.4...4... . .....p. .. ....44. .....HU.4.4.4 ...4. . 44. . 4......4... ......2... ... ...... 4.4. ...,4 . .. ...). .... Z .... 5;}... r . ..., ......4 3...... ..U ... .44. 37:7. l .4 4. ......4.4 4 .4 4 ......4 4 \. (4..........44:..4..4 44.5.4.4... 4.44....44..4.4.. 4.. .4 ... rt. ...4 ......4 r 14. 4........I.... ........H..........;.... ........_.....y -4.-. 5.4.1.1.. .... ......A..5.r.......4.. 4. ......4 ......44 IEStS I)ate 0—7 639 MIICHGAN IIII II I II IIIIIIIIIIIIIIIIIIIIIIII”IIIIIIIIII This is to certify that the thesis entitled Movement of Blueberry leaf mottle virus (BBLMV) within and between cultivated and wild Vaccinium presented by Claudio Roberto Sandoval Briones has been accepted towards fulfillment of the requirements for Botany Master of Science degree in Plant yPati-ll'iology Major professor /{/¢/71-—- MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michtxan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATEDUE DATE DUE MSU Is An Affirmative Action/Equal Opportunity Institution czwlmmducunS-n.‘ MOVEMENT OF BLUEBERRY LEAF MOTTLE VIRUS (BBLMV) WITHIN AND BETWEEN CULTIVATED AND WILD Vacciniumspp. by Claudio Roberto Sandoval Briones A THESIS Submitted to . Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Botany and Plant Pathology 1992 ABSTRACT MOVEMENT OF BLUEBERRY LEAF MOTTLE VIRUS (BBLMV) WITHIN AND BETWEEN CULTIVATED AND WILD Vacc1n1umspp. BY Claudio Roberto Sandoval Briones The in situ localization of BBLMV coat protein in blueberry pollen grains and anthers using inmunogold labeling revealed gold particles over the intine and exine of infected pollen grains and the anther membranes. Experiments were developed to determine vertical and horizontal transmission via pollen. The results demonstrated that fruits and mother plants became infected when pollinated with BBLMV-infected pollen. This opens the possibility that infected wild blueberry bushes in wooded areas surrounding commercial plantings are acting as a source of infection. To test this, a field survey was performed. BBLMV was present in Vaccinium corymbosum, V. myrtilloides and V. angustifolium growing in the wooded areas. Two patterns of virus distribution were found; the virus was moving from the wild bushes to the commercial blueberries and vice-versa. This was related in some cases to differences in nectar production and sugar concentration among wild and commercial bushes. Also, BBLMV was able to move via pollen long distances. Infected pollen grains were detected in pollen traps located up to 2 miles from the source field. In Dedication to Ma. Eugenia, Ma. Ignacia and Cristobal ACKNOWLEDGEMENTS I wish to thank Dr Donald Ramsdell for his advice, helpful suggestions, support and friendship during these two years. I would also like to thank the members of my graduate committee Drs. James Hancock and Karen Klomparens for their encouragment and constant advice. Special thanks to my friend Jerri Gillett for always being there, giving me not only technical assistance but also encouragment and support. Many thanks to all the people and friends who for one reason or another helped me along with this program. Thanks to the Michigan Blueberry Growers Association for their monetary support and help. Finally, I would like to thank my wife, Ma. Eugenia and my children, Ma. Ignacia and Cristobal for being always at my side during these years of hard work. This achievement is not only mine but also theirs. iv TABLE OF CONTENTS LIST OF TABLES..................... ...... .... ........ LIST OF FIGURES........ ..... . ........ .... ............ INTRODUCTION.......... ..... .... ...... .... ....... _..... LITERATURE REVIEW General Background.... ..... ....... ....... ......... Transmission...... ...... .... ......... ...... ....... Pollinization...... ..... ...... .................... Seed and pollen transmission............... ....... Virus movement. ...... ..... .......... . ...... .. ..... Virus localization.......................... ...... Techniques for virus localization..... .......... .. METHODS AND MATERIALS In situ localization of viral gene products in infected pollen grains and anthers Pollen and anthers.... ....... .... .......... .... Fixing and embedding.............. .......... ... Trimming and sectioning .............. . ......... Etching ........... .. ......... ...... ...... . ..... Immunogold labeling .......... ..... ...... . ...... Transmission Electron microscopy............_... Determination of Vertical and Horizontal transm1551on of BBLMV via pollen in blueberry Experiment.. .................. .. ............... Vertical transmission..... ..................... Horizontal transmission... ........ . ......... ... Field survey Samples........ ........ .. ..... ........ ...... ... Virus testing ........... . ...... .... ............ Taxonomic classification of Wild blueberry bushes............. ..... . ....... Nectar measurement........... ....... . .......... 40 43 48 51 Spread of BBLMV infected pollen Locations............ ......... ... ....... ....... 52 Pollen traps................................... 55 Pollen analysis................ ......... ....... 55 RESULTS In situ localization of viral coat protein in infected pollen grains and anthers Immunogold labeling localization...... ..... .... 59 Immunogold labeling technique........... ....... 59 Determination of Vertical and Horizontal transmission of BBLMV via pollen in blueberry Vertical transmission ..... ......... ......... ... 68 Horizontal transmission....... ..... ...... ...... 80 Field survey BBLMV-infected fields ..... . ........... .. ...... . 84 Healthy fields.......... ............... .... 109 Taxonomic classification of wild bushes ........ 119 Nectar production and sugar concentration...... 162 Spread of BBLMV-infected pollen ....... ............ 162 DISCUSSIONOCIOC0..........OID......IIOOOOOOOCOOOOOOIO 166 LIST OF REFERENCES. ........ ............ .............. 179 vi Table LIST OF TABLES Fruit clusters analyzed by ELISA from healthy, potted, 3 yr-old plants pollinated with BBLMV-infected pollen ..... ........ ....... .... Number of fruit clusters infected with BBLMV on healthy potted 3 yr-old bushes pollinated with BBLMV-infected pollen, at different harvest dates ...... ................. Summary of results from pollinization of healthy, 3 yr—old potted blueberry bushes with BBLMV—infected pollen.... ........ . ...... Percentage of BBLMV-infected clusters from five different blueberry cultivars pollinated with BBLMV-infected pollen ................... Analysis of roots, blossoms and leaves from 18 healthy, potted 4 yr— —old plants pollinated with BBLMV-infected pollen in 1991. .......... Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Grand Junction, Van Buren county (Holmquist Farm)...... ......... ... ........... Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Breedsville, Van Buren county (Crawford Farm). Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Grand Haven-1, Ottawa county (Zelinka Farm)... Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV— infected bushes. Grand Haven—2, Ottawa county (Novak Farm).... vii Page 77 78 79 81 83 85 86 87 88 Table 10 11 12 13 14 15 16 17 18 19 Page Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Fruitport , Muskegon county (Dirkse Farm)...... 89 Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. North Muskegon, Muskegon county (Paul Bros. Farm)............................. 90 Summary of distribution of BBLMV-infected wild blueberry bushes at different distances from the borders of six commercial fields containing BBLMV-infected bushes.............. 104 Summary of survey data for BBLMV-infected wild blueberry plants in woods surrounding commercial fields containing BBLMV-infected bushes in WestCentral Mich1gan..... .................... 108 Summary of distribution of BBLMV—infected wild blueberry bushes in the woods at four different compass directions from the border of commercial fields containing BBLMV—infected bushes ...... 110 Percentage of BBLMV—infected wild blueberry bushes in woods surrounding a healthy commercial field. South Haven-1, Van Buren county (DeGrandchamp Farm, House field, west)..................... 111 Percentage of BBLMV—infected wild blueberry bushes in woods surrounding a healthy commercial field. South Haven-2, Van Buren county (DeGrandchamp Farm, House field, east)...... ............... 112 Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a healthy commercial field. South Haven-3, Van Buren county .............. 113 Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a healthy commercial field. South Haven- -4, Van Buren county (DeGrandchamp Farm, Springett south)... ............. ....... 114 Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a healthy commercial field. South Haven-5, Van Buren county (DeGrandchamp Farm, Springett north). ......... .. .......... . 115 viii Table 23 24 25 26 28 29 30 Page Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a healthy commercial field. South Haven-6, Van Buren county.............. 116 Summary of distribution BBLMV-infected wild blueberry bushes at different distances from the borders of six healthy blueberry commercial fields, in woods............................. 117 Survey summary for BBLMV-infected wild blueberry bushes in the woods surrounding six healthy blueberry commercial fields in WestCentral Michigan............... ........... ........... 118 Taxonomic classification of some wild blueberry bushes sampled at the Grand Junction location (Holmquist Farm, north quadrant of woods).... 120 Taxonomic classification of some wild blueberry bushes sampled at the Grand Junction location (Holmquist Farm, south quadrant of woods).... 121 Summary of taxonomic classification of some wild blueberry plants sampled at the Grand Junction location (Holmquist Farm, north and south quadrant of woods) ............ . .............. 122 Taxonomic classification of some wild blueberry bushes sampled at the Breedsville location (Crawford Farm, north quadrant of woods) ..... 123 Taxonomic classification of some wild blueberry bushes sampled at the Breedsville location (Crawford Farm, west quadrant of woods) ...... 124 Summary of taxonomic classification of some wild blueberry bushes sampled at the Breedsville location (Crawford Farm, north and west quadrant of woods).................................... 125 Taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-1 location (Zelinka Farm, west quadrant of woods) ....... 126 Taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-l location (Zelinka Farm, south quadrant of woods)...... 127 ix Table 31 32 33 34 35 36 37' 38 39 40 41 42 Page Taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-1 location (Zelinka Farm, north quadrant of woods)...... 128 Summary of taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-1 location (Zelinka Farm, west, south and north quadrant of woods)............... ..... ....... 129 Taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-2 location (Novak Farm, north quadrant of woods) ......... 130 Taxonomic classification of some wild blueberry bushes sampled at the Grand Haven~2 location (Novak Farm, east quadrant of woods).......... 131 Taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-2 location (Novak Farm, south quadrant of woods)........ 132 Summary of taxonomic classification of some wild blueberry bushes sampled at the Grand Haven-2 location (Novak Farm, north, east and south quadrant of woods).. ...... ........ ........... 133 Taxonomic classification of some wild blueberry bushes sampled at the Fruitport location (Dirkse Farm, west quadrant of woods) ........ . ..... .. 134 Taxonomic classification of some wild blueberry bushes sampled at the Fruitport location (Dirkse Farm, east quadrant of woods) ................ 135 Taxonomic classification of-some wild blueberry bushes sampled at the Fruitport location (Dirkse Farm, south quadrant of woods).. ........... ... Summary of taxonomic classification of some wild blueberry plants sampled at the Fruitport (Dirkse Farm, west, east and south quadrant of woods)....... .................. . ............. 137 Taxonomic classification of some wild blueberry bushes sampled at the North Muskegon location (Paul Bros. Farm,east quadrant of woods)...... 138 Taxonomic classification of some wild blueberry bushes sampled at the North Muskegon location (Paul Bros. Farm, south quadrant of woods)... 139 Table 43 44 45 46 47 48 49 50 51 52 Page Taxonomic classification of some wild blueberry bushes sampled at the North Muskegon location (Paul Bros. Farm, west quadrant of woods).... 140 Summary of taxonomic classification of some wild blueberry bushes sampled at the North Muskegon -location (Paul Bros. Farm, east, south and west quadrant of woods)........................... 