24th Int. Symp. “Animal Science Days”, Ptuj, Slovenia, Sept. 21st−23rd, 2016. Acta argiculturae Slovenica, Supplement 5, 143–147, Ljubljana 2016 COBISS: 1.08 Agris category code: L73 MICROSPORIDIAN NOSEMA SPP. AS A MODEL GASTROINTES- TINAL MICROORGANISM OF CARNIOLAN HONEY BEE (APIS MELLIFERA CARNICA, POLLMAN, 1879): ASPECTS OF SPORE COUNTING Ilja Gasan OSOJNIK ČRNIVEC 1, 2 Microsporidian Nosema spp. as a model gastrointestinal microorganism of Carniolan honey bee (Apis mellifera carnica, Pollman, 1879): Aspects of spore counting 1 Agricultural Institute of Slovenia, Animal Production Department, Hacquetova 17, SI-1000 Ljubljana, Slovenia 2 New address: University of Ljubljana, Biotechnical Faculty, Department of Animal Science, Groblje 3, SI-1230 Ljubljana, Slovenia, e-mail: gasan.osojnik@bf.uni-lj.si ABSTRACT At high invasion levels, nosemosis can act as an underlying cause of colony collapse disorder, especially in cases of simultaneous invasion of other parasites or/and in combination with virus infections. Rapid quantitative methods for Nosema spp. spore detection are therefore required for various research and diagnostic needs. In this study, the use of haemocytometer (standard method) and Coulter counter is compared. All work was performed with late summer forager bees of Carniolan honey bee subspecies, prepared as isolates and homogenates of the posterior section of the worker bee’s intestine. It can be concluded that the Coulter counter provides greater sample dilution flexibility at suf- ficient repeatability (RSD 0.01–0.1) in the cases of: (i) low concentrations of spores, as well as (ii) samples that required high dilutions (c ≥ 107 spores ml–1). However, Coulter counter use required additional data processing for obtaining precise concentration estimates at ~ 3 µm mean spore size. Nevertheless, the uncorrected, as-measured values provided concentration ranges which were found to be suitable for general diagnostic purposes. Key words: bees, Carniolan honey bee, microbiology, microorganisms, cell counting, haemocytometer, Coulter counter 1 INTRODUCTION Nosema disease, or nosemosis of honey bees is presently known to be caused by two species of micro- sporidian parasites, Nosema apis (Zander, 1909) and Nosema ceranae (Fries et al., 1996). Spores of the two Nosema species have similar ovoid morphologies. Spores of N. ceranae are slightly smaller, reportedly measuring 3.3–5.5 µm length and 2.3–3.0 µm width, whereas N. apis spore ranges are 4–6 µm and 2–4 µm, respectively (Fries et al., 1996; Higes et al., 2010). Both species invade the epithelial cells of ventriculus (midgut) and have similar life cycles, resulting in release of large number of spores and in severe cases, in consequential death of bees due to destruction of the intestine wall. At high invasion levels, nosemosis can act as an underlying cause of colony col- lapse disorder, especially in cases of simultaneous inva- sion of other parasites and/or in combination with virus infections (Bromenshenk et al., 2010). Due to continuous honey bee colony losses, stand- ing requirements for data collection on parasite pathol- ogy, omnipresence within the honey bee population and simplicity of observation, the microsporidian Nosema spp. was selected as a model gastrointestinal microorgan- ism for a study on spore counting aspects, which can be extended to cell counting applications. Routinely, a haemocytometer is used for the quanti- tative particle analysis of bee gut contents, in particularly for assessment or spores of Nosema spp. Simultaneous analyses of other microorganisms residing in the bee gut increase the counting time and typically require prelimi- nary staining procedures. In this study, an insight in re- liability of a traditional haemocytometer counting tech- nique for Nosema spp. spore assessment is provided and the method is compared to the