Southeastern Naturalist 441 N.E. Harms and J.F. Shearer 22001199 SOUTHEASTERN NATURALIST 1V8o(3l.) :1484,1 N–4o5. 03 Spatial and Temporal Variability of the Alligatorweed pathogen, Alternaria alternantherae, in Louisiana Nathan E. Harms1,* and Judy F. Shearer2 Abstract - Alligatorweed leaf spot is a disease of invasive Alternanthera philoxeroides (Alligatorweed) in the southern US, caused by Alternaria alternantherae. However, little is known about when or where this pathogen naturally occurs. To better understand this species’ life history, we examined temporal (every 2–3 weeks) and spatial (latitudinal) patterns of A. alternantherae occurrence at sites in Louisiana for 2 y. Pathogen presence reflected clear within-year temporal and spatial patterns. Overall, the percentage of leaves infected with A. alternantherae was low during spring each year (0–20% infected) but increased throughout summer (maximum of 50% infected), and plants in northern sites had lower frequency of infection relative to southern sites until later in the year (late summer/early fall) but only in 1 of the 2 years of our study. The mean proportion of leaves infected with A. alternantherae declined with latitude both years (P = 0.01) and variability increased with latitude (P = 0.04), a pattern suggestive of range limitation in northern areas. We estimate a northern distributional limit of 34°N for A. alternantherae in Louisiana, but Alligatorweed occurs farther north. Although we did not directly examine disease impacts to Alligatorweed during the study, they may be greatest in southern areas, where the pathogen is more common early and throughout the growing season, and thus may be less likely to provide control in northern infestations of the invasive Alligatorweed. Introduction Biological control of the aquatic invasive plant Alternanthera philoxeroides (Mart.) Griseb. (Amaranthaceae; Alligatorweed) has been mostly successful along the Gulf coast of the US, where winters are warm and its primary control agent, Agasicles hygrophila Selman and Vogt (Coleoptera: Chrysomelidae; Alligatorweed Flea Beetle), overwinters successfully (Buckingham 1996, Coulson 1977, Vogt et al. 1992). In 1975, the pathogenic fungus, Alternaria alternantherae Holcomb & Antonop., was discovered producing characteristic purple leaf lesions on Alligatorweed plants near Baton Rouge, LA (Holcomb 1978, Holcomb and Antonopoulos 1976,). Subsequent studies found A. alternantherae to be highly damaging to Alligatorweed plants but posed a risk to several non-target species (Barreto and Torres 1999, Holcomb 1978, Pomella et al. 2007). Nonetheless, there has been recent interest in evaluating A. alternantherae for its potential as an inundative (i.e., applied to the target weed in a similar manner to herbicide application) biological control agent (Gilbert et al. 2005, Pomella et al. 2007). 1Research Biologist, Environmental Laboratory, US Army Engineer Research and Development Center, 3909 Halls Ferry Road, Vicksburg, MS 39180. 2Research Plant Pathologist (retired), Environmental Laboratory, US Army Engineer Research and Development Center, Vicksburg, MS 39180. *Corresponding author - Nathan.E.Harms@usace.army.mil. Manuscript Editor: Richard Baird Southeastern Naturalist N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 442 Despite its known presence in the US since the 1970s, and numerous reports from around the world (Akhtar et al. 2012, Barreto and Torres 1999, Gilbert et al. 2005), field dynamics of A. alternantherae, and especially interactions with introduced biological control agents (e.g., Alligatorweed Flea Beetle), remain unstudied. In other systems, insect–fungal interactions range from positive (e.g., infection facilitates herbivory or vice versa) to negative (e.g., infection increases plant resistance to herbivory or decreases palatability; Saikkonen et al. 1998), to both (e.g., herbivory reduces infection, but infection increases palatability to herbivores; Saikkonen et al. 1998). Although it is uncommon that biological control programs for weeds use a combination of insects and plant pathogens, their combined use may increase efficacy in areas where impacts from either individually are low (Kremer 2000). However, it is important to understand conditions suitable for use of control agents, which can be accomplished by field observations in areas where the agents are established. Although conditions favoring