141 Summary of taxonomic classification of some wild blueberry plants bushes in woods surrounding six different commercial fields containin BBLMV-infected bushes in WestCentral Mich1gan. (5 meters)................................... 152 Summary of taxonomic classification of some wild blueberry bushes sampled from woods surrounding six different commercial fields containing BBLMV- infected bushes in WestCentral Michigan. (50 meters)................. ...... ..... ...... 153 Summary of taxonomic classification of some wild blueberry bushes sampled from woods surrounding six different commercial fields containing BBLMV- infected bushes in WestCentral Michigan. (100 meters).. ...... ......... ........... ..... 154 Taxonomic classification of some wild blueberry bushes sampled at the South Haven-1 location (DeGrandchamp Farm, House field, west)....... 155 Taxonomic classification of some wild blueberry plants sampled at the South Haven- 2 location (Degrandchamp Farm, House field, east) ....... 156 Taxonomic classification of some wild blueberry bushes sampled at the South Haven-3 location (east and north quadrant of woods)........... 157 Taxonomic classification of some wild blueberry bushes sampled at the South Haven-4 location (DeGrandchamp Farm, Springett south, east and south quadrant of woods) ..... ...... ........ .. 158 Taxonomic classification of some wild blueberry bushes sampled at the South Haven-5 location (DeGrandchamp Farm, Springett north, south and north quadrant of woods)..................... 159 Table 53 54 55 56 Page Taxonomic classification of some wild blueberry bushes sampled at the South Haven—6 location (east and south quadrant of woods). ........ .. 160 Summary of taxonomic classification of some wild blueberry bushes sampled from woods surrounding six different healthy blueberry commercial fields........................................ 161 Nectar yield and sugar concentration from blossoms of three different types of wild blueberry and one commercial blueberry cultivar ....... ...... 163 Spread of BBLMV-infected pollen from a commercial planting containing BBLMV-infected bushes to five surrounding healthy commercial blueberry fields. 165 Figure 10 11 LIST OF FIGURES Page Wild blueberries in the woods surrounding a blueberry commercial field.................... Testing samples from wild blueberries in the woods by ELI SA.............................. Chenopodium quinoa used as a BBLMV indicator plant...... ...... ............................... Location of the BBLMV source field (A) and the five surrounding fields without BBLMV-infected bushes, from the experiment to determine spread of BBLMV-infected pollen between plantations in WestCentral Michigan.......... ............... ... Pollen traps used to trap BBLMV-infected pollen or healthy pollen from bee hives ............ .... Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in the intine and exine of BBLMV-infected blueberry pollen grains......... ............ .... Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV-infected blueberry pollen grain cell cytoplasm........................ Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV- infected blueberry anthers ......... .... ....... .. Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in pollen grains within BBLMV-infected blueberry anthers.. Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV— infected blueberry pollen grains ................ Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV— infected anthers.. ...... . ..... . ..... . ........... xiii 41 44 49 56 60 62 64 66 69 . 71 igure 13 14 15 16 17 18 19 20 21 22 23 Page Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV- infected blueberry pollen grains and anthers..... 73 Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in healthy blueberry pollen grains and anthers...... 75 Distribution of BBLMV-infected wild blueberry bushes in woods adjacent to a commercial planting containing BBLMV-infected bushes. Grand Junction, Van Buren county (Holmquist Farm) ............... 91 Distribution of BBLMV-infected wild blueberry bushes in woods adjacent to a commercial planting containing BBLMV-infected bushes. Breedsville, Van Buren county (Crawford Farm)..... ...... .... 93 Distribution of BBLMV-infected wild blueberry bushess in woods adjacent to a commercial planting containing BBLMV-infected bushes. Grand Haven-1, Ottawa county (Zelinka Farm)................... 95 Distribution of BBLMV-infected wild blueberry bushes in woods adjacent to a commercial planting containing BBLMV-infected bushes. Grand Haven-2, Ottawa county (Novak Farm).... ................. 97 Distribution of BBLMV—infected wild blueberry bushes in woods adjacent to a commercial planting containing BBLMV-infected bushes. Fruitport, Muskegon county (Dirkse Farm)..... ....... ...... 99 Distribution of BBLMV-infected wild blueberry bushes in woods adjacent to a commercial planting containing BBLMV-infected bushes. North Muskegon, Muskegon county (Paul Bros. Farm) ........ ...... 101 Summary of distribution of BBLMV-infected wild blueberry bushes in woods adjacent to six commercial plantings containing BBLMV-infected bushes. WestCentral Michigan. 1991. ............ .... 105 Vaccinium corymbosum growing in the woods.. 143 Vaccinium myrtilloides growing in the woods..145 Vaccinium angustifolium growing in the woods 147 INTRODUCTION Blueberry leaf mottle virus (BBLMV) has been described as causing an economically serious disease of highbush blueberry (Vaccinium corymbosum L.). The virus was detected for the first time, during 1977, in mature highbush blueberries ( Vaccinium corymbosum L. cv Rubel) growing in a four hectare field near Hartford, Michigan (Ramsdell and Stace-Smith, 1979). The bushes showed a general decline, dieback condition, leaf mottling and deformation. Later, the disease was found in Vaccinium corymbosum L. cv Jersey showing milder symptoms such as bush stunting and growth reduction (Ramsdell, 1987). The vector of BBLMV virus is the honeybee (Boylan-Pett, 1992; Childress and Ramsdell, 1987). Although some studies indicate that natural spread of the disease is by pollen, localization of virus particles in pollen grains has not been completely established. Also, there is little information about movement of the virus in the plant. It appears that infection may not in all cases be restricted to the gametophytic tissues of the pollen recipient. The virus can be transmitted by infected pollen to healthy ovules and in addition, it can invade and systematically infect the ovule—bearing mother plants. (Childress and Ramsdell, 1987). There has been a lack of information regarding he possible endemic presence of BBLMV in wild blueberry Jopulations surrounding commercial blueberry fields. Such plants could act as a reservoir of the virus and a source of infection to commercial plantings. The objectives of this thesis were : i) to determine the location of BBLMV in diseased pollen and anthers using transmission electron microscopy; ii) to gather information about the movement of the virus in healthy plants pollinated with diseased pollen; iii) to determine the possibility of vertical transmission of the virus to the seedlings from infected and non-infected mother plants pollinated with healthy or diseased pollen; iv) to establish whether BBLMV infection is endemic in the wild blueberry population, and v) to determine how far BBLMV-infected pollen moves from a blueberry commercial field with infection foci to other fields without infected bushes. LITERATURE REVIEW General Background Blueberry leaf mottle virus (BBLMV) has been reported from different areas of southwestern and western central Michigan. Even though the economic importance of the disease has not been well established, bushes infected with the virus are stunted and very unproductive (Ramsdell, 1987). Systemically infected plants show leaf mottling and sometimes deformation accompanied in severe cases by a general dieback of the main stem in the canopy along with a general reduction in plant vigor. The leaves appear puckered with narrowed leaf blades and are particularly noticeable on new growth from the crown during the growing season. Often, infected bushes are cut back by growers, but symptoms remain and are still noticeable on emergent infected shoots. In Vitis labrusca L. cv Concord, the virus causes a delay in bud break, irregular elongation of fruits, pale green foliage and straggly fruit clusters (Uyemoto et a1, 1977). Physical and chemical properties of the virus have been determined and blueberry leaf mottle virus (BBLMV) has been placed as a putative member of the nepovirus group. Purified virus particles separate into three components in sucrose density gradients. The middle and bottom components contain particles that measure 28 nm in diameter and the top component appears as empty shells when stained with uranyl acetate. The protein coat subunits have a molecular weight of about 57,000 daltons. Sedimentation coefficients of the T, M and B components are 53, 120 and 1285, respectively (Ramsdell and Stace-Smith, 1983). Two species of single stranded RNA have been observed and both are required for optimum infection. Two strains have been described. The MI strain is found in Blueberry in Michigan (Ramsdell and Stace Smith, 1979) and the NY strain, is found in grape vines in New York state (Uyemoto et al., 1977). Although the two strains may differ in their ability to infect blueberry and grapevine, they cannot be distinguished on herbaceous hosts. The virus is an excellent immunogen. Uyemoto et a1. (1977), reported a serological relationship between the BBLMV New York strain and grapevine Bulgarian latent virus (GBLV-Ev) . However, results obtained by Ramsdell et al., 1979 indicated that the relationship between both viruses is distant. Purified preparations of the Michigan (MI) and New York (NY) strains reacted weakly with antiserum to grapevine Bulgarian latent virus. On the other hand, MI and NY strains showed reactions of identity against antiserum to the MI strain. These results indicate that the MI and NY strains are almost identical and that the NY strain is too distantly related to GBLV to be considered merely a strain of this virus as suggested by Uyemoto et al., 1977. Transmission Plant viruses, in general, depend on various external agents for transmission or introduction into suceptible hosts (Sheperd, 1972). Transmission and spread of virus diseases may result from the activities of nematodes, insects, arachnids and humans, or passively through seed and pollen of commercial or naturally occuring wild hosts, with various combinations of modes of dispersal. Although the vector of BBLMV under commercial conditions appears to be the honeybee, there may be additional vectors, which remain unknown. Although physical and chemical properties of the virus have placed it as a putative member of the nepovirus group (Ramsdell and Stace-Smith, 1983), the suspected nematode vector, Xiphinema americanum, Cobb, 1913, is not associated with the disease (Childress and Ramsdell, 1986a). Uyemoto et al., 1977, found no transmission of the Y strain to bait seedlings of Chenopodium quinoa Willd., or o cucumber planted in x. americanum-infested soil taken rom beneath a diseased Concord grape vine. In controlled xperiments, this nematode did not transmit BBLMV to Chenopodium quinoa or Nicotiana Clevelandii (Childress and Ramsdell, 1986b). However, transmission of the virus through seed and pollen is characteristic of several members of the nepovirus group. Although this mode of dispersal is usually considered to be of secondary importance, the fact that blueberry leaf mottle virus antigen was detected by ELISA in the pollen from mature infected cv. Jersey bushes (Childress and Ramsdell, 1986b), indicates that natural spread is probably by pollen. Also, virus particles were easily removed from the surface of blueberry pollen grains by repeated washings in phosphate buffered saline (PBS) solution. The virions would he therefore located both on the surface and either within the pollen grains or in the lipid layer of the pollen exine. In addition, hand pollination of 15 2-year old blueberry bushes with BBLMV-infected pollen resulted in infection of shoots of one of the plants (Childress and Ramsdell, 1987). In a separate experiment, under field conditions, transmission occurred during bloom in a 2 to 3 week period from BBLMV—diseased source bushes to young virus-free potted bushes placed around a source bush (Childress and Ramsdell, 1987). Also, the virus was readiliy transmitted to healthy Chenopodium quinoa indicator plants by rub-inoculation of pollen obtained from BBLMV-infected blueberry bushes. The random and relatively rapid spread of BBLMV in a :ommercial planting does not resemble the