performance of a portable Acta agriculturae Slovenica, Supplement 5 – 2016144 I. G. OSOJNIK ČRNIVEC suming spherical morphologies. The procedure is auto- mated and typical sample handling time ranged from 1 to 3 minutes. For spore counts, sensors with 40 µm ap- erture were used. Prior use, the apparatus was assessed for reliability with a reference suspension of polystyrene beads (d = 15 µm, cp = 2 × 10 5 beads ml–1) using 40 and 60 µm sensors (sensor measuring ranges within 3–17 and 6–36  µm, respectively). Reference measurements con- sistently showed high repeatability of results with both sensor types (relative standard deviation, RSD  <  0.06, c p  =  2.3  ×105  beads  ml−1). Cell counts in the desired uniform particle size range were obtained with Scepter Software 2.1 (Merck) cellular analysis platform. In order to extrapolate particle size concentration and distribu- tion below the detection range of the 40 µm sensor, peak fitting (Gaussian amplitude / peak function) was per- formed with Origin 8.1 (OriginLab Corporation) data analysis software. In parallel to the novel automated counting meth- od, counting was performed using a Bürker chamber (12 × 12 fields) under Leica DM 2500 light microscope at 400 × magnification. Typical sample handling time for this method ranged from 10 to 20 minutes. 3 RESULTS AND DISCUSSION Initially, the standardised method was employed for spore counting for an array of samples, which were exhibiting a wide range of Nosema spp. levels of infec- tion (from 105–107  spores  ml–1), as shown in Table 1. Preliminary tests showed high variability (P  <  0.001, Coulter cell counter for examination of a complex sam- pling matrix of bee gut homogenate. 2 MATERIAL AND METHODS All work was performed with the subspecies Car- niolan honey bee (Apis mellifera carnica, Pollman, 1879). Forager bees of sufficient quantity (approximately 100  bees/hive) were sampled during July and August 2015 from hives located at various field stations operat- ing within the Slovenian national bee monitoring net- work. Bees were sampled at the hive entrance, in order to avoid underestimation of spore count via sampling of recently-hatched bees which are spore-free. Sample preparation (isolation and homogenisation of the posterior section of the worker bee’s intestne) and haemocytometer counting were performed following the standard methods, described by Fries et al. (2013) and Human et al. (2013), respectively. For all samples analysed, phosphate-buffer saline was used instead of DI water to ensure conductivity of the samples, which was required for particle size determination by means of the Counter principle. A portable Coulter counter was used to count sam- ples (Merck, model Scepter 2.0 Cell counter). The basic operational principle of the device consists of pumping a conductive suspension (sample) through a microchan- nel and determining the particle size of passing solids from measured changes in impedance, which are directly proportional to the volume of the particle traversing the orifice. The particle diameter is then calculated as- Replicate N° Hive I Hive II Hive III Hive IV Hive V Hive VI Hive VII 1 0.45 1.7 3.5 4.3 11.3 20.8 27.8 2 0.5 1.8 3.1 3.8 10.8 20.8 28.4 3 0.7 1.7 3.6 5.7 10.7 19.25 27.2 4 0.6 1.5 3.7 5.4 10.6 19.55 27.9 5 0.65 1.8 3.6 3.9 9.2 20.3 27.1 6 0.6 2.1 3.1 4.1 9.05 20.1 27.3 7 0.6 2.1 2.9 5.1 10.1 19.9 27.1 8 0.45 2.3 4.5 4.0 10.5 19.1 27.5 9 0.55 1.9 3.7 3.7 9.7 20.0 27.1 10 0.7 1.4 3.6 5.2 10.2 20.1 27.3 Average ± SD 0.58 ± 0.09 1.82 ± 0.28 3.49 ± 0.44 4.50 ± 0.75 10.22 ± 0.72 19.99 ± 0.57 27.45 ± 0.44 RSD 0.16 0.15 0.13 0.17 0.07 0.03 0.02 units of concentration: × 106 spores ml−1 Table 1: Spore concentration, average ± standard deviation and haemocytometer counting repeatability (RSD) for seven hives with different Nosema spp. levels of infection Acta agriculturae Slovenica, Supplement 