Alligatorweed Flea Beetle occurrence on Alligatorweed have been investigated (Coulson 1977, Harms and Shearer 2017, Vogt et al. 1992), there are no reports to date on occurrence of A. alternantherae at a range of field sites. Along with interest in its potential as a biological control agent, there has been additional interest in A. alternantherae due to recent reexamination of the seasonality and effectiveness of control of Alligatorweed by Alligatorweed Flea Beetle (Harms and Shearer 2017). An understanding of when and where the pathogen occurs may aid in predicting its potential as a biological control agent, in determining whether interactions with other agents might occur, and in predicting whether those interactions will be positive, negative, or neutral. In the current study, our goal was to document spatial and temporal patterns in A. alternantherae occurrence in Alligatorweed among study sites in Louisiana. We surveyed sites along a 315-km north–south transect in Louisiana multiple times per growing season for 2 y. Our findings represent the first documentation of A. alternantherae across a range of sites and over multiple years. Field-site Description We chose sites to cover a broad climatic gradient, which included 11 sites in Louisiana and 1 site (Openwood Pond) in Mississippi (Fig. 1, Table 1). For purposes of our temporal analysis, we classified sites into 3 regions of the state determined by latitude: South (28.0°N–30.5°N), Central (30.5°N–31.5°N), North (31.5°N–33.0°N). We made geographical classifications in an attempt to capture approximately equal portions of the state for each region. Site types varied but consisted of ponds (Openwood Pond, Simmesport Pond, Greenwood Community Park), rivers/bayous (Blind River, Bayou Chevreuil, Bayou Macon, Choctaw landing), wildlife management area wetlands (Maurepas WMA, Blackwater Conservation Area), and lakes (Martin Lake, Poverty Point Reservoir, Lake St. Joseph). We surveyed Alligatorweed for the presence of characteristic leaf lesions every 2–3 weeks in 2015 and every 3 weeks in 2016. In 2015, we surveyed 6 sites from March until August. In 2016, we expanded to 12 sites surveyed from March Southeastern Naturalist 443 N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 until October. Each sampling area within a site was limited to ~5 m2, and we sampled the same area during each visit. Methods Alternaria alternantherae sampling We employed 2 methods to assess A. alternantherae at our sites. We collected 30–50 Alligatorweed leaves per site from the 3rd and 4th apical node of plants, then randomly selected for pathogen processing a subsample of 5 leaves that displayed symptoms associated with Alligatorweed leaf spot. On some dates, we observed no diseased plants; thus, there were zero pathogen samples on those dates. We also collected no samples if plants had been defoliated (e.g., by the biological control agent Alligatorweed Flea Beetle). Therefore, the number of leaves processed from each site varied from 0 to 5 during each collection period. Several sites became inaccessible due to high water, so the number of sites sampled each date from each region varied from 1 to 4. We also randomly selected a subsample of 20 leaves at each site during each sample period, and scanned them on an Epson Perfection V550 Photo flat-bed scanner (Epson America Inc., Long Beach, Figure 1. Eleven study sites in Louisiana and 1 in Mississippi grouped by region in the state. Southeastern Naturalist N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 444 CA) at 600 dpi. To determine the proportion of leaves within sites that were diseased, we classified each one as diseased or healthy, based on visual observation of lesions. We were interested in whether plant damage status was informative of infection by A. alternantherae. Accumulated stress from previous damage events (from insect feeding or pathogen infection) may cause Alligatorweed plants to be more susceptible to future impacts if a defensive response (e.g., change in specific leaf area, C/N ratio) is not strong. As an indicator of accumulated damage at the time of our sampling, we quantified the number of missing leaves per plant. Leafabscission may occur in response to high damage levels (Akhtar et al. 2012, Faeth et al. 1981, Gilbert et al. 2005, Harms et al. 2017); thus, an examination of the proportion of missing leaves could be a fair representation of previously