characteristics of a virus spread by a nematode vector, but rather, it fits with a bee- mediated pollen-borne pattern of spread (Childress and Ramsdell, 1987). This would represent an alternate mode of BBLMV transmission to highbush blueberry in the absence of the nematode vector, or may represent an ecological adaptation for a more rapid mode of spread (Ramsdell, 1987). Free and Williams, 1972, determined the amount and distribution of pollen on the bodies of bees caught within their hive, leaving their hive, or foraging on flowers. They could detect some foreign pollen on the bodies of foraging bees, even though this was diluted with pollen from the crop being worked. These results, associated with the fact that BBLMV-contaminated pollen has been detected on the bodies of foraging honeybees would suggest that this virus would be spread mainly by these insects, which carry BBLMV-contaminated pollen. Childress and Ramsdell, (1987), found that 51.4 % of the pollen baskets from the legs of foraging honeybees trapped in a commercial field, containing BBLMV-infected bushes, were BBLMV-positive when tested by ELISA. The source of contaminated pollen could have been from BBLMV-infected plants within the field or wild blueberry plants surrounding the commercial field. Later, in 1992, Boylan-Pett et al., were able to prove in hive—pollen :ransfer of BBLMV-infected pollen from bee to bee. Their esults suggest that virus transfer occurs not only within a ingle colony but also between colonies in the same or ifferent apiaries. Bumblebees and solitary bees which are also blueberry pollinators, could be playing an important role in virus transmission. Free and Williams, 1972, determined that these insects usually had more pollen on their bodies than did honeybees and also some of them, had much more extraneous pollen, probably reflecting their greater tendency to visit more than one flower species per trip. This could be related with the movement of virus from wild plants to commercial blueberries and vice versa. Pollinization. The highbush and lowbush blueberry blossoms have all the characteristics of an entomophilous flower, adapted for insect pollination. The bell-shaped corolla of the flower is well-developed and fragant and at its base are located nectaries, which secrete the nectar that bees seek (Eck, 1988; Shutak, 1966). Pollen grains are comparatively heavy, clumped and not easily propelled by wind. Insect pollinated species tend to have a thick layer of lipids and carotenoids on the outer surfaces of the pollen exine. These characteristics might be related with virus transmission. A thicker exine could stabilize virions located on the surface of the pollen grain protecting them from the ultraviolet radiation. Also, the fact that blueberry pollen grains are clumped together gives an additional protection (Cooper et al., 1988). Each grain is a tetrad divided externally into four cells and generally produces just one germ tube as has been indicated by different authors. Merrill, (1936), obtained evidence that not more than one germ tube was produced by a single blueberry pollen grain. Brewer and Dobson, (1969b), reported that 98 % of germinated cv. Rubel pollen produced single germ tubes. . Highbush blueberry flowers are perfect flowers, with a range of self-fertility and capable of setting fruit when pollinated by their own pollen. However, the peculiar structure of the flower parts tends to make self pollination improbable and facilitate cross pollination by bees. (Eck, 1988). Pollen released from the stamens does not usually fall onto the stigma but is deflected due to its angular shape, thus reducing self pollination (Merrill, 1936). Three types of insects account for nearly all of the fruit set in blueberry, leading to inter-species hybridization. These are bumblebees, solitary bees and honeybees (Eck, 1988). The first two are numerous enough to pollinate commercial blueberry crops in isolated plantings. However, in intensively planted commercial blueberry fields, colonies of honeybees are essential to ensure adequate blossom pollination. Brewer, (1969a), recommended two strong colonies per acre for cv. Jersey in Michigan. Blueberry cultivars may vary in the ability of the flower to attract insects (Marucci, 1960). These differences in attractiveness could be explained based on nectar production, sugar concentration in the nectar or flower color and odor. The fact that floral nectar plays an important role in plant pollinator interaction has been established by different authors (Cruden et al., 1983). Filmer and Marucci, (1964), working with different blueberry cultivars concluded that in general, those more attractive to bees produced larger amounts of nectar. However, Brewer and Dobson, (1969c), working with cvs. Rubel and Jersey did not find differences in the quantity and quality of nectar produced by their flowers. As a result, the bee preference for cv. Rubel flowers observed in the field could not be explained in this particular case by differences in nectar production or sugar concentration in the nectar. In general, bees are equally attracted to sugar solutions ranging from 35 to 60 percent in soluble solids. However, bees have to expend more energy and time to fill their honey stomachs with a sweeter nectar because of its viscosity. This means that bees working in cv. Jersey flowers, which have a higher sugar concentration, will visit in a given time, a lesser number of flowers than they would visit if they were working in cv. Rubel. Seed and pollen transmission. In general, transmission of virus through infected seed or pollen of either cultivated or wild plants occurs in a number of different plant virus groups; this is especially true for the nepovirus group (Bennett, 1969). This type of transmission facilitates the establishment of new infection foci which serve as secondary reservoirs during low vector populations or adverse environmental conditions. However, transmission through virus infected seed and pollen takes place less often than transmission via insect vectors and is considered to be of only secondary importance. The percentage of virus transmission to the seed has been observed to depend on several factors. Whether the virus is transmitted via the male or female gametophyte will I I I I ‘, , I I I I 3determine the amount of transmission. This will be highest when both are infected (Medina et al., 1961). Froscheiser, (1974), found that alfalfa mosaic virus was transmitted at a much higher frequency to the seeds when pollen was the virus source. However in general, transmission of viruses through pollen as compared to ovule transmission is a relatively inefficient method for the viruses to become seed ,transmitted (Mandahar, 1981). Ryder, (1964), working with Ilettuce mosaic virus found that efficiency of pollen “transmission was about one tenth that of ovule transmission. jAlso, infected pollen played no significant role in transmission of tobacco ringspot virus in Soybean, so that seed transmission of this virus depended on ovule infection (Yang and Hamilton, 1974). Low efficiency of seed transmission of pollen borne viruses can be accounted for in several ways. First, the number of pollen grains produced per flower or anthers is reduced in virus- infected plants as compared to healthy plants. Also, the association of virus with the pollen grain can have an effect on the developmental pattern, morphology and vigor of the pollen, and consequently, over its ability to fertilize the ovule and transmit the virus. Yang and Hamilton, (1974), working with tobacco ringspot virus (TRSV) determined that pollen from TRSV infected soybean ( Glycine max ) plants was low in germination capacity and its germ tubes elongated slower than germ tubes of pollen from virus- :free plants. In 1983, Haight and Gibbs reported some Idileterius morphological effects of TRSV on germ tubes. The Isame authors reported that pollen grains from Chenopodium fquinoa plants infected with sowbane mosaic virus (SMV) were icollapsed, grooved and had sunken opercula, whereas those Ifrom flowers of virus-free plants were smooth, rounded and lwith protuberant opercula. Similar but less obvious symptoms Iwere found in Plantago lanceolata infected with either lribgrass mosaic virus or broad bean wilt virus. However, Iother authors have not seen major negative effects of Avirions on developmental patterns and vigor of the pollen igrain. Carroll et al., (1976), working with two different st 319 me pr 54 13 strains of barley stripe mosaic virus did not see any apparent interference of virions with normal mitosis and meiosis during development of anthers and pollen. Only a few precursor cells showed degeneration induced by the virus. Later in 1988, Massalski et al., found that the rates of germ tube elongation of pollen in vitro,from cherry leaf roll virus (CLRV)—infected birch (Betula pendula)did not differ significantly from those of pollen from virus-free trees. Percentage of germination of pollen from virus infected and virus- free trees did not differ greatly from one another. On the other hand, host-virus combination has been described as another factor affecting the rate of spread via seed and pollen (Shepherd, 1972). Studies on seed—borne viruses like soybean mosaic and bean mosaic have shown different rates of transmission depending on the host cultivar and virus strain. In the case of viruses transmitted by pollen, Cameron et al., (1973) found that the rate of spread of prunus necrotic ringspot virus (PNRSV) in orchards was not the same between and within Prunus species. Also, for PNRSV and sour cherry yellows viruses, the rate of spread in young orchards was dependent upon the age of the trees, the proximity of older diseased plants, and the amount of disease within the orchard. PNRSV spreads very slowly in orchards under 4 years old, but can spread rapidly in older orchards. On the other hand, sour cherry yellows virus does not spread rapidly until after the orchard is at least 10 14 years old. Seed transmission of plant viruses also depends on the virus' ability to infect the floral meristem and subsequently invade the gametophyte and gametes of its host (Bennett, 1969). Only those viruses which are present in situ at the time of differentiation of male and female sporogeneous cells and before the callose wall is laid down will be able to infect microspores, megaspores and embryos (Mandahar, 1981). Then, whether infection occurs and whether the virus becomes systematic prior to or after anthesis will play an important role in the amount of transmission through seed or pollen (Couch, 1955 and Crowley, 1959). Microspore mother cells become separated from the tapetum and the rest of the anther at the beginning of the meiotic prophase. Therefore, internal virus infection of pollen microspores is initiated at or before this time. Plasmodesmatal connections between the tapetum and the anther cell walls or between the tapetum and the microspore mother cell walls are few or non- existent before the meiotic phase. The microspore is protected from virus invasion through callose deposition in cell walls. Failure to invade the meristem early, will allow pollen mother cells and megaspore mother cells to escape from virus invasion, resulting in the production of virus— free pollen and egg sacs. Other factors which may also affect the amount of virus transmitted via contaminated pollen are the concentration and properties of the virus, the mode of transport of contaminated pollen, whether the 15 virus is found on the inside of pollen grains or outside, and the type of pollination. The process of transmission of contaminated pollen is likely to be more efficient in open— pollinated plants than it is in self-pollinated plants. (Cooper et al., 1988). Transmission of virus via contaminated pollen will occur in fairly localized areas around infected plants because of its short period of viability of the pollen. However, PNRSV and sour cherry yellows virus can spread by infected pollen over a considerable distance (Davidson and George, 1964). When an infectious agent is passed directly from a parent to its progeny, the process is termed vertical transmission. In plants this can occur via seed and/or pollen and it has been demonstrated to take place in 9 of the 36 recognized groups of plant viruses, and two viroids (C00per et al., 1988). For vertical transmission to occur through the seed, the embryo must be invaded early in its development by virus from either maternal tissue or from Virus infected pollen. The physical separation of the embryo Su99ests that infection must be introduced via one of the Parents at the time of fertilization. Entrance of pollen- borne viruses into ovules may be assumed to occur along with the male gametes, which