Microb Ecol (1994) 27:19-26

MICROBIAL ECOLOGY © 1994Springer-VerlagNewYorkInc.

Transmission Via Plants of an Insect Pathogenic Bacterium That Does Not Multiply or Move in Plants A.H. Purcell, 1 K.G. Suslow, 2 M. Klein 3 1Department of Environmental Science, Policy, and Management, University of California, 201 Wellman, Berkeley, CA 94720, USA; 2Plant Gene Expression Center, USDA Western Regional Research Lab, Albany, CA 94710, USA; 3Department of Entomology, Volcani Center, Bet Dagan 50 250, Israel Received: 22 June 1993; Revised: 22 October 1993

Abstract. A bacterial parasite (designated as BEV) of the leafhopper Euscelidius variegatus, which is passed transovarially to offspring, was transmitted from insect to insect via feeding of the insects in plants. The rate of bacterial infection of leafhoppers fed upon plants that had previously been exposed to BEV-infected leafhoppers declined with an increase in the time that infected leafhoppers had been off rye grass. Transmission of BEV also occurred on sugar beet and barley but not celery. The bacterium was also transmitted to and acquired from membrane-encased artificial diets. There was no evidence that the bacterium was transmitted via plant surfaces, but transmission and direct culture assays from plants indicated that the bacterium did not multiply or move within plants. This parasite-host relationship may represent a primitive stage in either the evolution of intracellular symbiosis with its insect host or to alternative parasitization of plant and insect hosts via insect transmission, as is the case for insect-vectored plant pathogens. Introduction Bacteria and other microparasites of both plants and insects use their insect hosts as vectors [15] to infect plants. Rarely are such prokaryotes transmitted vertically (parent to offspring), for example, transovarially [16]. In contrast, bacterial intracellular symbionts that are necessary for the reproduction of homopterous insects are transmitted only transovarially [3, 7]. An unclassified enteric bacterium, designated as BEV [14], has been placed phylogenetically in the gamma subdivision of the proteobacteria [4]. This intracellular parasite [5, 13] is efficiently transmitted transovarially in its definitive leafhopper host, Euscelidius variegatus [14]. But because of its pathogenicity to

Correspondence to: A.H. Purcell.

20

A.H. Purcell et al.

E. variegatus, vertical transmission alone should not sustain for an indefinite number of generations the survival of a microparasite such as BEV [8]. Therefore we examined the possibility that BEV could be transmitted horizontally among leafhoppers and whether the bacterium could multiply on or in plants. Materials and Methods

Insects and Plants Colonies of the leafhopper E. variegatus that were free of BEV (designated "healthy") were reared on barley (Hordeum vulgare) seedlings. Insects were transferred with an aspirator tube. Adults were caged to oviposit for 1 week on plants about 20 cm tall in 15-cm diameter pots and then removed from plants. Nymphal leafhoppers emerged 1-2 weeks after removing the adults and were transferred weekly to fresh barley plants. Colonies of E. variegatus that were congenitally infected with BEV were maintained in the same fashion except that transfers were made approximately every 2 weeks or sooner if plant condition deteriorated.

Transmission Experiments To test for transmission of BEV from infected to healthy leafhoppers via plants, we caged (10-cm diameter, 25-cm tall plastic cylinder) groups of 20 BEV-infected E. variegatus on a 10-cm pot containing 4 rye grass (Lolium perennae) plants. Four to six groups (replicates) were used for each plant transmission experiment. After 7 days, we removed the infected leafhoppers and replaced them with 10-20 uninfected ones, either immediately or after various intervals (Fig. 1). The healthy replacements remained on these "inoculated" plants for at least 1 week and then were transferred to fresh rye grass plants. Identical experiments were repeated using 'Atlas' barley or 1 plant per pot of celery (Apium graveolens, cultivar 'Tall Utah 72-50') or sugar beet (Beta vulgaris) instead of rye grass. In other experiments we caged BEV-infected E. variegatus on a single rye grass leaf surrounded by a tubular clear plastic cage (7.5 cm diameter), closed at the basal end by a 1-cm thick disk of foam rubber, and covered at the distal end with an aluminum foil diaphragm. The foil and foam rubber closures had a slit to allow a grass blade into the cage. A second identical cage was fastened onto the foil covering of the basal cage with rubber bands. BEV-infected E. variegatus were introduced into one cage and healthy E. variegatus into the opposite cage so as to allow BEV-infected leafhoppers to feed on the basal 2-3 cm of a leaf, and uninfected leafhoppers to feed distally on the same leaf, separated only by a foil barrier. The positions of the BEV-infected and healthy leafhoppers were reversed in an equal number of replicate cages. To evaluate leafhopper infection from plant surfaces in two experiments, we sprayed rye grass plants with 109 colony forming units (cfu) ml--I of BEV cells suspended in 0.01 u neutral phosphate buffered saline (0.85%). The plants were placed in a darkened room for 1 h to dry, then leafhoppers were caged on the plants overnight. The leafhopper acquisition of BEV was attempted by feeding noninfected E. variegatus adults on 50 p,1 of feeding solution E-2 [1] with a suspension of BEV sandwiched between two pieces of ethanol-sterilized, thinly-stretched Parafilm M (America National Can, Greenwich, CT) and presented to leafhoppers at one end of a plastic tube. The same method was used in attempts to transmit BEV with infected leafhoppers except that sterile feeding solutions were used and the sachet contents were extracted aseptically with a 27-gauge hypodermic needle and syringe and plated directly onto purple agar medium. To assay for BEV infection of originally healthy leafhoppers, we attempted isolations of BEV after 3-4 weeks following their initial exposure to "inoculated" plants.