5 – 2016 145 MICROSPORIDIAN NOSEMA SPP. AS A MODEL ... (APIS MELLIFERA CARNICA, POLLMAN, 1879): ASPECTS OF SPORE COUNTING et al., 2015; Emsen et al., 2016) and monitoring stud- ies report N. ceranae domination in population share, as well as in prevalence. As the need of differentiation amongst the two Nosema species has decreased in im- portance for general diagnostic purposes, spores in this study were regarded as one population with the Coulter counter device. Nevertheless, the counting method does allow for some basic recognition, taking in mind the particle diameters provided by the Coulter counter are represented as sphere diameters, whereas Nosema spores exhibit prolate ovoid morphologies. Assuming 50 % N. apis and 50 % N. ceranae population, prolate ovoids with 3.3–6 µm in length and 2.3–4.0 µm in width would have been observed in the sample, corresponding to an meas- ured average sphere diameter of 3.6 µm. However, as it is evident from Figure 1, the average sphere diameter has been extrapolated to fall below 3 µm (conforming to ovoid shapes with roughly 4 µm in length and 2.5 µm in width), indicating an expected dominance of N. ceranae spores and simultaneously confirming the coexistence of N. apis spores in the examined samples. The employed portable Coulter counter is capable of detection of particles, which are larger than 3 µm and at these limitations the device operates within the 5 × 104– 1.5 × 106 particle ml–1 concentration range. Using the ap- proach presented in Figure 1, it is possible to extrapolate the measured values to obtain total spore counts and dimensions. In comparison to counts obtained with the standard spore counting procedure, the adjusted values were analogous (90.4 % match). Furthermore, uncorrected, as-measured values ( c 1–3 = 1.22 × 107 spore ml–1) accounted for approximate- 0.2–0.5  RSD) of spore counts from bees sampled from the same hive for separately prepared and thoroughly ho- mogenized suspensions (n = 20 / 3 suspensions / 4 hives). Therefore in order to increase the level of repeatability, all subsequent analyses were performed with the same pri- mary material per hive. Furthermore, ten replicates of the same sample were evaluated by counting either twelve or five fields in the counting chamber and the results were found to be comparable (an average count ratio for 12/5 fields of 0.99/1 was obtained). From there on, five fields were used for counting (1/1; 1/12; 12/1; 12/12; 6/6), at it is mostly done in internal routine counting appli- cations. Using this approach it is evident that at lower concentrations (up to 4.5  ×  106  spores  ml–1), where 1–20  spores can be counted for each field, the coeffi- cient of variation between replicates was relatively high (Table 1). To increase the repeatability of these results, more fields could have been counted (consequently in- creasing the time of analysis). At higher concentrations (20–140 spores per field), repeatability was high and the coefficient of variation was well below 10 %. For the purpose of examination of the counting method based on the Coulter principle, we selected a sample exhibiting high Nosema spp. level of infection. For each sample, two series of triplicates were measured and respective particle-size distributions were extrapo- lated below the lower limit of measurement to the full ex- pected diameter range (as shown in Figure 1 for the case of a 100 × diluted sample). Within samples, good repeat- ability of measurements, as well as a good fit to normal distribution were observed. In the recent years, several researchers (e.g. Csáki 2.5 3.0 3.5 4.0 4.5 0 300 600 900 1200 0 300 600 900 1200 0 300 600 900 1200 2.5 3.0 3.5 4.0 4.5 Fd = 100 c3= 2.53 × 10 7 spore ml-1 Diameter (µm) triplicate 3 Fd = 100 c2= 2.46 × 10 7 spore ml-1 Co un t triplicate 2 Fd = 100 c1= 2.46 × 10 7 spore ml-1 R2= 96.5 % R2= 96.8 % triplicate 1 R2= 96.6 % Figure 1: Particle diameter and count in 100 × diluted sample of bee gut homogenates. Difference between triplicates not statistically significant (P > 0.05). Acta agriculturae Slovenica, Supplement 5 – 2016146 I. G. OSOJNIK ČRNIVEC ly two fold lower concentrations, yet they remained in the same order of magnitude in comparison to the refer- ence value which was determined by the standard count- ing method. As both counting techniques require sample dilu- tion at high spore concentration, a dilution series was prepared to study possible effects. The haemocytomet- ric spore count of the undiluted sample was set as the reference value (Figure  2, denoted as dashed line). For the case of undiluted sample, high repeatability of counts was obtained at  >  100  spores for each counting field. However, analyses of larger sets of samples having such a concentration range require a lot of time and focused concentration of the microscope operator. It is therefore not uncommon to count Nosema spp. spores in 10 × di- luted samples, where reliable counts can be achieved, as Figure 2 shows. At higher dilutions (100  and  1000  ×) beyond the recommended spore concentration of 5–50  spores per field (Human et al. 2013), the spore numbers became in- creasingly overestimated (up to 180 %). In the case of the standard method, increasing the dilution did not affect the coefficient of variation (in order of dilution: 13  %, 9 %, and 7 %). Using the Coulter counter adjusted values at 10  × sample dilution (operating slightly above the recom- mended concentration) 30  % underestimation of the spore concentration was observed. In contrast to stand- ard counting method, Coulter counter provided relevant values (up to 93 % of the reference value) in subsequent dilutions. In this case, the coefficient of variation was in- creased from 1 % at 10 × dilution to 10 % at 1000 × dilu- tion. Similarly to previous observations, uncorrected val- ues corresponded to 40–77 % (2.5–1.3 fold) of the adjust- ed concentration at 10–1000 × dilution, respectively. This apparent discrepancy can be explained by the fact that as the number of the spores in the suspension decreases due to dilution, the size fraction of the spores which is numerically most represented in the initial (undiluted) population, gains an even greater effect on the overall population estimate. Furthermore, adjusted and unad- justed values exhibited the same coefficients of variation, indicating that the extrapolation has been performed correctly as it preserves the initial characteristics of the measurements within the expanded set of data. 4 CONCLUSIONS Nosema spp. was selected as a model gastrointesti- nal microorganism for a study on spore counting aspects, which can be extended to cell counting applications. Two methods, (i) standard haemocytometer counting tech- nique and (ii) the use of a Coulter counter are compared. It can be concluded, that the Coulter counter pro- vides greater dilution flexibility and exhibited sufficient accompanying repeatability (RSD 0.01–0.1) for the anal- ysis of: (i) spores at low concentrations, or (ii) samples at high dilutions. Counting spores at low and medium concentrations (6 × 105–4 × 106 spore ml–1) using the standard method resulted in borderline repeatability (RSD up to 0.17), indicating that prolonged sample observation was re- quired, as an increased number of fields would have had of been visually examined. 