move through the pollen tube as it grows into the embryo sac (Shepherd, 1972)- Virus may accompany the male gametes, one of which unites with the egg cell during fertilization. The other male gamete unites Wlth 16 the polar nuclei, which gives rise to the endosperm. In this case, the infectious agent is passed directly from a parent to its progeny. Only in a few instances have viruses been shown to be introduced via pollen into the plants pollinated (horizontal transmission). In general, pollen-transmitted viruses remain restricted to the developing embryo because of the surrounding callose layer. They normally cannot escape this isolation and cause systemic infection of the parent plant. However, about a dozen cases are known where virus brought by pollen to the flower—bearing plants becomes systemically distributed in the latter and incites disease (Mandahar, 1981). Some Ilarvirus, for example, are now considered to be naturally transmitted in this way among woody rosaceous hosts (tobacco streak virus in Rubus spp.) (Cooper et al., 1988). Nepoviruses provide the other well- characterized example of horizontal spread via pollen. The route through which viruses are transferred horizontally from the pollen to the pollinated plant is unknown. It is assumed that the virions must escape from the ovule after their introduction via fertilizing germ tubes. An internal infection of the pollen grain would be a Prerequisite. However, the fact that surfaces of pollen grains are contaminated with infectious virions suggests new ways of infection. Insects, through feeding in and wounding floral or vegetative tissue, may directly inOCUIate stable POIIen-associated viruses in the mother plants (COOPer et 17 al., 1988). Infection of healthy ovules (vertical transmission) and mother plants (horizontal transmission) from virus infected pollen, has been demonstrated for barley stripe mosaic virus, elm mosaic virus, apple chlorotic leafspot virus, black raspberry latent virus, avocado sunblotch virus, sour cherry yellows virus, cherry leafroll virus (CLRV-W), necrotic ringspot virus (PNRSV) and prune dwarf virus (PDV) (Gold et al., 1954; Callahan, 1957; Way and Gilmer, 1958; Gilmer and Way, 1960; Das and Milbrath, 1961; George and Davidson, 1963; George and Davidson, 1964; Davidson and George, 1964; Gilmer, 1965; Desjardins et al., 1979, and Mircetich et al., 1982). Gold et al., 1954, using transmission electron microscopy, could detect the presence of rod-shaped particles associated with barley stripe mosaic virus in seeds produced from healthy barley plants pollinated with infected pollen. They concluded from these results that the virus was being transmitted vertically via pollen. Four years later, Way and Gilmer (1958), reported vertical transmission to five of 18 cherry seedlings that resulted from a cross of healthy English morello (female parent) by virus- infected Montmorency (male parent). These were found to be infected with PNRSV. Two years later, Gilmer and Way, (1960) presented conclusive evidence that both PNRSV and PDV are vertically transmitted by infected P011en to seeds produced on virus-free sour cherry trees. The data Presented sh of in wh show that about 25 % of seeds resulting from fertilization of healthy ovules with pollen from diseased trees were infected with PNRSV, PDV or both viruses. Other viruses which have been reported to be transmitted vertically by infected pollen are cherry leafroll virus (CLRV-w), the causal agent of the black line disease and avocado sunblotch virus. CLRV is commonly transmitted vertically through both pollen and ovules, infecting seeds and seedlings of woody perennial hosts including birch (Cooper et al., 1984). Mircetich et al., (1982), and Cooper et al., (1984), concluded from their work that CLRV-w is transmitted through seed and pollen from black line-affected English walnut trees. Flowers on healthy English walnut trees pollinated with CLRV-w-infected pollen resulted in virus-infected nuts. On the other hand, Desjardins et al., (1979), experimentally demonstrated pollen transmission of the sunblotch viroid in (avocado. ‘ Horizontal transmission has been proven for several viruses. In 1961, Das and Milbrath reported 10 % transmission of PNRSV through the flower via infected pollen. Nine of the 97 squash plants which were pollinated with PNRSV-infected pollen developed symptoms in the tips of the plants. Later, infection of healthy stone fruits via movement of virus-containing pollen by honey bees was demonstrated for PNRSV and PDV in Montmorency cherry (George et al., 1963). Infection was not detected until three years after initial inoculation. 19 The association of honey bees with horizontal transmission of virus by pollen has also been established by other authors. Howell, (1988), working with sweet cherry could detect viable pollen contaminated with infectious PNRSV on the bodies of bees that emerged from the hives. The practice of moving commercial beehives directly from earlier blooming orchards in California to sweet cherry orchards in Washington could be related with the appearance of new PNRSV and cherry rugose mosaic infected sites. However, considering that hundreds of honeybees repeatedly visit thousands of blossoms on adjacent healthy and PNRSV-infected trees during the bloom period and only an average of 10 % of the adjacent trees become infected each year, the rate of PNRSV transmission via pollen to a mature tree per bee visit would not be very high. However,later in 1976, Davidson working with sour cherry orchards in Ontario, Canada concluded that the transmission of virus-infected pollen from diseased trees is the major and, perhaps the only means of natural spread of PNRSV in the Niagara peninsula. vi th fo 20 Virus movement There is little information concerning how different viruses move from the infected plant to the pollen grains in} the anthers. In general, the process has been explained following the different stages of pollen grain formation. Archesporial cells, present in Young anthers, will differentiate in to two constituents, the parietal cell layer and the sporogeneous tissue. The parietal cell layer will generate the outer wall of the anther and the tapetum (inner layer) which subsequently lyses, providing the pollen kit that coats the sculptured outer suface of the pollen grain (Cooper et al., 1988). Therefore, those viruses which are located on the surface of pollen grains, should move into tapetal tissues before lysis. On the other hand, sporogenous tissue will differentiate in pollen mother cells which later, after mitotic divisions, become microspores that mature into pollen grains. Movement of the virus from the infected plant to the anther and then to the microspore should occur before meiotic prophase. It has been demonstrated for different species that at this moment cells become separated from the tapetum and the rest of the anther (Heslop-Harrison, 1964). Plasmodesmatal connections between the tapetum and the anther wall cells or between the tapetum and the microspore mother cells are few or non-existent during the meiotic prophase. virions located inside the pollen grain should move from the outer wall of the anther and inner layer into the microspores before this stage. Carroll et al., (1976), working with two different strains of barley stripe mosaic virus in barley, could see in anther and pollen sections, virions attached to microtubules during the pre-meiotic and meiotic stages. Nothing is known about the mechanism by which pollen transmitted viruses are transferred from the pollen grain to the flower. In general, two mechanisms of infection can be described. First, infection of the stigma may occur from virus carried on the pollen surface and second, infection of the egg during fertilization may result from virus carried within the pollen, followed by infection of the mother plant. The lack of plasmodesmatal, vascular or other direct protoplasmic connection between embryo and maternal tissue is regarded as an effective, major obstacle to the transmission of virus from embryo to the maternal plant. Infection of maternal tissue may take place in the stigma or style as the result of penetration by an intimate contact with diseased pollen tubes some of which may burst (George and Davidson, 1963). In plant-to-plant transmission, therefore, pollen appears to play the additional role of vector in that it acts as a carrier and introduces the virus into the tissue of the maternal plant of which the pollen does not become an integral part. me: me tu by in ZY th 22 Childress and Ramsdell, (1987), proposed different mechanisms through which BBLMV could enter into blueberry ~mother plants via pollen: Mechanical inoculation by the germ tube during disruption of the exine, infection of the stigma by contact with virus contaminated pollen, infection of the integuments of the ovule, infection of the developing zygote, and finally, transmission through the cytoplasm of the vegetative cell to the embryo. Not much is known about virus movement within the plant after infection. The sequence of symptom development in cv. large yellow glass sweet cherry infected by pollination, demonstrated that sour cherry yellows virus (SCYV) moved rapidly from flowers or young fruits into woody tissue (Gilmer, 1965). George and Davidson in 1964, made some attempts to determine how long blossoms or fruits must remain on the trees for infection to occur. The results (indicate that the virus PNRSV, carried in the pollen, probably passed from the blossoms or fruits into twigs of the female parent tree within two weeks after the late shuck stage. This supports the evidence that infection occurs at or shortly after full bloom. Direction of movement of virus after infection has been investigated by a number of researchers. Samuels, (1934) and Capoor, (1949), working with tomato plants infected with tobacco mosaic virus (TMV) and potato virus X (PVX), found that the virus moved downward toward the roots first and then upward via the roots to the shoot tips. TMV can also tr: re; be an travel both upward and downward simultaneously after reaching the stem (Kunkel, 1939). Another virus which has been reported to follow the same movement pattern is tomato aucuba mosaic virus. Robb, (1964), working with tomato plants got the same general results. Maduewesi and Hagedorn, (1965), working with Wisconsin pea streak virus, found that the virus moved upward initially, passing some nodes and then moved down to the roots. They also reported some simultaneous up and down movement. Long distance transport of virus within a plant has been related to the direction of food transport. Bennett, (1940), working with TMV in tobacco and tomato plants found that translocation of the virus was accelerated in an acropetal direction if the top of the plant was removed or defoliated. Later, in 1960 working with beet yellows virus and beet curly top virus, he found the same phenomenon. He associated this response with transportation of carbohydrates. There is direct and indirect evidence that most of the viruses which spread systemically within the plant, are translocated in the phloem. However, there are few reported cases where the xylem also plays a certain role in long distance movement. Schneider and Worley, (1959a, 1959b), working with southern bean mosaic virus (SBMV) in Pinto bean, found that SBMV was able to move long distance upward and downward in the phloem as well as long distances through the xylem upward in the water stream of a local lesion host 24 such as the Pinto bean. Later, in 1966, Chambers and Francki, working with lettuce necrotic yellows virus (LNYV) could find virions in the young xylem cells of Nicotiana glutinosa leaf veins. They did not find virions associated with the phloem. Urban, (1987), using dot—blot immunoassay and a cDNA probe was able to detect Blueberry shoestring virus in the xylem of infected blueberry plants. Long distance transport of the virus was through both the xylem and the phloem. Virus localization Virus association with pollen has been known for over a half a century. Early workers inferred the presence of virus by homogenizing pollen taken from infected plants and inoculating indicator plants which were the monitored for symptoms (Cooper et al., 1988). Since then, many experiments have been done in search of the location of virus-like particles in the pollen and its importance related to vertical or horizontal transmission. Pollen-transmitted virus can be located inside pollen grains or on the pollen surface. This could determine the mechanism by which the virus is transmitted to flowers and the mother plant. The location of virions on the surface of pollen grains has subsequently been demonstrated in a number of different studies. Cory