A novel system for maintaining Varroa destructor mites on artificial diets and its application for studying mites as a vector for honey bee viruses

The mite Varroa destructor is one of the most destructive parasites of the honey bee (Apis mellifera) and the primary cause of colony collapse in most regions of the world. These mites cause serious injury to their hosts, especially during the larval and pupal stages, and serve as the vector for several viruses, which affect honey bee health causing colony death. Attempts by beekeepers to control these mites have yielded limited success. The inability to rear populations of mites in vitro that excludes contact with their honey bee hosts has stymied research of Varroa biology. Previous attempts to rear and/or maintain Varroa mites in vitro by feeding them on artificial diets have had limited success. Several methods were plagued by mechanical failures including leaking membranes and, thus far, none have been widely adopted. Here we report a robust system for maintaining Varroa mites that includes an artificial diet, which does not contain honey bee tissue-derived components, thus making it particularly valuable in studying mite vectoring of honey bee viruses. With our system we demonstrated for the first time that Varroa mites maintained on an artificial diet supplemented with the particles of honey bee viruses, cDNA clone-derived genetically tagged Varroa destructor virus-1 and wild-type Deformed wing virus, can acquire and later transmit these viruses to recipient honey bee pupae. Along with providing an opportunity to study parasites and pathogens in the absence of honey bee hosts, this in vitro system for Varroa mite maintenance is both scalable and consistent. These features can be used to better understand mite nutritional needs, metabolic activity, responses to chemicals and other biological functions.


Introduction
The ectoparasitic mite, Varroa destructor (Anderson and Trueman) (Acari: Varroidae), is believed to be the major factor in the widespread collapse of honey bee colonies in North America as well as many parts of Europe and Asia [1]. Attempts by beekeepers to control There have been several attempts to rear populations of Varroa mites in vitro using artificial diets. Perhaps the first successful attempt to feed Varroa on an artificial diet was reported by Bruce et al. [4], who stretched a thin (~10 µm) parafilm membrane containing a dietary medium over modified queen cells, which are used for queen-rearing. Mites were observed to survive for up to 120 h, and many mites also laid eggs. However, this system was plagued by evaporation and contamination of the diet. Further modifications to this system using queen rearing cells covered with parafilm minimized leakage and were successful for maintaining a number of the much smaller tracheal mite, Acarapis woodi [7]. Later, Talbart et al. [5] produced synthetic chitosan membranes for holding an artificial diet with which to rear Varroa mites but this system also had limited success and has not yet been adopted by other researchers. This issue is likely due to earlier systems that required relatively complex components; only a few devices could be manufactured at a time, with the result that tests were done with very few mites [5,8]. Attempts to reproduce these published methods revealed many setbacks (authors' unpubl. data), especially related to membrane leakage, which drowned the mites. These physical limitations made it difficult, if not impossible to carry out repeatable tests for research. Because a system for rearing populations of Varroa mites in vitro remains elusive, a system is still needed for maintaining sufficient numbers of mites for research purposes.
The success of maintaining Varroa mites on artificial diets for research purposes depends upon three major issues: 1) availability of numerous healthy adult female mites; 2) the composition of the artificial diet; and 3) a device and membrane system for housing mites and containing the diet, respectively. Unfortunately, sufficient numbers of Varroa mites for research remain seasonally restricted. Nonetheless, for further progress in maintaining large numbers of mites in vitro with artificial diets, the first task is to create a simple but reliable system that can allow mites housed in a device to have access to a diet that does not leak, evaporate or is prone to contamination [9]. The second task is to provide a membrane thin enough (10 to 15 µm), to allow Varroa mites to reach the diet with their mouthparts, which are very short [10]. Finally, methods must contain a sufficient volume of liquid to feed mites for many days or even weeks without compromising the quality and/or integrity of the diet.
In this report, we describe a robust system including a membrane based on that of Avila and colleagues [11] for maintaining large numbers of Varroa mites numbers on completely hostfree diets. Further, we validate the system's utility in an experiment of Varroa mite vectoring of honey bee viruses, specifically, the novel cDNA clone-derived Varroa destructor virus-1 (VDV1 or DWV-B).