sustained damage. We calculated this fraction by subtracting the number of missing leaves from the total number of potential leaves (with 2 per node), then dividing by the total number of potential leaves and subtracting the result from 1. This calculation provided the proportion of missing leaves for each date and location. Other types of stressors (e.g., herbicide, drought) could result in similar leaf abscission Table 1. Study sites and coordinates. Latitude (lat) in °N and longitude (long) in °W. Waterbody Site Name Region Lat, Long type Location description Poverty Point Reservoir N 32.53, 91.49 Reservoir State park, north of Delhi, LA Openwood Pond N 32.40, 90.79 Residential Residential pond in Vicksburg, pond MS Bayou Macon N 32.09, 91.56 River Public boating access Lake St. Joseph N 32.08, 91.23 Lake Oxbow Lake at Newellton, LA Spring Bayou C 31.14, 92.01 River Spring Bayou WMA, North of Marksville, LA Simmesport Pond C 30.97, 91.81 Public pond Public pond, Simmesport, LA. Greenwood Community Park C 30.57, 91.17 Public pond Community park/golf course in Baton Rouge, LA Blackwater Conservation Area C 30.54, 91.09 Wetland Public park wetland near Baton Rouge, LA Martin Lake S 30.22, 91.90 Lake Public access near Breaux Bridge, LA Maurepas WMA S 30.15, 90.81 Swamp Maurepas WMA south of Sorrento, LA Blind River S 30.09, 90.78 River Blind River WMA near Gramercy, LA Bayou Chevreuil S 29.91, 90.73 River Public access near Thibodaux, LA Choctaw Landing S 29.85, 90.68 River Sanchez boat landing near Choctaw, LA. Southeastern Naturalist 445 N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 (Grossmann 2010, Parker and Pallardy 1985), but these conditions were not common at our sites. Herbicide application occurred infrequently (only twice that we recognized) in the areas around our sample locations and sites remained saturated during the study period. A 0.5 x 1.0-cm section from each unique lesion on each diseased leaf was excised, surface-sterilized in 10% bleach for 1 min, rinsed in sterile water, then plated onto Martin’s agar (maximum of 3 leaf pieces per plate) (Dhingra and Sinclair 1995). We incubated the plates in the dark at 25 °C for 1 week, at which time we cut and placed onto potato dextrose agar (PDA) (Difco, Detroit, MI) a 1 mm x 1mm section from the leading edge of each distinct colony. We incubated the isolates for 1 week at 25 °C and sorted them into morphospecies based on colony morphology and color. We identified morphospecies to species when possible by plating each onto PDA and potato–carrot agar (PCA) plates. We used 2 agars because PCA stimulates sporulation and PDA provides morphological characteristics (distinctive colony colors and growth patterns) for identification. After we identified the isolates, we determined the proportion of leaves in which A. alternantherae was isolated for each site and date. We multiplied this proportion by the proportion of diseased leaves at sites (determined by earlier leaf-scan classification) to estimate A. alternantherae abundance as the total proportion of leaves at sites infected by A. alternantherae. This abundance value was the response variable in our statistical analyses. Statistical approach. We took a number of approaches to examine spatial and temporal patterns of A. alternantherae occurrence during our study. We normalized all proportion data through arcsine–square root transformation (Gotelli and Ellison 2004). We used general linear models to identify spatial (i.e., latitudinal) patterns in pathogen occurrence. We averaged the proportion of diseased plants at sites on each date and used overall site means (per year) as the response variable in the model. We included year, latitude, and a year x latitude interaction in the model. We calculated overall mean proportion of diseased leaves by site and year, and used that average as the data point; thus, there were 6 data points in 2015 and 12 in 2016. We were also interested in variability of disease occurrence within sites, so we calculated the coefficient of variation (CV) of the proportion of diseased plants at each site and used the same linear model to compare CV between years and along the north–south gradient. We used Pearson correlation to examine the relationship between accumulated stress (estimated by proportion of missing leaves per plant) and infection by A. alternantherae. All statements of statistical significance are based on α = 