and Hewitt, (1968), working with Vitis vinifera recovered grapevine viruses from the pollen of three 0‘ fanleaf grapevi inocula (1974) , soybean contami observe surface and Vig brome 1 were c< examin; infect. a proq the ta develo and Wi P011en and TN Campbe Canta] °°ntan four 1 inOCui after 25 three cultivars infected with three strains of grapevine fanleaf virus (GFV): GFV—yellow mosaic, GFV-veinbanding and grapevine yellow vein virus. Pollen grains were washed and inoculated to Chenopodium quinoa. Later, Yang and Hamilton, (1974), working with tobacco ringspot virus (TRSV) infected soybean plants demonstrated the presence of virions on contaminated pollen grains. Then, in 1977, Hamilton et al., observed by transmission electron microscopy (TEM) that the surfaces of pollen grains from plants of Phaseolus vulgaris and Vigna unguiculata infected with tobacco mosaic (TMV), brome mosaic (BMV) and southern bean mosaic virus (SBMV) were covered with virus-like particles. Electron microscopic examination of ultrathin sections of stamens from SBMV- infected bean plants at various stages of development showed a progressive infection of the filament, the anther wall and the tapetal layer. In the later stages of pollen development, crystalline masses of virions accumulated on and within the outer surface of the exine of maturing pollen. Triturates of pollen from plants containing SBMV and TMV were infective. One year later, Alvarez and Campbell, working with squash mosaic comovirus-infected cantaloupe melons, demonstrated the presence of virions on contaminated pollen grains. Infected pollen grains washed four times and homogenized, did not cause symptoms when inoculated in indicator plants, whereas pollen triturated after a single washing cycle did. surf: Coop¢ dem01 and1 pres in p TEM with endc bar] wor} virf vege loo: det. Wor of ‘ wal Sat in wit Han 00! Sec Cy1 26 Alternatively, some viruses have been found on the surface and internally in pollen grains.Massalski and Cooper, (1983), working with cherry leafroll virus (CLRV) demonstrated the presence of antigens both on the surface and within birch pollen. Later, in 1984, the same authors presented further evidence. Explicit evidence for the internal location of viruses in pollen relies upon TEM data. Gold et al., (1954), using TEM detected the presence of rod shaped particles associated with barley stripe mosaic virus (BSMV) in leaves, embryos, endosperm, pollen and unfertilized pistils of infected barley plants. Later, Gardner, (1967), and Carroll, (1974), working with the same virus, reported the presence of virions in the nucleus and cytoplasm of the sperm and vegetative cells of BSMV-infected barley plants. Internal location of viruses in pollen grains has also been determined for other viruses. Wilcoxson et al., (1975), working with alfalfa mosaic virus (AMV) found the presence of virus aggregates in leaves, embryonic cotyledons, ovary walls, anthers and pollen from infected alfalfa (Medicago sativa) plants. Virus—like particles were located internally in the pollen grain. These results corroborate those related with vertical transmission of the virus by pollen. Yang and Hamilton, (1974), working with tobacco ringspot virus (TRSV) could detect by TEM virus—like particles in ultrathin sections of the intine of the pollen wall, the wall and cytoplasm of the generative cell, the integuments, nucellus, soyt wit} dwal for and of ‘ alm- Kel PNR of The of cor p0] frc 10¢ im wi‘ se mi ha th 27 embryo sac wall and in the megagametophytic cell of infected soybean. The strongest evidence for virus localization related with horizontal transmission via pollen lies with prune dwarf virus (PDV) and PNRSV. Cole et al., (1982), searched for the location of PNRSV associated with pollen in almond and cherry trees. They concluded that most or perhaps all of the PNRSV antigens associated with pollen from infected almond and cherry trees were located on the pollen surface. Kelley and Cameron, (1986), investigated the location of PNRSV and PDV in almond and cherry. They supported a model of an internal location of the viruses with their results. They observed virus-like particles in ultrathin sections of pollen from trees infected with PDV . Particles appeared concentrated in the areas of the pores in the corner of the pollen grain. Particles were not observed in pollen grains from healthy or PNRSV-infected trees. Alternatively, some other authors have reported localization of virus particles in infected anthers and immature pollen grains. Pacini and Cresti, (1977), working with developing pollen grains of olive (Olea europaea) could see that virus particles were present in the cytoplasm of microspore mother cells. However, just before the first haploid mitosis, virions were found only in the pores inside the invaginations formed by the plasmalemma. Later, Massalski and Cooper, (1984), working with immature anthers, obse arra cell CLR) ant] (Jm sur: par suc som whi and use a f cel SUC V11 Pa] la] 28 observed virus-like particles in packed paracrystalline arrays in cells forming the anther walls and in the sperm cell cytoplasm of immature pollen grains developing within CLRV-infected birch. This phenomenon was also observed in anther cells and pollen grains of CLRV-infected walnut (Juglans regia). These particles rarely coated the outer surfaces of developing grains. Techniques for virus localization TEM has been used to detect and locate virus-like particles in host plants since the late 19505. The value of such observations has been greatly enhanced by incorporating some other methods with this technology. . One technique which has been used to determine the location, distribution and replication of the virus within the plant, has been the ‘use of fluorescent antibodies. Lei and Agrios, (1986), used a fluorescent antibody technique to look at the number of cells infected with maize streak virus in resistant and suceptible corn lines. Urban, (1987) used also this technique to determine the presence of blueberry shoestring virus in infected tissue in highbush blueberry cv. Jersey. Another technique used for localization of virus particles in infected tissue is immunogold labeling . This technique combines the serological specificity of a gold- labeled antibody binding to an antigen with the resolution and is ex; of is la} Pr< St; Chi ch. in wi SP nu wi and ultrastructural detail provided by TEM. This technique is based upon the interaction between antibody and antigen exposed on the surface of a tissue section. Once an antigen of interest has been attached by an antibody, the antibody is detected by a gold-labeled secondary antibody or gold- labeled Staphylococcus aureus protein-A (Bendayan, 1984). Protein A is located in the cell wall of most strains of Staphylococcus aureus and consists of a single polypeptide chain having a molecular weight of 42,000. It is characterized by its ability to interact with the immunoglobulin-G of almost all mammals and in some species with Ig A and Ig M as well. However, Craig and Goodchild, (1982) and Bendayan et al., (1980), have reported non- specific labeling to an unknown component of plant cell nuclei regardless of whether antibody had previously reacted with the tissue section. It appears that less protein A~gold binds non-specifically to Lowicryl K4M-embedded ultrathin sections than to Spurr's resin-embedded sections. For application in immunoelectron microscopy, protein-A must be tagged with an electron-dense marker. Colloidal gold has been chosen because of its versatility and several other advantages. As a negatively charged hydrophobic solid, it binds to macromolecules by non-covalent electrostatic adsorption. The interaction is stable and does not affect the biological activity of the tagged molecule. Also, colloidal gold is particulate, allowing for an accurate identification of the labeled structures and quantitative eve 1a} in an re‘ e1 in fi re GC evaluations of the labeling intensity. In immunogold labeling studies, there is a compromise between excellence in tissue fixation and the preservation of protein antigenicity. The antigenicity of target proteins must be retained throughout destructive preparative procedures for electron microscopy including fixation, dehydration, resin infiltration and heat polymerization. After standard fixation in 3% glutaraldehyde and embedding in Spurr‘s resin, protein antigenicity is reduced by 90 % (Craig and Goodchild, 1982). Another common fixative, osmium tetroxide, which preserves and stains phospholipids, also severely reduces protein antigenicity (Bendayan and Zollinger, 1983). However, Sodium meta-periodate etching seems to restore the antigenicity of proteins in osmium—fixed tissues embedded in Spurr's resin (Bendayan and Zollinger, 1983). Reinke, (1989), obtained a two-fold increase in labeling sensitivity by etching. Results obtained by the same author working with Nicotiana tabacum L. cv Xanthi-infected with tobacco mosaic virus and Brassica rapa L. cv Just right-infected with cauliflower mosaic virus, showed that more highly crosslinked fixative (glutaraldehyde) and resin (Spurr‘s) resulted in better ultrastructural resolution of plant tissue, but poorer gold labeling sensitivity. A second factor which can also have an effect on the final labeling, is the incubation time with antiserum and gold probe. Fasseas et al., (1989), examined the distribution of parsnip yel Nic inn obt prc USE of US: ra1 yea an fi' C0 in an (1 tu no St an. 31 yellow fleck virus particle antigen in infected cells of Nicotiana clevelandii and Spinacia oleracea, utilizing immunogold labeling of ultrathin sections. Best results were obtained by long incubation times with antiserum and gold probe. During the last few years, immunogold labeling has been used by different researchers to determine the localization of viral protein in infected tissue. Huang et al., (1983), using this technique, determined the presence of tobacco rattle virus particles in high concentrations in the cytoplasm of infected Paeonia emodi pollen grains. The same year, Na-sheng Ling and Langenberg, used gold labeled antibody to locate viral protein in ultrathin sections of fixed barley stripe mosaic-infected wheat cells. They concluded that this procedure could be used in the intracellular localization of any protein to which an antibody can be prepared. Three years later, Garmer et al., (1986), was able to detect in ultrathin sections of fixed turnip yellow mosaic virus ( TYMV )-infected cabbage leaves not only TYMV and its coat protein but also some non- structural proteins of the virus such as the RNA replicase and its viral-encoded subunit. One year later Hills et al., (1987), was able to localize within TMV-infected tobacco leaf cells TMV viral coat protein and RNA replicase. Also, immunoelectron microscopy using gold-labelled antibodies has een used to determine virus protein associations with ylindrical inclusion bodies. Langenberg, (1986), using this technf are a: mosaic 32 technique showed that homologous capsid proteins or virions are associated with cylindrical inclusions of wheat streak mosaic virus in infected wheat cells. In BBLMV healt These Adam: oral blue (Dif P011 slut and fixa Were h a1 39a] b11f: METHODS AND MATERIALS In situ localization of viral gene products in infected pollen grains and anthers. Pollen and anthers Pollen grains and anthers from 10 different flowers in BBLMV-infected blueberry bushes and 10 different flowers in healthy blueberry bushes were collected from the field. These plants were previously tested by ELISA (Clark and Adams, 1977 and Voller et al., 1976) to confirm the presence or absence of the virus. Fixing and embedding Pollen grains obtained from BBLMV—infected and healthy blueberry plants were embedded in 0.1 % (w/v) Bacto agar (Difco) in distilled water. Next, pieces of agar containing pollen grains in it, were diced in fixative [4 % (v/v) glutaraldehyde in 0.1 M sodium phosphate (PH 7.2) buffer] and fixed at room temperature for 3 h. After primary fixation, three buffer washes were given, and then tissues were postfixed in 1 % (v/v) Osmium tetroxide in buffer for 2 h at room temperature. Fixed pollen grains in agar [1% (w/v) agarose in distilled water] were washed three times in buffer and then dehydrated in a graded ethanol series [25, 33 50 pit to pr 5P we 34 50 and 75 % (v/v)] and stored in 100 % ethanol overnight. Anthers removed from healthy or BBLMV-infected blueberry plants, were directly fixed in 4 % (v/v) glutaraldehyde, followed by the same procedures mentioned above. Samples prepared as described were infiltrated, under vacuum, with Spurr's low viscosity embeding medium (Spurr, 1969). These were polymerized at 60 C for 24 h. Trimming and sectioning Ultrathin sections (60-80 