To minimize contamination, fresh diet was prepared in a fume hood and stored frozen (-20 o C) in small aliquots. No antibiotics or antifungal agents were included to avoid compromising microbiota that may provide essential nutrients for mite nutrition, since species of Diplorickettsia, Arsenophonus, Morganella, Spiroplasma, Enterococcus, and Pseudomonas have been reported to inhabit Varroa mites [12].
Varroa mites. The sugar roll method [13] was used to collect phoretic mites from honey bee colonies that had not received miticidal applications. Following collection, and after rinsing mites with tap water , they were placed on petri dishes lined with dry tissue paper. Mites were checked for activity prior to experimental selection, and only active mites used.
Device and membrane. Snap-cap polypropylene 1.5 mL microcentrifuge tubes were chosen for constructing the device housing the mites (Fig 1). The tubes were cut crosswise at the point that allowed a 1 cm diameter opening. Cut surfaces were smoothed with sandpaper to avoid damaging the parafilm (American National Can Company, NY) membranes that would later be attached. Prior to using the devices, while wearing sterile nitrile gloves, containers were transferred to a laminar flow hood, disinfected for 2 min with 1.0 % bleach, followed directly by 6 light. Following disinfection and while still in the laminar flow hood, each device was covered with a thin sheet of sterilized parafilm stretched to 16.6±5.86 µm and wrapped over the open end of the device. Membrane thickness was checked with a Marathon digital micrometer (Marathon Corp., Canada). Diet solution (10 µL) was pipetted onto the center of the parafilm membrane, after which the membrane plus diet solution was covered with a second layer of sterile parafilm, forming the parafilm sachet [11,14,15]. The snap-cap lid on the other side of the device was opened to insert a female Varroa mite. Before closing, the lid was punctured with a sterile no. 1 stainless steel pin to form a minute hole (approximately 0.5 mm) to allow air to escape so the parafilm sachet would not rupture when the lid was snapped shut (Fig 1). The completed devices were placed in a dark incubator (Thermofisher Scientific, San Jose, CA) set at 32.1 ± 0.3 °C, and having 82.2 ± 1.3 % relative humidity.
leakage of the diet, or whether air bubbles or other obstructions blocked access to the diet.
Finally, the diet was inspected for contamination and evaporation as described previously [9].
Design of infectious cDNA clone of Varroa destructor virus-1 (VDV1 ) and production of clone-derived inocula. We designed a full-length infectious cDNA clone of a Californian isolate of VDV1, GenBank Accession number MN249174 (S1 Text, S2 Text Transmission of the acquired viruses to honey bee pupae. Following the 36h acquisition period the surviving mites were placed individually on a pink-eyed honey bee pupa housed gelatin capsules and placed in a dark incubator (see above) for 48h to allow for the transmission of the virus to the pupae. Pupae were then incubated for an additional 72h to allow virus replication (Fig 2). To determine whether viral transmission occurred, total RNA was extracted from each of pupae from the experimental (n=15) and control (n=9) treatment groups, then extracts subject to RT-qPCR to quantify copy numbers of VDV1 and DWV genomic RNA and assess transcription levels of honey bee actin by RT-qPCR using the primers described in S1

Results
Device, membrane and mite survival. The feeding device (Fig 1) allowed Varroa mites to feed on the artificial diet without honey bee-derived components through a ~10 µm parafilm membrane. No evidence of diet leakage, evaporation or contamination, or other mechanical failures precluding mite's access to the diet was recorded. We compared survival of mites from the three trials having devices with parafilm sachets containing the V-BRL artificial diet against survival of mites from both the positive and negative control groups (Fig 3; S2 Table). In contrast to diet fed-mites, most mites survived up to 25 days while feeding on bee pupae. Mite survival declined precipitously after day 10. The average survival time while feeding on pupae was 9.42 days (Fig 3, green dotted line). We found that mites began dying within 6h after confinement in the artificial feeding device without diet. Of the 32 mites held in the container without food (negative control group), all but two were dead within one day; the remaining two died by day two (Fig 3, Table) and the finding of FITC-labeled fluorescent beads in excreta of the mites fed on the bead-containing diet ( Fig 4B) but not in excreta for those feeding diet without fluorescent beads (Fig 4 C). The daily average deposition of excreta was 2.55 ± 0.52 deposits per mite. Statistical analysis of the excreta data for each of the four days showed that the differences in excreta per live mite per day were not statistically significant (one-way ANOVA, P = 0.3797, Fig 4 D). C. D.