Bacterial Culture and Identification To confirm BEV infection in leafhoppers, we isolated BEV on Difco (Difco Labs., Detroit, MI) purple agar [13] acidified with 1 N hydrochloric acid to pH 6.3 and incubated in total darkness (light kills

Bacterial Transmission Via Plants

21

BEV) at 24-26°C. Positive cultures were confirmed by colony growth after 4-6 days or longer, colony appearance, cell morphology, and inability to grow in light. Control leafhoppers from healthy colonies were always negative for BEV. To estimate titers of viable (cultivable) BEV within plants, we surface sterilized rye grass leaf segments that had been exposed to BEV-infective E. variegatus by immersion for 1 min in 70% ethanol, 2 rain in 1.25% sodium hypochlorite, then 3 successive rinses in sterile water. We then aseptically cut a 1-cm length from the leaf segment and pulverized it at high speed in 2 ml of neutral 0.01 phosphate buffer with a Brinkrnann Polytron (Westhury, NY) tissue homogenizer. Concentrations of viable BEV were estimated from three tenfold dilutions of the pulverized leaf plated onto purple agar. We measured concentrations of viable BEV per centimeter length of rye grass leaf to facilitate rapid processing of culture attempts because of the sensitivity of BEV to light. In addition, the rye grass leaves were relatively uniform in width where feeding cages were placed, and estimates of bacterial concentration per unit length of leaf could be readily interpreted relative to analyses of systemic spread within the plant, because the vascular tissues of the grass are parallel to the leaf axis. To increase the concentration of BEV within leaves in some experiments, we restricted the feeding of BEV-infective E. variegatus on a portion of a leaf by caging 3 insects in 2.5-cm leaf cages on a single leaf.

Results The percentage of E. variegatus that acquired BEV from "inoculated" rye grass plants dropped steadily as the time interval between removing BEV-diseased leafhoppers and replacing them with healthy ones was increased (Fig. 1). This was the opposite of what would be expected if BEV multiplied in the plants. Healthy leafhoppers acquired BEV from similarly inoculated sugar beet and barley plants but not from celery (Table 1). We estimated by culture assays the number of viable BEV cells in plants at successive time intervals after feeding by BEV-infective E. variegatus. Our first attempts to isolate BEV from leaf segments from our transmission experiments on rye grass were unsuccessful. To increase the initial densities of BEV in leaves for culture assays, we concentrated the feeding of 3 leafhoppers on 2.5-cm lengths of rye grass leaf. After 4 days of exposure to feeding leafhoppers, the numbers of viable BEV recovered from surface-sterilized, homogenized leaf segments ranged from less than 20 (detection threshold) to 1.8 × 103 cfu cm -~ and averaged 6.6 x 102 cfu c m - ~ of leaf (SE = 1.5 x 102, n = 11). After 4 weeks, the number of viable BEV had declined to an average of 2.0 x 102 cfu cm -1 (SE = 6 x 10, n = 8; Student's t = 2.1; P < 0.05). No BEV were recovered from 12 adjacent segments of the same leaves that had been fed upon by BEV-infected leafhoppers and from which BEV was recovered, suggesting that BEV did not move systemically from leafhopper feeding sites. In a repeat of this experiment using one leafhopper for one d a y ' s exposure, we recovered BEV from 5 of 12 plants at an average density of 5 × 10 (SE = 9) cfu cm -1 after one day. No BEV were recovered from 12 leaf segments adjacent to feeding cages. After 18 days, we recovered BEV from 4 of 14 plants at an average density of 3 x 10 (SE = 10; t = 1.24; P > 0.1) cfu cm -1. No BEV were recovered from 10 leaf segments adjacent to feeding cages. All feeding-inoculated leaves in the second experiment had senesced and completely dried by 26 days after inoculation. BEV was not recovered after 8 or 15 weeks or from control plants exposed to uninfected E. variegatus. We did not observe any differences in the appearance of plants exposed to feeding BEV-infected leafhoppers compared to those fed on by healthy leafhoppers.