nm) for TEM were cut with a Delaware diamond knife in a Sorvall MT—2 ultramicrotome. These sections were supported by 300 mesh nickel grids. Etching Tissue sections were etched by floating grids, section side down, on drops of 0.1 % (w/v) sodium metaperiodate in sodium phosphate (pH 7.2) buffer. Three different etching times were tested (30, 45, and 60 min). Sections were then washed by floating grids on three drops of glass distilled water for 5 min each. Immunogold Labeling Grids were floated on drops of l % (W/V) dehydrated milk in sodium phosphate buffer (pH 7.2) for 30 min to block non— specific binding of antibody or protein-A to the section. Then, grids were transferred to drops of BBLMV— antiserum dii in va de‘ la 15 di a1 35 diluted 1:100 to 1:500 in 1 % (w/v) dehydrated milk and incubated at room temperature for two h. Grids were then washed three times (5 min each) on fresh drops of 1 % (w/v) dehydrated milk in sodium phosphate buffer. Next, grids were labeled on drops containing protein A / 15 nm gold (A 520 nm = 2.5) for 5 and 10 nm diameter and (A 520 nm = 3.5) for 15 nm diameter colloidal gold (Amersham International) diluted 1:20 with 1 % (w/v) dehydrated milk. Labeling was allowed to proceed for 1 h. Grids were then washed with glass distilled water. Three different controls were used : First, some grids with ultrathin sections of BBLMV-infected pollen and anthers were floated on drops of Pre-immune serum diluted 1:100 to 1:500 in 1 % (w/v) dehydrated milk and incubated at room temperature for 2 h. Next, grids were washed and gold labeled as above. Second, some grids with ultrathin sections from infected tissue were directly gold-labeled without antibody, after etching. Finally, grids with sections from healthy anthers and pollen were labeled following the same procedure mentioned above, using BBLMV antiserum. All the sections were stained in aqueous uranyl acetate ‘(saturated solution) for 30 min at room temperature followed iby a lead citrate (1.33 g lead nitrate mixed with 1.76 g sodium citrate, 8 ml of 1 N sodium hydroxide and 42 ml of C02 free water) staining for 3 min. L‘ Si micro: seasc trans and I infec were Winte buds Sone buds used heal HaVe abse for 36 Transmission Electron microscopy Samples were examined with a Philips 201 electron microscope operating at 80 kv. Determination of Vertical and Horizontal transmission of BBLMV via pollen in blueberry Experiment Three experiments were developed during the growing season in 1991 to determine whether vertical and horizontal transmission of BBLMV occurs to the seedlings, fruits, seeds and healthy small bushes pollinated with pollen from BBLMV- infected bushes. Diseased bushes used in each experiment were obtained from fields containing such plants during the winter of 1991. To confirm the presence of virus, flower buds from these bushes were tested using ELISA following the same procedure described later in the field survey. Flower buds from healthy bushes and BBLMV-infected bushes were ‘used as negative and positive controls, respectively. The ‘healthy bushes obtained from Tower View nursery, South Haven, MI, were also tested using ELISA to verify the absence of virus. Healthy and diseased bushes were stored for about two months in a cold room (4 C) to keep them dormant until used. Two experiments (May and June) were done ut EX} dii bu: bu: ca th ca an ut of doors at the Michigan State University Horticulture Experimental Station. In both experiments two cages with different combinations of diseased and healthy blueberry bushes were used. In one cage, a BBLMV-infected blueberry bush was placed together with healthy bushes. The other cage, used as control, contained only healthy bushes. During the blooming period, a hive of bees was placed within each cage for pollination. Three different combinations of fruits and seeds were analyzed in each experiment. First, those produced by flowers from the BBLMV- infected bushes pollinated with healthy or infected pollen. The second combination consisted of fruits and seeds produced by flowers from the healthy bushes pollinated with infected pollen. The third combination consisted of, fruits and seeds produced by flowers from the healthy bushes pollinated with healthy pollen. Vertical transmission To determine if vertical transmission of BBLMV was occuring with progeny, the seedlings, fruits and seeds were collected from the bushes in the cages. The three different combinations of fruits and seeds mentioned above were analyzed in each experiment. Immature fruits were collected from the pollinated plants, 4 wk after pollination and analyzed by ELISA to determine the presence or absence of virus. For this analysis, 50 % of the fruits in each clue matr dill pot: and and big] bec; reg me for bus bus res ext roc vet See des dii den in bi] Bil 38 cluster were collected, leaving the rest on the bush until maturity for further analysis. Fruits were weighed and diluted 1:10 (w/v) in 10X extraction buffer [0.1 M sodium- potassium phosphate buffer containing 0.05% tween 20 (v/v) and 2 % (w/v) polyvinylpyrrolidone, molecular weight 40,000 and 0.2 % (w/v) egg albumin (w/v)] and then ground with a high speed homogenizer. A 10X extraction buffer was used because of the low pH obtained when grinding fruits with regular ELISA extraction buffer (1 X). The rest of the analysis was performed using the same procedure described for ELISA in the field survey. Immature fruits from healthy bushes pollinated with healthy pollen and from diseased bushes were used as negative and positive controls, respectively. Later, at maturity, fruits were harvested, seeds extracted and sown in pots. The pots were stored in a cold room at 4 C to vernalize the seeds. About 6 mo later, pots were moved to a greenhouse for germination and growth. Seedlings were analyzed by ELISA using the procedure described for leaf analysis in the field survey. To determine differences in vertical transmission among different blueberry cultivars, a similar experiment was developed during the spring, 1990. In this case, mature fruit clusters from healthy plants of five different 1neberry cultivars (Spartan, Elliot, Bluecrop, Jersey and Blueray) which were previously pollinated with BBLMV— infe in 1 dEtl (ho: 199 C f gre vit San blo ELI the des di] f0] an; BBI h We Dr 39 infected pollen using honeybees, were analyzed by ELISA. Horizontal transmission The healthy bushes pollinated with BBLMV—infected pollen in the assays mentioned above, were analyzed periodically to determine the presence of the virus in the mother plants (horizontal transmission). At the beginning of the autumn, 1991, mother plants (bushes) were stored in a cold room at 4 C for vernalization. Three no later, they were moved to a greenhouse (22 to 30 C) under cool-white fluorescent light with a 15 h day length to break dormancy. After 12 wk, samples from different parts of the bushes, including blossoms, leaves and roots, were collected and analyzed by ELISA to determine the presence and location of the virus in the plant. The procedure followed for ELISA was the same as described for the field survey. Tissues were weighed and diluted 1:5 (w/v) in extraction buffer. In order to determine differences in pollen transmission for the five different blueberry cultivars, the same analysis was performed. Healthy bushes pollinated with BBLMV-infected pollen that were ELISA-positive in 1990 (fruit clusters) were stored in a cold room at 4 C and 3 mo later, moved to a greenhouse. Samples from roots and leaves were collected and analyzed by ELISA to determine the Presence of the virus in the mother bush. the u SOUI‘C and BBLM‘ infe. fieL comm Thes the the from voo< Werc tip: ide inf In orj no: 40 Field survey Samples It was not known whether BELMV infection is endemic in the wild blueberry population, thus, possibly serving as a source of infection to commercial plantings. A field survey was performed during the summers of 1991 and 1992 next to six different commercial fields containing BBLMV-infected bushes and in six fields without BBLMV- infected bushes in Western— Central, lower Michigan. In the fields containing diseased bushes, samples were taken from commercial bushes showing symptoms characteristic of BBLMV. These were analyzed by ELISA to reconfirm the presence of the virus. Samples from wild blueberries in the woods adjacent to the commercial planting were taken from bushes radiating out from the commercial field at 5 m , 50 m and 100 m into the woods (Figure 1). From each bush, a minimum of 10 leaves were taken, preferentially selecting young tissue from the tips of new growth. Each wild blueberry sampled was identified with a number and a tag. This assured that BBLMV- infected plants could be found later for further analyses. In each location, woods were divided according to their orientation with respect to the commercial field, i.e. north, south, east and west . From each direction, a max1mum 41 Figure 1. Wild blueberries (V. angustifolium)in the good: surrounding a blueberry commercial 1e . 42 \. of: ver Thi Vol 91: mac' sex su] teI CE] ro' ph co 50 m1 MC 43 of 100 samples was taken from bushes in the woods. Virus testing Samples collected during the summers of 1991 and 1992 were tested for presence of the virus by ELISA (Figure 2). This technique was developed by Clark and Adams, 1977 and Voller et al., 1976. In order to purify the anti-BBLMV gamma globulin from a specific polyclonal antiserum previously made in rabbit (Ramsdell and Stace-Smith, 1979), 1 m1 of serum was precipitated by adding 9 ml of saturated ammonium sulfate. After incubation for 30 to 60 min at room temperature, the precipitate was collected by a low speed centrifugation at 5,000 rpm for 5 min in a Beckman No. 30 rotor. The precipitate was resuspended in 2 m1 1/2 strength PhOsphate buffered saline (PBS, 0.015 M potassium phosphate containing 0.8% (w/v) NaCl) and dialized three times against 500 ml 1/2 strength PBS. The IgG was then passed through a 6 ml bed of DEAE-23 cellulose (Sigma Chemical Co., St. Louis, MO 63178) which had been pre-equilibrated in 1/2 strength PBS. Two ml fractions were collected from the column and monitored for absorbance activity at 280 nm. The fractions containing IgG were saved, adjusted to a concentration of 1 mg/ml (A280 of 1.4 = 1 mg/ml) . To conjugate the IgG with alkaline phosphatase (Sigma No. P—5521). a 2 mg/ml solution of ammonium sulfate—precipitated enzyme was added to 1 ml 0f purified anti-BBLMV-IgG and then dialized three times 44 Figure 2. Testing samples from wild blueberries in the woods by ELISA. 45 aga an of (e1 631 die BBI alt bui COI Sp: Th SO- In ra be it 46 against 500 ml full strength (0.015 M) PBS. Conjugation of enzyme to the IgG fraction was facilitated by the addition of 0.05 % (v/v), final concentration, glutaraldehyde (electron microscope grade, Sigma Chemical Co., St Louis, MO 63178) for 4 h at 20 C. Glutaraldehyde was removed by dialysis against PBS, three times. The conjugated-anti- BBLMV-IgG was stored with 5 mg/ml (w/v) of bovine serum albumin and 0.01 % (w/v) sodium azide at 4 C. Leaves were weighed and diluted 1:5 (w/v) in extraction buffer [0.01 M sodium-potassium phosphate buffer, pH 7.4, containing 0.05 % Tween 20 (v/v) and 2 % polyvinylpyrrolidone, molecular weight 40,000 (Sigma) and 0.2 % egg albumin (w/v)]. Tissues were ground with a high speed homogenizer (Tekmar company, Cincinatti, OH 45222)). The coating anti-BBLMV—IgG was diluted to 1 ug/ml in 0.5 M sodium-carbonate-bicarbonate buffer, pH 9.6 and added to an Immulon-l, 96-well flat bottomed polystyrene ELISA plate (Dynatech laboratories, Inc, Chantilly, VA 22021) at the rate of 200 ul per well. The plate was sealed in a plastic bag for incubation at this step and for all succeeding incubation steps. Plates were incubated at 37 C for 2 to 4 h. Plates were then rinsed three times in PBS containing 0.05 % Tween—20 (PBS tween) . Ground samples were added to the plate at the level of 200 ul/well and incubated overnight at 4 c. The plate was again rinsed three times with PBS-tween as above. The anti-BBLMV—IgG— alkaline phc ext am twv va di th ts 47 phosphatase conjugate was diluted to 1:800 (v/v) in extraction buffer, added to the ELISA plate at 200 ul/well, and incubated 2 to 4 h at 37 C. It was then rinsed in PBS— tween as above. The enzyme substrate, P-nitrophenyl phosphate (Sigma) was diluted to 1 mg/ml in substrate buffer (10 % diethanolamine, pH 9.8, in distilled water), and added to the plate. The reaction was allowed to proceed at room temperature for 30 to 60 min, depending on the rapidity and strength of the enzymatic reaction. Reaction values in wells were read spectrophotometrically at A410 nm with a model MR 590 micro- ELISA minireader (Dynatech Instruments, Santa Monica, CA). Test