0.05. We performed statistical analyses in Statistica ver. 12 (Statsoft, Inc., Tulsa, OK). Results Clear spatial and temporal patterns emerged during our study into A. alternantherae occurrence. During both years, the proportion of diseased plants increased during the growing season (Fig. 2A, B). In 2015, by May, we documented Southeastern Naturalist N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 446 A. alternantherae in southern sites (nearly 50% plants infected), but occurrence in northern sites did not peak until the last sampling period in August. However, in 2016, all regions displayed similar patterns in infection during the year (Fig. 2B). Although the maximum proportion of infected leaves in 2016 did not reach levels from 2015, all regions had low rates of infection early, which increased during the year, represented by peaks in May and September. The proportion of Alligatorweed leaves infected with A. alternantherae decreased 99% from the maximum of 0.36 in a southern site to less than 0.01 in multiple northern sites (Fig. 2C; Latitude: df = 1, F = 12.51, P = 0.003) but varied only marginally between years (df = 1, F = 4.22, P = 0.06). Extrapolation from the linear model (Fig. 2C) suggests that the northernmost distribution limit of A. alternantherae may be near 34°N. Variability in the proportion of infected leaves nearly doubled with latitude (df = 1, F = 6.42, P = 0.02), but not between years (P = 0.83) (Fig. 2D). We detected no significant interactions between latitude and year for either mean abundance of variation in mean abundance. We also found no significant relationship between the proportion of missing leaves per plant and the proportion of remaining leaves infected by A. alternantherae (P > 0.05). Figure 2. (A) Proportion of leaves infected by A. alternantherae during (A) 2015 and (B) 2016, and (C) mean and (D) coefficient of variation of the proportion of leaves infected by A. alternantherae. Mean (± SE) proportion infected by year is displayed in inset. Best-fit lines in (C) and (D) were fit by least squares regression. Southeastern Naturalist 447 N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 Discussion We were able to demonstrate, for the first time, that the occurrence of the Alligatorweed leaf spot pathogen A. alternantherae varied considerably during multiple years and across sites. We found that incidence of the pathogen generally decreased (and variability increased) at northern latitudes (Fig. 2). Although we did not directly examine the relationship between infection by A. alternantherae and damage due to Alligatorweed Flea Beetle feeding, it is possible that the patterns we observed are partially the result of seasonal variation in insect biological-control damage to plants. For example, Harms and Shearer (2017) documented defoliation due to feeding by Alligatorweed Flea Beetle in these same southern, but not northern, sites in 2015. Thus, a possible explanation for the pattern we observed in 2015 is that plants weakened by flea beetle damage were more susceptible to infection early in 2015, whereas northern plants never reached comparable maximum levels of feeding damage or infection. However, in 2016, overall damage levels from Alligatorweed Flea Beetle feeding were higher across sites (N.E. Harms, unpubl. data), but maximum infection rates were lower (Fig. 2A, B). Our examination of the relationship between accumulated damage (missing leaves) and the infection of remaining leaves failed to uncover a significant relationship. The apparent synchronization in timing of presence, but not abundance, between populations during 2016 could be related to either Alligatorweed Flea Beetle feeding or overwintering success, or both. In 2016, we observed feeding by Alligatorweed Flea Beetle at all sites early in the year, presumably a consequence of high overwintering survival during the mild winter of 2015–2016 (mean winter temperature averaged across all sites = 13.59 ± 0.53 °C) as compared to the previous colder winter in 2014–2015 (9.97 ± 0.69 °C). It may be that overwintering survival and subsequent occurrence of both Alligatorweed Flea Beetle and A. alternantherae are correlated but not causal. Also, host availability in northern sites may be low if plants are killed back to the water line by hard frosts. Whether low winter temperatures in our study sites have a direct mortality effect on A. alternantherae or are an indirect