Evidence of acquisition of viruses by mites and transmission to honey bee pupae.
There was no statistically significant difference in the survival of mites fed on the diet containing the viruses versus mites fed on the control diet over a 72-h study period (S1 Fig,   S4 Table, S5 Table). While there was no significant difference between the levels of honey bee actin mRNA in the treatment and control groups (P=0.1395, S6 Table), the levels of DWV and VDV1 were statistically significantly higher in pupae exposed to virus-fed mites compared to those in the control group (P<0.01 and P<0.001 respectively; Fig 5A). One-way ANOVA of the virus accumulation in the control and treatment groups were highly  (Fig 5A). This result could be explained by a 1000-fold difference between the levels of DWV and VDV1 in the diets containing 3.2 × 10 4 and 3.3 × 10 7 genome equivalent per µL respectively.
We confirmed that VDV1 derived from the artificial diet replicated in the treatment group pupae to high levels based on the diagnostic AsiSI restriction site ( Fig 5B). This restriction site was not present in the wild-type isolates of VDV1 or DWV (Fig 5B, lane "wt"). Further evidence showing that the VDV1 transmitted to pupae was the cDNA clone VDV1 acquired from the artificial diet comes from digesting the RT-PCR fragment corresponding to the 5' terminal 1200 nt with AsiSI restriction enzyme. All seven of the recipient pupae that developed high virus levels showed that the AsiSI site was present in VDV1 ( Fig 5B). Notably, the sample isolated from the 27 pupae of the treatment group contained high levels of both DWV and VDV1 (  fragments corresponding to the 5'-terminal region of DWV and VDV1 RNA genomes were amplified with the primers specific to both DWV and VDV1 using RNA extracts from the experimental pupae with the virus levels exceeding 10 9 genome copies from the "Virus extract" group (S6 Table), and from the wild-type VDV1 and DWV-infected pupae, "wt".
The undigested 1250 nt fragments (left) and AsiSI-digested (right). Expected fragment sizes, undigested (black arrow) and AsiSI-digested (red arrows), are shown on the left, DNA ladder sizes are shown on the right.

Discussion
Here we describe a simple, but robust, system including an artificial diet, which was free of honey bee-derived components, that successfully maintained Varroa mites for a study of their vectoring of honey bee viruses to host pupae. In contrast to the previous artificial systems for rearing/maintaining Varroa mites, we used the accumulation of mite excreta and the presence of the fluorescent microbeads acquired from the diet in the excreta to confirm that mites had successfully fed through the membrane on the artificial diet. A significant challenge in this experiment was potential presence of viruses in both the recipient honey bee pupae and the Varroa mites sourced from the honey bee colonies maintained in apiaries known to harbor viruses. Therefore, it was also necessary to distinguish between the virus introduced through feeding on the diet and those that may have already been present either in the mites or in the recipient honey bee pupae. This challenge was resolved by using cDNA clone-derived VDV1 (also known as DWV-B) that was tagged with a rare AsiSI restriction enzyme site to differentiate it from the wild type ( Fig 5B). The absence of pupae with high DWV and VDV1 levels (above 10 9 genome equivalents) in the control group (Fig 5A), i.e. by mites exposed to 36 h feeding on virus-free diet, was in agreement with the recent report by which indicated that DWV is vectored by Varroa mites via a non-persistent mechanical route [17].
The virus acquisition and subsequent transmission experiment presented in this report provides a "proof of concept" that validates the utility of the in vitro system for maintaining Supporting information S1 Figure. Mite survival analysis of the virus acquisition and transmission experiment. S1 Table. Primers and the synthetic gene used in this study. Supporting information S1 Figure. Mite survival analysis of the virus acquisition and transmission experiment.