samples were considered positive for virus antigen if the A410 nm value of a sample was greater than the A410 nm value plus three standard deviations of the healthy leaves control samples (usually three per plate). Leaves from a BBLMV—infected blueberry bush from a commercial planting and a Chenopodium quinoa Willd. plant inoculated with BBLMV, were used as positive controls. For those bushes which were ELISA-positive, new samples were obtained from the woods. These samples were diluted 1:5 (w/v) in extraction buffer, ground with a mortar and pestle and rub-inoculated onto carborundum—dusted (320 meSh) C. quinoa plants in the fourth leaf stage. Plants were kept for about 1 mo in a greenhouse (22 to 30 C) under cool-white fluorescent light with a 15 hour day length. Leaf tissue was tes ver was bus the keg till my: hyi pl me to le 48 tested by ELISA for BBLMV. BBLMV-infected C. quinoa plants were used as positive controls (Figure 3). Taxonomic classification of wild blueberry bushes While sampling wild blueberry bushes in the woods, it was noted that different morphological types of blueberry bushes were present. As a result, during the spring of 1992, these wild plants were classified as to species using the key trials outlined by Vander Klyet (1988) that distinguish the Vaccinium species found in Southwestern Michigan (V. myrtilloides, V. angustifolium, V. corymbosum and their hybrids). Five different criteria were used to classify the plants : Leaf margin (entire or serrate), leaf size (small, medium or large), height (less than 30 cm, 30 to 60 cm, 60 to 90 cm, 90 to 120 cm or taller than 120 cm); twigs and leaves pubescent or glabrous; and, presence or absence of rhizomes. V. myrtilloides has pubescent leaves with an entire margin and shrubs that are less than 100 cm high and rhizomatous. V. angustifolium leaves are sharply serrate with shrubs 9 to 27 cm high and rhizomatous. Finally, V. corymbosum shrubs are crown forming and ussually more than 150 cm high. This taxonomic classification was later compared with the results obtained in the survey to determine if Vaccinium species other than V. corymbosum and V. australe were infected with BBLMV. 49 Figure 3. Chengpodium quinoa used as a BBLMV-indicator pan. 50 T. blueb in qu oonce colle durir plant threv half high blos anal affe Dob: app] var inv the Cer 196 an: Su. ne Sa 51 Nectar measurement To determine if bees were moving from commercial blueberry plants to wild blueberries because of differences in quantities of flower nectar production and sugar concentration, seven samples of twenty blossoms each were collected from wild plants and from commercial cultivars during the spring of 1992 in west-central Michigan. Wild plants were divided according their characteristics into three categories : Lowbush blueberry (V. angustifolium), half-high hybrids (V. corymbosum x V.angustifolium) , and highbush (V. corymbosum) blueberry. Only fully opened blossoms without signs of senesence were included in the analysis to minimize variation. The age of the blossoms affects nectar yield and sugar concentration (Brewer and Dobson, 1969c). Also, the blossoms were removed at approximately the same time from the plants to reduce variation caused by nectar changes during the day. Nectar was extracted from the flowers by placing inverted blossoms in microcentrifuge tubes and centrifuging them for 5 min at approximately 2,000 rpm in a clinical centrifuge (technique described by Brewer and Dobson, 1969c). Because in general, blueberry flowers produce small amounts of nectar, twenty blossoms were extracted successively to obtain enough for analysis. The amount of nectar extracted was determined using a 20 ul-pipetman. Six samples of twenty flowers were run for each type of wild blueber Sug extract equal t initial Water a was re: Tokyo 2 Suv analyz. Duncan move ; locatc Petfo: field 5 95) 1’111110 (Fiqu that 52 blueberry or commercial cultivar sampled. Sugar concentration was determined by diluting the extracted nectar with an amount of glass distilled water equal to two times the volume of nectar extracted. The initial volume of nectar was not sufficient for a reading. Water and nectar were mixed using a pipetman and the mixture was read using an Atago hand-held refractometer (Atago Co., Tokyo 173, Japan). Sugar concentration and nectar production data were analyzed using a standard analysis of variance procedure and Duncan's multiple range test for mean comparisons. Spread of BBLMV infected pollen Locations In order to establish how far BBLMV-infected pollen can move from an infection focus in a field to other fields located around it, at different distances, an experiment was performed during the spring of 1992. A large commercial field which contained BBLMV-infected bushes (approximately 5 %) was used as a source field. Five surrounding fields without infected bushes were selected for the experiment (Figure 4). Candidate fields were inspected for BBLMV. Any that were suspected as being infected were tested by ELISA 53 Figure 4. Location of the BBLMV source field (A) and the five surrounding fields without BBLMV- infected bushes from the experiment to determine spread of BBLMV-infected pollen between plantations in WestCentral Michigan. 54 LAKE MICHIGAN AIZelinka form Bil. uian farm CI Reender farm DI Rank form EIDe Vries form FIBrower form clmiles during differ Severa due to trap v pollel hive seven and t] and 7 1:5 ( Torte of pc Probe teSte PrSV: heal‘ nega‘ dilu 55 during the winter of 1992. Flower bud samples from 20 different suspect plants were taken each different field. Several fields were rejected for inclusion in the experiment due to presence of BBLMV-infected bushes. Pollen traps During the blossom period (50 % to full bloom), a pollen trap was attached to the hive entrance. The trap scraped the pollen basket from the bee's hind legs as they entered the hive (Figure 5). Seven pollen traps were attached to each of seven different hives at the five healthy field locations and the source field location. Pollen was collected at 48 and 72 hours after setting the pollen traps on the hives. Pollen analysis Each pollen sample was weighed (mg quantities), diluted 1:5 (w/v) in ELISA extraction buffer and then ground with a mortar and pestle. Those samples containing higher amounts of pollen were divided into sub—samples to increase the probability of finding infected pollen grains. These were tested by ELISA following the same procedure described previously. Pollen samples collected from ELISA-tested healthy and BBLMV-infected blueberry bushes were utilized as negative and positive control controls, respectively. To verify the results obtained by ELISA, pollen samples diluted 1:5 (w/v) in extraction buffer were ground and rub— 56 Figure 5. Pollen traps used to trap BBLMV-infected pollen or healthy pollen from bee hives. 57 inoculated < plants in t1 1 mo in a g: fluorescent tested by E quinoa plan negative co 58 inoculated onto carborundum dusted (320 mesh) C. quinoa plants in the fourth leaf stage. Plants were kept for about 1 mo in a greenhouse (22 to 30 C) under cool—white fluorescent light with a 15 hour day length. Leaf tissue was tested by ELISA for BBLMV. BBLMV infected and healthy C. quinoa plants respectively, were used as positive and negative controls. In situ In most observed (a over the e: grain (Fig1 was Possibi Dellen gra and Genera Particles In the 0f the 901 Surroundir “me sold the“ pol: Gold Secti(ms SOdium ph RESULTS In situ localization of viral coat protein in infected pollen grain and anthers. Immunogold labeling localization In most of the BBLMV-infected pollen grain sections observed (about 20 grids), the densest labeling was found over the exine (outer surface) and intine of the pollen grain (Figure 6). However, in some sections (4 grids), it was possible to observe some immunogold labeling within the pollen grain in the cytoplasm and nucleus of the vegetative and generative cells. Some of these sections show virus particles associated with the labeling (Figure 7). In the case of BBLMV—infected anthers (15 grids), most of the gold labeling was located in the membranes surrounding the immature pollen grains (Figure 8). Also, some gold labeling was observed in the intine and exine of these pollen grains (Figure 9). Immunogold labelling technique. Gold labelling was successful with pollen and anther sections etched for 60 min in 0.1 % sodium metaperiodate in sodium phosphate buffer (PH 7.2) and treated with BBLMV antiserum diluted 1:100 (v/v) in 1 % dehydrated milk. The 59 60 Figure 6. Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in the intine and exine of BBLMV-infected blueberry pollen grains. Infected pollen grains were fixed in 4 % (v/v) glutaraldehyde, post- fixed with 1 % (v/v) osmium tetroxide, dehydrated in ethanol and embedded in Spurr's low viscosity medium. Sections were etched in 0.1 % (w/v) sodium metaperiodate, labeled with a 1 : 100 (v/v) dilution of BBLMV antiserum followed by protein-A/15 nm gold. Sections were stained with uranyl acetate followed by lead citrate. Arrows ' indicate the 15 nm gold particles surrounding a virion. Bar represents 500 nm. Ezexine; 1: intine #1 > I k 62 Figure 7. Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in ’ BBLMV-infected blueberry pollen grain cell cytoplasm. Infected pollen grains were fixed . in 4 % (v/v)glutaraldehyde, post-fixed With 1 % (v/v) osmium tetroxide, dehydrated in _ ethanol and embedded in Spurr's low visc051ty medium. Sections were etched in 0.1 % (W/V) sodium metaperiodate, labeled with a 1 : 100 dilution (v/v) of BBLMV antiserum followed by protein-A/ls nm gold. Sections were stained with uranyl acetate followed by lead citrate. Bar represents 100 nm. Vzvirions; Va: vacuole. 64 Figure 8. Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV-infected blueberry anthers. Infected anthers were fixed in 4 % (v/v) glutaraldehyde, post-fixed with 1 % (v/v) osmium tetroxide, dehydrated in ethanol and embedded in Spurr's low viscosity medium. Sections were etched in 0.1 % (w/v) sodium metaperiodate, labeled with a 1 : 100 (v/V) dilution of BBLMV antiserum followed by protein-A/15 nm gold. Sections were stained with uranyl acetate followed by lead Citrate. [A Anther sacs [B] Epidermis [C] Connective tissue. Arrows indicate the 15 nm gold particles. Bar represents 500 nm. 66 Figure 9. Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in pollen grains within BBLMV-infected blueberry anthers. Infected anthers were fixed in 4 % (v/v) glutaraldehyde, post- fixed with 1 % (v/v) osmium tetroxide, dehydrated in ethanol and embedded in Spurr's low viscosity medium. Sections were etched in 0.1 % (w/v) sodium metaperiodate, labeled with a 1 : 100 (V/V) dilution of BBLMV antiserum followed by. protein-A/15 nm gold. Sections were stained with uranyl acetate followed by lead Citrate. [A] intine and exine, [B] cytoplasm. Arrows indicate the 15 nm geld particles. Bar represents 500 nm. EzeXine: I:1ntine. 68 various controls used were essential to provide confidence in the specificity of labeling. Virtually no labeling was observed when IgG from pre-immune serum was used (Figures 10 and 11), or when the antiserum step was omitted in staining the BBLMV-infected pollen grain and anther sections (Figure 12). Slight labeling was observed when sections of healthy pollen grains and anthers were treated with BBLMV—antiserum and gold (Figure 13). Determination of Vertical and Horizontal transmission of BBLMV via pollen in blueberry Vertical transmission The results for the analysis of fruit clusters in healthy plants pollinated with BBLMV-infected pollen are presented in tables 1, 2 and 3. From a total of 201 fruit clusters analyzed (Table 1), 15 became infected with the virus during the first pollination date (May, 1991), and six in the second pollination experiment (June, 1991) as is shown in Table 2. This represented about 10 % of the total number of fruit clusters analyzed (Table 3) on both dates. To determine if the virus was present in the seedlings from healthy plants pollinated with diseased pollen, seeds were germinated after a vernalization period. Because of mice and 69 Figure 10. Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV-infected blueberry pollen. grains. Infected pollen grains were fixed in 4 % (v/v) glutaraldehyde, post-fixed with 1 % (v/v) osmium tetroxide, dehydrated in . ethanol and embedded in Spurr's low ViSCOSity 'medium. Section were etched in 0.1 % (W/V) sodium meta eriodate, labeled with a 1 : 100 (v/v) dilution of pre-immune serum followed by protein-A/ls nm gold. Sections were stained with uranyl acetate followed by lead citrate.[A] Intine and exine. [B] Blueberry pollen grain cytoplasm. Bar represents 500 nm. E:exine, Izintine. 7O 71 Figure 11. Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV-infected anthers. Anthers were fixed in 4 % (v/v) glutaraldehyde, post fixed with 1 % (v/v) osmium tetroxide, dehydrated in ethanol and embedded in Spurr' 5 low viscosity medium. Sections were etched in 0.1 % (w/v) sodium metaperiodate, labeled with a 1 : 100 (V/V) dilution of pre—immune serum followed by protein-A/15 nm gold. Sections were stained with uranyl acetate followed by lead -citrate. [A] Epidermis [B] Anther sac. Bar represents 500 nm. :" ‘L x 2 an ‘ I sdl' m in. 72 Figure 12. 73 Control for immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in BBLMV-infected blueberry pollen grains and anthers. Sections were fixed in 4 % (v/v) glutaraldehyde, post-fixed with 1 % (v/v) osmium tetrox1de, dehydrated in ethanol and embedded in Spurr's low . viscosity medium. Sections were etched in 0.1 *% (w/v) sodium metaperiodate and treated with protein- A/15 nm gold only. Sections were stained with uranyl acetate followed by lead citrate. [A] Pollen grain [B] Anther. Bar represents 500 nm. E:exine; Izintine; Va: vacuole. 75 Figure 13. Immunogold localization of blueberry leaf mottle virus (BBLMV) coat protein in healthy blueberry pollen grains and anthers. Sections were fixed in 4 % (v/v) glutaraldehyde, post—fixed with 1 % (v/v) osmium tetroxide, dehydrated in ethanol and embedded in Spurr's low viscosity medium. Sections were etched in 0.1 % (w/v) sodium metaperiodate, labeled with a 1 : 100 (V/V) dilution of BBLMV antiserum followed by. protein-A/lS nm gold. Sections were stained with uranyl acetate followed by lead citrate. [A] Pollen grain [B] Anther. Bar represents 500 nm. E:exine; Izintine; Vazvacuole. 76 77 man Table 1. Fruit clusters analyzed by ELISA from healthy, potted 3 yr-old plants pollinated with pollen ___ from BBLMV-infected blueberry bushes. Ham Date of Plant Cluster number __T6ta1 pollination designation "‘ May, 1991 HXS 1 l,2,3,4,5,6,7,8,9,10,11 11 Hxs 2 1,2,3,4,5,6,7,8,9,10,11,12,13,15 14 HXS 3 1,2,3,7,8,9,10,11,12,13,14,15,16, 17,18,19,20,21,22,23 20 HXS 4 1,2,3,4,5,6,7,8,9,10,11,12,13 13 HXS 5 1,2,3,4,5,6,7,8,9,10,11 11 HXS 6 1,2,3,4,5,6,7,8,9,10,11,12,13 13 HXS 7 1,2,3,4,5,6,7,8 , 8 HXS 8 1,2,3,4,5,6,7,8,9,12,13 11 HXS 9 1:213:41516171819I1’or11112!13:14 1'4 Hxs io 1,2,3,4,5,6,7,9,10 9 HXS 11 i,2,3,4,5,6,7,8,9,io,11,12,13,14. 15 15 Hxs 12 1,2,3,4,5,6,7,8 8 subtotal of plants 1Dfi analyzed 12 147 ' June, 1991 HXS 1 1,2,3,4,5,6,77§,9,1o,11 11 ... HXS 2 1,2,3,4,5,6,7,8,9,10,ll 11 Hxs 3 l,2,3,4,5,6,7,9,10,11 1° HXS 4 1,2,3,4,5,6,7,8,9 9 HXS 5 1,2,3,4,5,5,7,3,9 9 HXS 6 1,2,3,4,5,6,7,8,9,10,11,12,13,14 14 subtotal of plants analyzed 5 54 Total plants analyzed 13 201 Table 2. Number of fruit clusters infected with BBLMV on healthy potted 3 yr-old bushes pollinated with BBLMV-infected pollen, at different harvest dates. Harvest date of Date of ELISA positive clusters clusters pollination 9-23-91 June, 1991 AAAAHMS2-5 9-17-91 May, 1991 HXSS-lO, HXSll-4, HXSQ-lo, HXSl-lO June, 1991 HXS4-3, HXS3-2, HX56-4 9-10-91 May, 1991 HXS3-ll 8-15-91 May, 1991 HXSG-S 7-15-91 May, 1991 HXSll-7, HXSlO-l, HXS9-4, HXS4-6 June, 1991 HXSl-l, HXS6-2 7-05-91 May, 1991 HXSll-Z 6-27-91 May, 1991 HXSZ-Z, HXS4-4, HXSS-l, HXS6-2 Total number of infected fruit clusters May, . June, 1991 1991 Tabl (12 EXpe (8 r We: 79 Table 3. Summary of results from pollination of healthy, 3 yr-old potted blueberry bushes with BB BLMV- infected pollen. Total No. of’fruit Fruit clusters ELISA—pOSitive clusters analyzed number % Experiment 1 147 15 10.2 (12 bushes) Experiment 2 54 6 11.1 (8 bushes) Total 201 21 10.4 fungal 1 sufficie ELISA-n1 decide . Eight 51 represe seeds 5 The healthy negativ seedlir The transmi Commerc rate 0: low. T] infect detect cvs. E 0'3! 0 with 1 daten (hOri; 80 fungal problems, only eight seedlings survived and grew sufficiently to be analyzed by ELISA. All of them were ELISA-negative for BBLMV. From these results we cannot decide if BBLMV is being transmitted vertically via pollen. Eight seedlings is not enough if we consider that they represent only a small percentage of the total number of seeds sown. The healthy control plants which were pollinated with healthy pollen, yielded clusters which were all ELISA- negative for BBLMV. The same results were obtained when seedlings from these plants were analyzed. The results from the experiment to determine rates of transmission of BBLMV via pollen in five different commercial blueberry cultivars are presented in Table 4. The rate of virus transmission to fruit clusters was generally low. The cv. Spartan presented the highest level of infection (2.8 %). No BBLMV-infected fruit clusters were detected from cv. Blueray (only 11 clusters tested). The cvs. Elliot, Blue crop and Jersey had levels of infection of 0.3, 0.4 and 0.8 % respectively. Horizontal transmission After a dormancy period, the healthy plants pollinated with BBLMV-infected pollen were analyzed by ELISA to determine if the virus had moved into the mother plants (horizontal transmission) The results for the analysis of Ta 9| 81 Table 4. Percentage of BBLMV-infected clusters from five different blueberry cultivars pollinated with BBLMV-infected pollen. Cultivar Total number of ELISA-positive fruit clusters fruit clusters analyzed number % Spartan 179 5 2.8 Elliot 287 1 0.3 Bluecrop 217 1 0.4 Jersey 126 1 0.8 Blueray 11 0 0.0 roots, ] the firs with di: to ELIS. just in plant. was pre mother by ELIE and 11 that h; in 199 the vi In or ho: diffe] Blue < BBLMV P0111 Perio not h 82 roots, leaves and flowers are presented in Table 5. During the first year, about 78 % of the healthy plants pollinated with diseased pollen had infected fruit clusters according to ELISA tests. However, it was not clear if the virus was just in the seeds or had already moved into the mother plant. Plants retested by ELISA in 1992, showed that BBLMV was present in roots, leaves and flowers. Almost 40 % of the mother plants tested BBLMV-positive when roots were analyzed by ELISA. Virus was located in leaves and flowers in 22.2 % and 11.1 % of the plants, respectively. When mother plants that had BBLMV-negative fruit clusters in 1991 were retested in 1992 , 11.1 % of them, had become positive. In this case the virus was detected only in the roots. In the other experiment designed to determine vertical or horizontal transmission of BBLMV via pollen among different commercial blueberry cultivars (Spartan, Elliot, Blue crop, Jersey, Blueray), it was not possible detect BBLMV in the roots or leaves of those healthy plants pollinated with BBLMV—infected pollen after the dormancy period. Samples were taken several times, but virus could not be detected. Tab Total p] fruit c] Plants 1 a dormer Total p] Cluster: Plants 1 a dormal 83 Table 5. Analysis of roots, blossoms and leaves from 18 healthy, potted 4 yr-old blueberry plants cv. Jersey pollinated with BBLMV-infected pollen in 1991. ELISA p051tive samples fifimBer % of total plants analyzed Total plants with BBLMV-infected fruit clusters in 1991 14 77.7 Plants retested in 1992, after a dormancy period with BBLMV-infected : roots 7 38.8 leaves 4 22.2 flowers 2 11.1 Total plants with fruit clusters negative for BBLMV in 1991 4 22.3 Plants retested in 1992, after a dormancy period, with BBLMV-infected : roots 2 11.1 Ta differ to dei growix contai appeal found BBLMV- distal diffEI 7, 8 a 11 an< the N< FrUit] Wild ] from . Junct, (West Side) the b' from . T .iki 84 Field Survey BBLMV—infected fields Tables 6 to 11 present the results obtained from six different locations in West-central Michigan from the survey to determine the presence of BBLMV in wild blueberries growing in woods surrounding commercial blueberry fields containing BBLMV—infected bushes. From these results it appears that BBLMV is present in some of the wild bushes found in the woods. However, the highest percentage of BBLMV-infected wild bushes was not always found at the same distance from the border of the commercial field at the six different locations. If one compares for example, Tables 6, 7, 8 and 9 and Figures 14, 15, 16 and 17 with Tables 10 and 11 and Figures 18 and 19, it can be seen that in the case of the North Muskegon location (East and South sides) and the Fruitport location (East side), most of the BBLMV-infected wild blueberries were located in the woods at 100 meters from the border. Conversely, in places like the Grand Junction location (North side), the Breedsville location (West and North sides), the Grand Haven-1 location (North side) and the Grand Haven-2 location (South side), most of the bushes infected with BBLMV were located only 5 meters from the commercial field. These two different patterns of virus distribution in Tab ‘ Locati l woc relati commer fie] Sou Table 6. 85 Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containin BBLMV-infected bushes. Grand Junction, Van Buren County (Holmquist Farm). C. quinoa Location of’ Distance from Total ELISA _ ' woods the border bushes positive pOSitive relative to sampled commercial number i number field South 5 m 7 0 0 O O 50 m 61 0 0 o o 100 m 25 0 0 O 0 Subtotal 93 0 0 0 0 North 5 m 40 2 5.0 0 O 50 m 35 2 5.7 O O 100 m 25 0 0.0 0 0 Subtotal 100 4 4.0 O 0 East 5 5 0 0 0 0 a 50 m 0 - - - — 100 m 0 - - - - Subtotal 5 0 0 0 0 Total 198 4 2 0 0 a No wild bushes available for sampling Table 7. Location woods relative ' commercia field West 86 Table 7. Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Breedsv111e, Van Buren county (Crawford Farm). Location of Distance from Total ELISA C. quinoa woods the border bushes positive positive relative to sampled commercial field number E n er West 5 m 29 1 3.4 1 3.4 50 m 33 2 6.1 1 3.0 100 m 39 2 5.1 0 0.0 Subtotal 101 5 4.9 2 1.9 North 5 m 93 2 2.2 2 2.2 15 m 6 0 0.0 0 0.0 50 m 1 0 0.0 O 0.0 SEBtotal 100 2 2.0 2 2.0 Total 201 7 3.5 4 1.9 Table 8 . Location woods relative commercia field SOUEF: Subtotal West Wot-a; W No wil 87 Table 8. Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Grand Haven-1, Ottawa county (Zelinka Farm). Location of Distance from Total ELISA C. inoa woods the border bushes positive positive relative to sampled commercial n er n er field South 5 m 99 7 7.1 4 4.0 50 m 0 - - - - 100 m 0 - - - - Subtotal 99 7 7.1 4 4.0 West 5 m 100 8 8.0 1 1.0 50 m 0 - - - - 100 m 0 - - - - SfiBtotal 100 8 8.0 1 1.0 North 5 m 50 7 14.0 3 6.0 50 m 24 2 8.3 0 0.0 100 m 26 1 3.8 O 0.0 Subtotal 100 10 10.0 3 3.0 Total 299 25 8.4 8 2.7 a I No wild bushes available for sampling Table 9. Location woods relative commercia field 88 Table 9. Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Grand Haven-2, Ottawa county (Novak Farm). Location of Distance from Total ELISA C. quinoa woods the border bushes positive positive relative to sampled commercial number number field North 5 m 116 6 5.2 0 0 a 50 m 0 - - - - 100 m 0 - - - - Subtotal 116 6 5.2 0 0 South 5 m 72 7 9.7 2 2.8 50 m 0 - - - - 100 m 28 0 0.0 0 0.0 SfiBtotal 100 7 7.0 2 2.0 East 5 m 33 4 12.1 0 0.0 50 m 0 - - - - 100 m 0 - - - - Subtotal 33 4 12.1 0 0.0 Total 249 17 6.8 2 0.8 a No wild bushes available for sampling. Table 10 relative t. commercial field East Sfibtota. South U) 7157?? 03. RH ta. 'TOTal‘ W Subto 7 No wi: 89 Table 10. Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. Fruitport, Muskegon county (Dirkse Farm). Location of Distance from Total ELISA C. quinoa woods the border bushes positive positive relative to sampled commercial n er n er field East 5 m 59 10 16.9 4 6.8 50 m 36 7 19.4 3 8.3 100 m 30 10 33.3 3 10.0 Sfibtotal 125 27 21.6 10 8.0 South 5 m 64 2 3.1 0 0.0 a 50 m 0 - - - - 100 m 36 1 2.8 0 0.0 Subtotal 100 3 3.0 0 0.0 West 5 m 56 9 16.1 0 0.0 50 m 17 3 17.6 0 0.0 100 m 29 3 10.3 0 0.0 Subtotal 102 15 14.7 0 0.0 Total 327 45 13.8 10 3.1 a No wild bushes available for sampling. Table 11 W woods relative commercia field 90 Table 11. Percentage of BBLMV-infected wild blueberry bushes in woods surrounding a commercial field containing BBLMV-infected bushes. North Muskegon, Muskegon county (Paul Bros. Farm). Location of Distance from Total ELISA C. quinoa woods the border bushes positive positive relative to sampled commercial number number field East 5 m 32 5 15.6 0 0.0 50 m 36 3 8.3 0 0.0 100 m 32 18 56.2 1 3.1 Subtotal 100 26 26.0 1 1.0 South 5 m 65 10 15.4 2 3.1 50 m 25 4 16.0 0 0.0 100 m 30 8 26.7 2 6.7 SHBtotai 120 22 18.3 3 3.3 West 5 m 100 4.0 2 2.0 50 m 20 1 5.0 1 5.0 100 m 15 0.0 0 0.0 Subtotal 135 5 3.7 3 2.2 Total 355 53 14.9 8 2.3 91 Figure 14. Distribution of BBLMV-infected blueberry bushes in woods adjacent to a commercial planting containing BBLMV-infected bushes. grand Junction, Van Buren county. (Holmquist arm). 92 5 50 100 5 50 100 South East _ 6 5 A.» 3 m; ._I 8525 w>£moo Ewoo .._.......,. ...-1 ...:r..:.. 1.71.4. ......L..a..:...:..... ,............». .31.: ...... .. I. . . ....