonies in many
cases. Colonies with a low annual average number of damaged mites (< 30%)
showed a stronger increase in the natural mite mortality than colonies with
a high number of damaged mites (> 40%).
 
Besides the grooming activity of the bees, the damage rate of mites under
field conditions is certainly influenced by several other factors such as
mite mortality inside the brood cells, the physiological status of the
mites, and the presence of ants and other potential mite predators. In
order to exclude these uncertainties, Hoffmann24 developed a laboratory
test method: groups of about 300 bees are kept in cages and are
artificially infested with mites (fig. 4). The mite mortality is checked up
to 7 days after infestation, and the number of damaged mites found is
evaluated.
 
 
Brood removal behaviour of the bees
 
In the papers of Peng et al.(41) and Rath and Drescher(43), brood removal
behaviour is described as a defence reaction by A. cerana towards varroa
infested brood cells. Boecking and Drescher(3) have found a similar
behaviour in A. m. carnica colonies, and in further investigations(4) they
observed that the comb material influenced the removal reaction. The
removal rate on Jenter-combs (plastic queen raising comb with removable
bases) (fig. 5) and standard plastic combs surpassed the rate on natural
wax combs.
 
Hygienic behaviour of bees plays an important role in resistance to
different brood diseases. Rothenbuhle(52) described this behaviour in
relation to resistance against Bacillus larvae. For screening colonies he
put slices of freeze-killed brood into healthy brood combs and checked the
time of uncapping and removing reactions (fig. 6). Comparing this test
method with the brood removal reaction against brood cells infested with
two mites, Boecking and Drescher(4) found a positive correlation between
both tests (r =3D 0.6-0.65). The authors, therefore, recommended the
comparatively simple and well standardized method of using freeze-killed
brood for testing the varroa dependent removal reaction.
 
The removal rate is greatly influenced by environmental effects and the
repeatability of single tests is therefore low(23). The geneticcomponent
can be demonstrated more clearly by repeated measurements. The average
mite-removal rate for nine genetically different lines of A.m.carnica
hybrids for 8-10repetitions using Jenter-comb was correlated with an
average of four repetitions of the test using freeze-killed brood
(r=3D0.79)(23)
 
Hoffmann(23) suggests a further simplification of the testing procedure
involving killing a definite number of pupae by simply puncturing
individual cells within healthy brood combs with a fine insect pin and
measuring the time until the consequent removal reaction (fig.7). For tests
repeated 3-8 times correlations to the cited Jenter-comb test (r=3D0.65) and
to the test with freezed-killed brood (r=3D0.78) were found.
 
B=FCchler(9) observed a negative correlation between the strength of the
removal response and susceptibility to mites in four different A. m.
carnica lines. Offspring from the colonies with the greatest removal
response turned out to show a very pronounced removal behaviour(13). A
considerable additive genetic component of the removal behaviour is
indicated.
 
 
Attractiveness of the brood
 
B=FCchler(6) combined sections of brood of the same age from seven different
origins in an individual frame, which covered A. m. carnica, A. mellifera
and Buckfast strains. These mixed combs were placed into highly infested
colonies one day before cell capping. Later the level of infestation of the
different brood samples was checked. Differences were found which depended
on the origin of the brood. The A. mellifera brood had a significantly
lower infestation, while the infestation rate of the Buckfast brood was
higher than average. These results could be confirmed in a laboratory
choice test . The average attractiveness values obtained for the different
strains of bees tested showed a positive correlation with the varroa
population development in the colonies(7). Also, Fuchs(20) observed
differences in the attractiveness of brood to mites in tolerance tests with
small bee-units and confirmeded the positive correlation between brood
attractiveness and infestation increase.
 
A delayed invasion of brood resulting in an extended interval between
generations may be considered as an explanation of the effect of reduced
brood attractiveness on the population dynamics of the mite. Otten(38)
noted a significant correlation between the proportion of mites in brood
cells (brood mites) to the total level of infestation in a colony and the
growth of the mite population in a test with 47 colonies.
 
After acaricide treatment in several performance test groups Buchler(9)
found, in most cases, a positive correlation between the rate of brood
mites and the infestation level of the colonies.
 
Bienefeld and Stroh(2) tested the attractiveness of brood from several
pairs of reciprocally inseminated queens and stated that a maternal genetic
effect is present.
 
 
Colony influence on the fertility of V. jacobsoni
 
Undoubtedly, the number of infertile mites developing in worker brood is a
key factor for the population dynamics of V. jacobsoni. From the
investigations cited that comment on varroa tolerance in natural and
experimental populations, it is quite evident that geographic and climatic
differences(18,32,46) as well as host specific effects(16,19,46) play an
important role. Within Europe and in the comparison of European strains no
such differences are known. Nevertheless there are some reports about
infertility rates above the usual value of 10-20% under certain
circumstances.
 
Otten and Fuchs(39) reported considerable seasonal effects on the
occurrence of infertile mites in worker brood. Most mites stayed infertile
in the brood during winter, with a continuous increase in fertility
occurring during March/April, and from May to August values below 20% were
usually found; in autumn the fertility slowly decreased.
 
A significant seasonal effect on the fertility of mites was also noted by
Kulincevic et al(26) for brood samples in the period from the end of June
until the end of September, in Yugoslavia. The effect did not occur during
a repetition of the test in the following year.
 
Buchler(10) found a significant increase in mite infertility as a result of
using the trapcomb technique. In this method, the queen is trapped on one
comb for 8-9 days, then moved to other combs in three successive steps.
This results in a bee population composed of older bees in the first weeks
after the queen is released. Rosenkranz and Sturme(49) demonstrated by
systematic manipulation of the mites the dependence of mite fertility on
their feeding shortly before cell invasion and during the first hours after
cell capping. The infertility rate was clearly increased as a consequence
of a period of starvation or an elongated phoretic periodbefore brood
invasion. Mites reproduced normally when they were introduced into brood
cells immediately after cell capping but showed a strictly reduced
fertility when the introduction was delayed for some hours. In accordance
with these observations Fuchs(20) assumed that the observed effects of the
strain of the bees on the fertility of the mites were caused by differences
in the adult host bees and not by differences in the brood.
 
It is not known which components of the diet of the mites are responsible
for inducing oogenesis (egg production). The juvenile hormone theory, which
has been intensively discussed, is now refuted by the investigations of
Rosenkranz et al.(50). Possibly the quantity of haemolymph available at a
certain time may regulate reproduction, especially the initiation of
vitellogenesis (egg protein production) in  jacobsoni. The considerable
weight increase observed in the mites shortly after cell capping may
indicate this(48). As the haemolymph intake of mites is influenced by
certain characteristics of the bees such as grooming, defence and brood
removal behaviour, brood activity and age structure of the bee population
the fertility of the mites may prove to be an indirect tolerance
characteristic.
 
--============_-1379316406==_============
Content-Type: text/plain; charset="us-ascii"
 
                __________________________________________________________
Hans-Ulrich THOMAS. Beekeeper & collector of books about:
 
- bees and beekeeping
- ants (yes these small little buggers!)
- nature printing
 
e-mail: [log in to unmask]
CompuServe: 100045,2556
Fax: ++41 1 633 10 77
__________________________________________________________
 
--============_-1379316406==_============--