22

A.H. Purcell et al. 10090-

'13 •~

BO-

'~ 702 o 60t,,b

.c_ 50X

I

0

II) 40t~ "~ t'-

30-

~

20-

(310 '

~

'

I

20

'

'

'

I

40

'

'

'

I

60

'

'

,

'Vl

80

,

,

i,

100

,

,i

120

,i,

i,O,

140

160

D a y s inoculating i n s e c t s off plant 1. Percentages of leafhoppers (Euscelidius variegatus) that became infected with the BEV bacterium at various intervals after they replaced BEV-infected leafhoppers on rye grass. Replacement leafhoppers surviving 3-6 weeks later were assayed for infection with BEV by culturing buffer-diluted hemolymph spread onto solid medium (Difco purple agar acidified to pH 6.3). Each datum point represents the average overall infection rate with BEV in a separate experiment in which 3-6 replicates of 20 BEV-infected E. variegatus were caged on rye grass plants for 4--7 days, removed, and replaced after various intervals (horizontal axis) by uninfected late instar to young adult E. variegatus for 1 week. Vertical lines from each datum point represent the range among replicates. Curved lines represent regression equations (least-squares criterion) for the two data sets, using an exponential model (Deltagraph Professional, DeltaPoint, Inc., Monterey, CA). Three treatments (circles) were run simultaneously (% transmission = y = 44.5 e -°'56x, where x = days infective insects off plant before replacement; r 2 = 0.988, P < 0.05), the remainder (diamonds) at various other times (y = 106.6 e-°szTx; P = 0.891, P < 0.01). Fig.

Additional transmission experiments also indicated that BEV did not move systemically within rye grass. Groups of 3--4 E. variegatus nymphs free of BEV were caged on 2-3 cm of a rye grass leaf inserted into a slit in an aluminum foil diaphragm that served as a barrier to 3 BEV-infective E. variegatus caged on the opposite side of the same leaf for 7 days. None of 17 surviving leafhoppers that had been confined on the basal side of the leaf opposite BEV-infected leafhoppers nor 13 uninfected leafhoppers that had been confined on the side of the leaf that was distal from BEV-infected leafhoppers became infected with BEV within 4 weeks. In a repeat of this experiment in which 5 leafhoppers per cage were used, none of 9 surviving leafhoppers that had been confined on the basal side of the leaf opposite BEV-infected leafhoppers nor 8 uninfected leafhoppers that had been confined on the side of the leaf that was distal from BEV-infected leafhoppers became infected

Bacterial TransmissionVia Plants

23

Table 1. Transmissionof BEV to the leafhopperEuscelidius variegatusvia rye grass, celery, sugar beet, and barley. Plant Celery Sugar beet Barley Rye grass (positivecontrol)

Number of insects tested 20 7 16 10

Number positive for BEV 0 4 l0 9

Percentage positive 0 57 63 90

with BEV within 4 weeks, but all of 5 surviving healthy nymphs confined with the BEV-infected adults as positive controls became infected. In experiments in which artificial feeding solutions were used, we found that E. variegatus were infected by BEV after feeding on suspensions of BEV. After a 16 h access to BEV suspensions (6 x 106 cfu ml-1), 3 of 11 surviving E. variegams became infected with BEV. After 6 h of access to 108 cfu ml -x , 2 of 8 became infected. Infected leafhoppers also transmitted BEV to sterile feeding solutions. In three separate experiments, we individually fed BEV-infected E. variegatus on sachets of sterile diet for 19-23 h and then plated the solutions onto solid media. We recovered BEV from 6 of 24, 23 of 44, and 9 of 20 sachets. Healthy control insects did not transmit BEV. The viability of BEV in the feeding solutions was short. The concentrations of colony forming units, as estimated by dilution plating, decreased by 90-95% during 7-h exposures within a feeding membrane in separate experiments at temperatures of 24-26°C. The poor survival (range 23% to 52%) of leafhoppers kept overnight on feeding solutions and our direct observations of the insects' feeding while on artificial diets suggested that this method was not an accurate appraisal of bacterial acquisition from or inoculation of plants by E. variegatus. But the experiments clearly illustrated that BEV could be acquired by feeding and that infected E. variegatus could introduce the bacterium by feeding through a membrane. Leafhoppers might be expected to acquire BEV from plant surfaces, as in an oral-fecal mode of transmission [18]. But we found no evidence for this in two experiments where we sprayed BEV onto plants, placed healthy insects on the plants, and then tried to recover BEV from the exposed insects. Suspensions of BEV sprayed onto rye grass plants failed to infect E. variegatus placed onto the plants. BEV was not cultured from any of the 38 E. variegatus that had survived for 30 or more days following exposure to foliar sprays of BEV, but was isolated from all 4 congenitally infected leafhoppers (positive controls) cultured on the same batch of medium. We presume that poor survival of BEV outside of plants at least partly explains this observation. Discussion To our knowledge, this is the first report of transmission of bacteria among insects via plant intermediates where the bacteria neither multiply nor move systemically within plants and where the bacteria are not transmitted by the insects from plant surfaces but must be acquired from within plant tissues. This further implies that