RT-PCR amplification of the VDV1 and DWV IRES sequence, complementary to positions 1236-1252 of VDV1 genomic RNA.   M  i  t  e  e  x  c  r  e  t  a  d  a  t  a  .   E  v  a  l  u  a  t  i  o  n  (  D  a  y  )   D  i  e  t  c  h  a  m  b  e  r  n  u  m  b  e  r   V  a  r  r  o  a  n  u  m  b  e  r   E  x  c  r  e  t  a  c  o  u  n  t  E  v  a  l  u  a  t  i  o  n   F  e  c  e  s  p  e  r  d  a  y  E  v  a  l  a  u  t  i  o  n   A  l  i  v  e  o  n  t  h  e  d  a  y  o  f  o  b  s  e  r  v  a  t  i  o  n   D  e  a  d  m  i  t  e  s  i  n  c  h  a  m  b  e  r   1  1  2  9  9  1  1  1  2  2  6  6  1  1  1  3  2  1  1  2  0  1  4  2  0  0  0  2  1  5  0  0  0  0  0  1  6  0  0  0  0  0  1  7  3  1  6  1  6  2  1  1  8  3  0  0  2  1  1  9  4  1  1  1  1  1  3  1  1  0  4  1  6  1  6  3  1  1  1  1  4  1  1  1  3  1  1  2  4  0  0  0  4  1  1  3  4  3  1  3  1  3  1 T  r  a  n  s  m  i  s  s  i  o  n  -P  u  p  a  e   3  0  h  o  u  r  s  F  i  r  s  t  d  a  y  =  2  4  h  o  u  r  s  S  e  c  o  n  d  d  a  y  =  4  8  h  o  u  r  s  T  r  e  a  t  m  e  n  t  D  e  v  i  c  e  s  a  m  p  l  e  s   V  a  r  r  o  a  m  i  t  e  s  p  e  r  d  e  v  i  s  e   T  o  t  a  l  v  a  r  r  o  a   A  c  q  u  i  s  i  t  i  o  n  -D  i  e  t  S  5  T  a  b  l  e  .  V  i  r  u  s  a  c  q  u  i  s  i  t  i  o  n  a  n  d  t  r  a  n  s  m  i  s  s  i  o  n  e  x  p  e  r  i  m  e  n  t  :  M  i  t  e  s  u  r  v  i  v  a  l  a  n  a  l  y  s  i  s  .   T  r  e  a  t  m  e  n  t  H  o  u  r  s  S  u  r T  a  b  l  e  .  V  i  r  u  s  a  c  q  u  i  s  i  t  i  o  n  a  n  d  t  r  a  n  s  m  i  s  s  i  o  n  e  x  p  e  r  i  m  e  n  t  ,  R  T  -q  P  C  R  q  u  a  n  t  i  f  i  c  a  t  i  o  n  o  f  V  D  V  1  ,  D  W  V  a  n  d  h  o  n  e  y  b  e  e  a  c  t  i  n  .   T  r  e  a  t  m  e  n  t  P  u  p  a  I  D  q  P  C  R  p  r  o  b  e  s  C  t  (  u  n  d  e  t  e  c  t  a  b  l  e  C  t  =  5  0  )   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  1  H  o  n  e  y  b  e  e  a  c  t  i  n  2  3  .  0  5  5  4  0  5  4  1   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  2  H  o  n  e  y  b  e  e  a  c  t  i  n  2  2  .  5  4  1  8  7  1  5   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  3  H  o  n  e  y  b  e  e  a  c  t  i  n  2  2  .  2  8  1  1  6  8  0  5   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  4  H  o  n  e  y  b  e  e  a  c  t  i  n  2  4  .  0  5  9  8  6  6  9  8   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  5  H  o  n  e  y  b  e  e  a  c  t  i  n  2  2  .  4  7  1  9  4  8  6  5   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  6  H  o  n  e  y  b  e  e  a  c  t  i  n  2  3  .  1  7  1  0  4  3  9  6   B  u  f  f  e  r  c  o  t  r  o  l  ,  P  B  S  7  H  o  n  e  y  b  e  e  a  c  t  i  n  2 A  s  e  -f  r  e  e  T  u  r  b  o  D  N  A  s  e  (  A  m  b  i  o  n  )  ,  t  h  e  f  u  l  l  -l  e  n  g  t  h  V  D  V  1  t  r  a  n  s  c  r  i  p  t  s  w  e  r  e  p  u  r  i  f  i  e  d  b  y   R  N  e  a  s  y  c  o  l  u  m  n  (  Q  i  a  g  e  n  )  .   T  o  p  r  o  d  u  c  e  c  l  o  n  e  -d  e  r  i  v  e  d  V  D  V  1  i  n  o  c  u  l