effect from a reduction in host plants is unknown. Although we did not sample plant material during winter months, it would be worth investigating whether A. alternantherae persists in infected tissues underwater during these months or is dispersed to northern sites during the year. The latter explanation is plausible, given the large-scale dispersal patterns of Alligatorweed Flea Beetle from southern to northern sites (Coulson 1977, Harms and Shearer 2017) and because spores of Alternaria spp. can disperse by wind, water, or insect vectors (e.g., Chen et al. 2003, McCartney and Fitt 1998, Meredith 1966). Future research could benefit from examining Alligatorweed Flea Beetle as a disease vector by sampling dispersing individuals for presence of A. alternantherae. We detected an increase in variability of A. alternantherae occurrence with latitude. Geographic distributions of species reflect the biotic (e.g., predation, competition, symbioses) and abiotic (e.g. geology, climate) conditions they experience at varying spatial and temporal scales relative to their niche requirements (Brown 1995, Gaston and Blackburn 2008). Thus, variability in species abundance is often Southeastern Naturalist N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 448 predicted to increase towards range limits (Sexton et al. 2009) because relatively small fluctuations in limiting factors (e.g., winter temperatures) may have disproportionately large effects on the organism. Implications of this variability are that impacts from A. alternantherae, although likely lower overall due to reduced occurrence at higher latitudes, will also be more difficult to pre dict. We observed leaf lesions on 46 occasions for which we did not isolate A. alternantherae from leaves. There were a number of other foliar fungi associated with Alligatorweed at our study sites (N.E. Harms and J.F. Shearer, unpubl. data); thus, we conclude that a different pathogen(s) was responsible for damage on those dates. Interactions between foliar fungi may be complex and interconnected, potentially related to nutrient supply or various kinds of stressors (Carroll 1988, Liu et al. 2011); the patterns we observed in differences with latitude or time of the year could be related to as yet unknown relationships between different Alligatorweedassociated fungi. Ours is the first study to examine spatial patterns of pathogen occurrence relative to a climatic gradient in the context of the Alligatorweed biological control program. Although our study lasted only 2 growing seasons, we provide evidence for spatial and temporal patterns in occurrence of the common Alligatorweed leaf spot pathogen A. alternantherae. Future applied efforts might examine whether the pathogen could be used as an inundative agent in northern areas outside the range of the insect agents (e.g., Arkansas, Tennessee) or experimentally test cold hardiness of the pathogen. Potential interactions between A. alternantherae and introduced insect agents (e.g., Alligatorweed Flea Beetle ) may prove fruitful if they are not antagonistic to each other. Despite the number of years since Alligatorweed was first introduced in northern areas, it is still important to identify appropriate management tools to reduce the impact of infestations. Alligatorweed populations in the US may be a mix of genotypes (Pan et al. 2013), so future research may also account for the importance of plant genotype on management effic acy. Acknowledgments We thank Jonathan Winslow, Julie Nachtrieb, Sam Kirk, Schuyler Cool, Andrew Flick, and Tri Tran for assistance in the field. We are grateful to Lynde Dodd, Gary Dick, Aaron Ellison, and 2 anonymous reviewers for suggestions to improve this manuscript. This work was conducted under the US Army Corps of Engineers Aquatic Plant Control Research Program under management of Dr. Linda Nelson. Literature Cited Akhtar, K.P., N. Sarwar, and Imran-Ul-Haq. 2012. Nimbya alternantherae reported for the first time to cause leaf and stem necrosis of Alternanthera philoxeroides (Alligatorweed) in Pakistan. Tropical Plant Pathology 37:428–430. Barreto, R.W., and A.N. Torres. 