24

A.H. Purcellet al.

BEV is very efficient in infecting leafhoppers after feeding uptake, in view of the small numbers of bacteria (hundreds per centimeter) deposited even when leafhopper feeding was highly concentrated. Viruses that replicate in aphids and can be transmitted via plants without replication in plants have been reported, but the viruses are apparently systemic within plants [10]. The intraplant multiplication and systemic movement of leafhopper-vectored plant pathogenic prokaryotes increase the likelihood of vectors acquiring these microbes by amplifying their numbers and the space that the pathogens occupy within plants. Multiplication and spread within plants also provides a reservoir for the pathogen when vectors are scarce or inactive [12]. Our findings of BEV transmission via plants suggest new possibilities for the insect transmission of some plant pathogenic bacteria that do not cause disease in some plant species that are "host plants" based on vector acquisition of the pathogen. For example, bacteria that cause Stewart's wilt of maize [11] and Pierce's disease of grapevines [9] have symptomless plant hosts from which potential vectors can acquire these plant pathogens. Whether these and other bacteria (including nonpathogens) that are transmitted by sucking insects via plants in which the bacteria neither multiply nor move systemically remains to be investigated. Our findings demonstrate that the consistent appearance within insects of intracellular prokaryotes that are transovarially transmitted should not be assumed to indicate an obligate or mutualistic relationship or that such parasites are transmitted only transovarially. In such cases, the possibility of horizontal transmission, for example via plants, should be examined where possible. Transmission via plants of bacteria similar to BEV might explain some of the "facultative" or "secondary symbionts" or "rickettsialike organisms" reported in leafhoppers [5] and other homopterans [3]. The "secondary symbiont" of the pea aphid [17] and of a whitefly [6] are phylogenetically near-relatives of BEV [4]. Molecular data suggest that aphid endosymbionts have coevolved for at least 250 million years [2]. Similar comparisons of sequences from 16S ribosomal genes from BEV relative to other proteobacteria imply a relatively more recent divergence of BEV from enteric bacteria such as Escherichia coli. Leafhoppers (Cicadellidae) provide numerous examples of vector insects with both complex plant pathogen transmission mechanisms (propagative, circulative, vector-specific) and complex microbial symbioses (obligate for host reproduction, distinct mycetome, transovarial transmission). The generalized distribution of BEV in leafhopper hemolymph and intracellularly in a variety of tissues [5], rather than restriction to a special organ such as a mycetome [3], is further evidence that the association of BEV with its insect host is recent compared to mycetomal endosymbionts [7]. Nevertheless, BEV appears to have specialized sufficiently to colonize E. variegatus' gut epithelium [5], to invade the body cavity, salivary glands, and other tissues of its insect host, and to be transmitted transovarially [ 14]. We propose that BEV represents a primitive evolutionary stage toward becoming either a leafhopper-transmitted plant pathogen or--more likely--an endosymbiont of its leafhopper host. One of the earliest stages in this evolution would be an oral-fecal transmission route, followed by intracellular, enteric infections of the insect host. If this has been the case for BEV, it has already lost its ability to survive outside of plants and insect hosts. For transmission via plants to become a more dominant feature of its life cycle, BEV would have to adapt to multiply and move

Bacterial Transmission Via Plants

25

within plant tissues. This might lead to its becoming a plant pathogen. For transovarial transmission alone to perpetuate BEV in nature, BEV must evolve towards benefiting its leafhopper host. The most damaging effect of BEV on the fitness of its insect host is reduced fecundity [13]. Less virulent variants of BEV might be rapidly selected, but it is difficult to predict how or if beneficial strains might arise. For strains that benefit their hosts, horizontal transmission via plants should be unnecessary [8] and rapidly lost. The high frequency of endosymbioses of prokaryotes with leafhoppers [3, 7]---often with multiple types of bacterial symbionts found in a single leafhopper--may reflect the relative success of gut-invasive, enteric bacteria in exploiting established physiological mechanisms in leafhoppers that now insure the transmission of long-associated symbionts. Molecular phylogenetic studies of leafhopper symbionts might be used to test the hypothesis that the symbionts evolved polyphyletically from a proteobacterial lineage.