0 10 100 1000 0 1 2 3 4 5 Co nc en tra tio n (× 1 07 s po re s m l-1 ) Dilution factor Standard method Corrected Coulter counter Coulter counter re fe re nc e v al ue Figure 2: Spore concentration average at different sample dilutions for each counting method (bars denote standard deviation) Acta agriculturae Slovenica, Supplement 5 – 2016 147 MICROSPORIDIAN NOSEMA SPP. AS A MODEL ... (APIS MELLIFERA CARNICA, POLLMAN, 1879): ASPECTS OF SPORE COUNTING High sample dilutions may be required, e.g. when analysing concentrations ≥ 107  spores  ml–1. At higher concentrations (3  ×  107  spores  ml–1), the repeatability of the traditional technique was high (RSD < 0.05), and the sample handling time was longer as a result of higher spore counts. In the case of unsuitably high dilutions, the measured concentrations were imprecise. The downside of using Coulter counters is that ad- ditional data processing techniques may be required. It is furthermore important to note that the entire spore concentration range, and simultaneous presence of other microorganisms (e.g. yeasts with typical dp  =  3–4  μm) were not evaluated in this study and that these effects on Nosema spore counting should be considered further. Nevertheless, in comparison to reference value, similar spore concentrations were observed when using adjusted Coulter counter counts. At 100 and 1000 × dilution, ad- justed values consistently corresponded to roughly 90 % of the set reference value, showing that this approach can be used for obtaining precise total counts. To allow for time efficient analyses, such devices should allow for the integration of advanced data processing techniques. Uncorrected values show two to three fold lower concentrations, yet they remain in the same concen- tration range than the reference value (e.g. 107 for the presented case) and could therefore be used for routine diagnostic purposes, where the information of the exact concentration is not required. 5 ACKNOWLEDGEMENTS The author gratefully acknowledges various techni- cal help, sample selection assistance and diligent work with biological material from David Kozamernik, Mitja Nakrst and Mateja Soklič (Agricultural institute of Slo- venia) as well as from high school students Teja Rus and Anja Seliškar (Biotechnical Educational Centre Ljublja- na). 6 REFERENCES Bromenshenk, J.J., Henderson, C.B., Wick, C.H., Stanford, M.F., Zulich, A.W., Jabbour, R.E., Deshapande S.V., et al. (2010). Iridovirus and Microsporidian Linked to Honey Bee Colo- ny Decline. PloS ONE, 5(10), e13181. Retrieved from http:// dx.doi.org/10.1371/journal.pone.0013181. Csáki T., Heltai M., Markolt F., Kovács B., Békési L., Ladányi M., Péntek-Zakar E., et al. (2015). Permanent prevalence of Nosema ceranae in honey bees (Apis mellifera) in Hungary. Acta Veterinaria Hungarica, 63(3), 358–69. Emsen, B., Guzman-Novoa E., Hamiduzzaman M.M., Ec- cles L., Lacey B., Ruiz-Pérez R.A., Nasr M. (2016). Higher prevalence and levels of Nosema ceranae than Nosema apis infections in Canadian honey bee colonies. Parasitology Re- search 115(1), 175–81. Fries, I., Feng, F., Silva, A.D., Slemenda, S.B., Pieniazek, N.J. (1996). Nosema ceranae n. sp. (Microspora, Nosematidae), morphological and molecular characterization of a mi- crosporidian parasite of the Asian honey bee Apis cerana (Hymenoptera, Apidae). European Journal of Protistology, 32, 356–365. Fries, I., Chauzat, M.P., Chen, Y.P., Doublet, V., Genersch, E., Gisder, S., Higes, M., et al. (2013). Standard methods for nosema research. In V. Dietemann, J.D. Ellis, P. Neumann (Eds.), The COLOSS BEEBOOK: Volume II: Standard meth- ods for Apis mellifera pest and pathogen research. Journal of Apicultural Research 52(1). Retrieved from http://dx.doi. org/ 10.3896/IBRA.1.52.1.14. Higes, M., Martín-Hernández, R., Meana, A. (2010). Nosema ceranae in Europe, an emergent type C nosemosis. Apidolo- gie, 41, 375–392. Human, H., Brodschneider, R., Dietemann, V., Dively, G., Ellis, J., Forsgren, E., Fries I., et al. (2013). Miscellaneous stand- ard methods for Apis mellifera research. In V. Dietemann, J.D. Ellis, P Neumann (Eds.) The COLOSS BEEBOOK, Vol- ume I: standard methods for Apis mellifera research. Journal of Apicultural Research 52(4). Retrieved from http://dx.doi. org/10.3896/IBRA.1.52.4.10.