1999. Nimbya alternantherae and Cercospora alternantherae: Two new records of fungal pathogens on Alternanthera philoxeroides (Alligatorweed) in Brazil. Australasian Plant Pathology 28:103–107. Brown, J.H. 1995. Macroecology. University of Chicago Press, Chicago, IL. 284 pp. Southeastern Naturalist 449 N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 Buckingham, G.R. 1996. Biological control of Alligatorweed, Alternanthera philoxeroides, the world’s first aquatic weed success story. Castanea 61:232–243. Carroll, G. 1988. Fungal endophytes in stems and leaves: From latent pathogen to mutualistic symbiont. Ecology 69:2–9. Chen, L., T. Price, and Z. Park-Ng. 2003. Conidial dispersal by Alternaria brassicicola on Chinese Cabbage (Brassica pekinensis) in the field and under simulated conditions. Plant Pathology 52:536–545. Coulson, J.R. 1977. Biological control of Alligatorweed, 1959–1972: A review and evaluation. Department of Agriculture, Agricultural Research Service, Washington DC. 98 pp. Dhingra, O., and J. Sinclair. 1995. Basic Plant Pathology Methods. CRC Press, Boca Raton, FL, 448 pp. Faeth, S.H., E.F. Connor, and D. Simberloff. 1981. Early leaf abscission: A neglected source of mortality for folivores. The American Naturalist 117:409–415. Gaston, K., and T. Blackburn. 2008. Pattern and Process in Macroecology. John Wiley and Sons, Hoboken, NJ. 377 pp. Gilbert, R.L., B.A. Auld, and B.R. Hennecke. 2005. Leaf and stem spot of Alternanthera philoxeroides (Alligatorweed) in Australia caused by Nimbya sp. Plant Pathology 54:585–585. Gotelli, N.J., and A.M. Ellison. 2004. Primer of Ecological Statistics. Sinauer Associates Publishers, Sunderland, MA. 614 pp. Grossmann, K. 2010. Auxin herbicides: Current status of mechanism and mode of action. Pest Management Science 66:113–120. Harms, N.E., and J.F. Shearer. 2017. Early-season dynamics of Alligatorweed biological control by Agasicles hygrophila in Louisiana and Mississippi. Journal of Aquatic Plant Management 55:89–95. Harms, N.E., J.F. Shearer, and M.J. Grodowitz. 2017. Folivory and disease occurrence on Ludwigia hexapetala in Guntersville Reservoir, Alabama. Journal of Aquatic Plant Management 55:19–25. Holcomb, G. 1978. Alternaria alternantherae from Alligatorweed also is pathogenic on ornamental Amaranthaceae species. Phytopathology 68:265–266. Holcomb, G.E., and A.A. Antonopoulos. 1976. Alternaria alternantherae: A new species found on Alligatorweed. Mycologia 68:1125–1129. Kremer, R. 2000. Combinations of microbial and insect biocontrol agents for management of weed seeds. Pp. 799–806, In N.R. Spencer (Ed.). X International Sympiosium on Biological Control of Weeds, 4–14 July 1999, Montana State University, Bozeman, MT. 1010 pp. Available online at https://www.invasive.org/publications/xsymposium/ proceed/11pg799.pdf. Liu, Q., A.J. Parsons, H. Xue, K. Fraser, G.D. Ryan, J.A. Newman, and S. Rasmussen. 2011. Competition between foliar Neotyphodium lolii endophytes and mycorrhizal Glomus spp. fungi in Lolium perenne depends on resource supply and host carbohydrate content. Functional Ecology 25:910–920. McCartney, H.A., and B.D.L. Fitt. 1998. Dispersal of foliar fungal plant pathogens: Mechanisms, gradients and spatial patterns. Pp. 138–160, In D.G. Jones (Ed.). The Epidemiology of Plant Diseases. Springer, Dordrecht, Netherland. 460 pp. Meredith, D. 1966. Spore dispersal in Alternaria porri (Ellis) Neerg. on onions in Nebraska. Annals of Applied Biology 57:67–73. Pan, X.Y., X. Jia, D.J. Fu, and B. Li. 2013. Geographical diversification of growth-defense strategies in an invasive plant. Journal of Systematics and Evolution 51:308–317. Southeastern Naturalist N.E. Harms and J.F. Shearer 2019 Vol. 18, No. 3 450 Parker, W.C., and S.G. Pallardy. 1985. Drought-induced leaf abscission and whole-plant drought tolerance of seedlings of seven Black Walnut families. Canadian Journal of Forest Research 15:818–821. Pomella, A.W.V., R.W. Barreto, and R. Charudattan. 2007. Nimbya alternantherae a potential biocontrol agent for Alligatorweed, Alternanthera philoxeroides. BioControl 52:271–288. Saikkonen, K., S.H. Faeth, M. Helander, and T.J. Sullivan. 1998. Fungal endophytes: A continuum of interactions with host plants. Annual Review of Ecology and Systematics 29:319–343. Sexton, J.P., P.J. McIntyre, A.L. Angert, and K.J. Rice. 2009. Evolution and ecology of species range limits. Annual Review of Ecology, Evolution, and Systematics 40:415–436. Vogt, G.B., S. Kay, and P. Quimby Jr. 1992. Effects of weather on the biological control of Alligatorweed in the lower Mississippi Valley region, 1973–83. US Department of Agricultuer Technical Bulletin 1766. USDA Agricultural Research Service, Washington, DC. 143 pp.