Acknowledgments. We thank Jonathan Dariyanani, Brad Marder, Reggie Martinez, and especially Stuart Saunders for technical assistance.

References 1. Alivizatos AS (1982) Feeding behavior of the spiroplasma vectors Dalbulus maidis and Euscelidius variegatus in vivo and in vitro. Ann Inst Phytopathol Benaki 13:128-144 2. Baumann P, Munson MA, Lai C-Y, Clark MA, Baumann L, Moran N, Campbell BC (1993) Origin and properties of bacterial endosymbionts of aphids, whiteflies, and mealybugs. ASM News 59:21-24 3. Buchner P (1965) Endosymbiosis of animals with plant microorganisms. Wiley-Interscience, New York 4. Campbell BC, Purcell AH (1992) Phylogenetic affiliation of BEV, a bacterial parasite of the leafhopper Euscelidius variegatus, on the basis of 16S rDNA sequences. Curr Microbio126:37-41 5. Cheung WWK, Purcell AH (1993) Ultrastructure of the digestive system of the leafhopper Euscelidius variegatus Kirshbaum (Homoptera: Cicadellidae), with and without congenital bacterial infections. Int J Insect Morphol Embryol 22:49-61 6. Clark MA, Baumann L, Munson MA, Baumann P, Campbell BC, Duffus JE, Osborne LS, Moran NA (1992) The eubacterial endosymbionts of whiteflies (Homoptera: Aleyrodoidea) constitute a lineage distinct from the endosymbionts of aphids and mealybugs. Curt Microbio125:119-123 7. Douglas AE (1989) Mycetocyte symbiosis in insects. Biol Rev Camb Philos Soc 64:409-434 8. Fine PF (1975) Vectors and vertical transmission: an epidemiological perspective. Ann NY Acad Sci 266:173-194 9. Freitag JH (1951) Host range of Pierce's disease virus of grapes as determined by insect transmission. Phytopathology 41:920-934 10. Gildow FE, D'Arcy CJ (1988) Barley and oats as reservoirs for an aphid virus and the influence on barley yellow dwarf virus transmission. Phytopathology 78:811-816 11. Poos FW (1939) Host plants harboring Aplanobacter stewartii without showing external symptoms after inoculation by Chaetocnema pulicaria. J Econ Entomol 32:881-882 12. Purcell AH (1982) Evolution of the vector relationship. In: Mount MS, Lacy GH (eds) Phytopathogenic prokaryotes, vol. 1. Academic Press, New York, pp 121-156 13. Purcell AH, Suslow KG (1987) Pathogenicity and effects on transmission of a mycoplasmalike organism of a transovafially infective bacterium on the leafhopper Euscelidius variegatus (Homoptera: Cicadellidae). J Invert Pathol 50:285-290 14. Purcell AH, Steiner T, Bov6 JM (1986) In vitro isolation of a transovarially transmitted bacterium from the leafhopper Euscelidius variegatus (Hemiptera: Cicadellidae). J Invert Patho148:66-73

26

A.H. Purcell et al.

15. Rosenberger DA (1982) Fastidious prokaryotes: epidemiology of the hidden pathogens. In: Mount MS, Lacy GH (eds) Phytopathogenic prokaryotes, vol. 2. Academic Press, New York, pp 72-100 16. Sinha RC (1981) Vertical transmission of plant pathogens. In: Mckelvey JJ, Eldridge BF (eds) Vectors of disease agents. Praeger, New York, pp 109-121 17. Unterrnan BM, Baumann P, McLean DL (1989) Pea aphid symbiont relationships as determined by analysis of 16S rRNAs. J Bacteriol 171:2970-2974 18. Whitcomb RF, Shapiro M, Richardson J (1966) An Erwinia-like bacterium pathogenic to leafhoppers. J Invert Pathol 8:299-307

Transmission via plants of an insect pathogenic bacterium that does not multiply or move in plants.

A bacterial parasite (designated as BEV) of the leafhopper Euscelidius variegatus, which is passed transovarially to offspring, was transmitted from i...
505KB Sizes 0 Downloads 0 Views