Understanding how biotic and abiotic factors influence vector-borne pathogen spread is essential for effective management. Cotton leafroll dwarf virus (CLRDV), transmitted by the cotton aphid, Aphis gossypii, can cause yield loss in cotton, Gossypium hirsutum CLRDV has a variable incidence across the U.S. Cotton Belt, but factors underlying this variation are unknown. Field surveys of commercial cotton fields conducted in Alabama and Georgia from 2021 to 2022 allowed for collection of data on A. gossypii, natural enemies, CLRDV weed hosts, CLRDV incidence, landscape composition, temperature, and precipitation. A structural causal modeling analysis using a directed acyclic graph framework was used to test the significance of these factors on CLRDV incidence in cotton while controlling for potentially confounding effects. A. gossypii abundance, prevalence of cotton in the landscape, and temperature had significant and positive effects on CLRDV incidence, whereas other variables were not significant. Temperature had a significant and positive effect on A. gossypii abundance, which likely contributed to the overall effect of temperature on CLRDV incidence. Cotton is a host for both the vector and virus, which may increase the abundance of both. Novel findings from this study identify factors associated with regions at high risk for CLRDV. Future work can leverage these findings to better predict the spatial and temporal variability in disease incidence.

Insect-associated viruses (viromes) shape insect biology and agroecosystems, yet aphid viromes remain undercharacterized. The cotton aphid, Aphis gossypii, is a globally distributed pest with a broad plant host range and demonstrated virus-vector competence, making it well suited for investigating virome composition at the plant–insect–virus interface. In this study, we profiled the virome of field-collected aphid alates, which were dominated by A. gossypii but included additional aphid species in some samples, across 15 cotton fields in Alabama, USA, to assess taxonomic diversity, relative viral abundance, and focal aphid-associated viruses. Metatranscriptomic sequencing of aphid alates revealed differences in taxonomic composition among county-level libraries, including viruses and diverse prokaryotic and eukaryotic taxa. Fifty-eight viral contigs > 1 kb were assembled, of which 20 were assigned to seven families: Dicistroviridae, Iflaviridae, Mitoviridae, Nudiviridae, Partitiviridae, Phasmaviridae, and Solemoviridae. Representative contigs from six families (all except Phasmaviridae) and five additional family-unassigned viruses were validated by PCR and Sanger sequencing. Three iflaviruses (RrIV, AgIV1, and AgIV2), positive-sense single-stranded RNA viruses, were discovered, including one associated with Rhopalosiphum rufiabdominale and two with A. gossypii. Their complete genomes were determined using PCR-based resequencing and 5′/3′ RACE, each comprising a single open reading frame that encodes a polyprotein. Sequence identity and phylogenetic analyses indicate that the newly identified viruses are putative new species and form a distinct clade together with previously reported aphid iflaviruses within Iflaviridae. AgIV1 and AgIV2, despite being associated with the same host species, were not monophyletic within this clade, consistent with cross-species transmission among aphid hosts. Strand-specific RT-PCR detected negative-strand RNA for AgIV1 and AgIV2, suggesting replication in aphids. Given their low field prevalence, we propose the names Iflavirus furtiva (RrIV), Iflavirus obscurata (AgIV1), and Iflavirus rarivira (AgIV2). The aphid virome is taxonomically diverse and shows county-level differences in relative viral contig abundance. We identified seven viral families and three complete iflavirus genomes, providing a foundation for further investigating their host range, transmission, and potential impacts on aphid biology and cotton production.

Tobacco thrips, Frankliniella fusca Hinds (Thysanoptera: Thripidae), are economically important early-season pests of seedling cotton and peanut in the southern United States, causing stand loss, delayed growth, and transmission of tomato spotted wilt virus (TSWV) in peanut. Historically, managing F. fusca has depended on insecticide seed treatments and in-furrow applications. However, control efforts are becoming more limited due to insecticide resistance and the partial suppression of virus transmission. ThryvOn cotton, which expresses the Bt protein Cry51Aa2.834_16, suppresses thrips feeding and reproduction without causing high mortality, raising questions about how this trait may influence thrips host use in mixed cropping systems. We evaluated oviposition preference and population abundance of F. fusca on peanut, ThryvOn cotton, and non-ThryvOn cotton using greenhouse no-choice, two-choice, and three-choice assays, as well as a field experiment in which all crops were planted side by side. Across all greenhouse assays, F. fusca consistently laid the greatest number of eggs on peanut, intermediate numbers on non-ThryvOn cotton, and the fewest on ThryvOn cotton. Field results closely reflected these patterns, with peanut supporting significantly higher egg, nymph, and adult densities throughout early vegetative growth, while ThryvOn cotton consistently harbored the lowest thrips populations. Our findings suggest that F. fusca exhibits a preference for peanut over cotton, particularly ThryvOn cotton, indicating that reduced host suitability of ThryvOn cotton may influence early-season thrips distribution in diversified cropping systems. Although thrips movement was not measured, this preference highlights the need to consider interactions when developing regional management strategies.

High-throughput sequencing (HTS) has expanded our perspective on the distribution and diversity of plant viruses. Furthermore, improvements in HTS and decreasing sample costs have enabled the discovery of novel plant viruses in field-collected samples. This study examined the putative virome of cotton samples collected from fields across the southern United States. Leaf samples were collected, and total RNA was extracted. Library preparation was performed from pooled samples within locations before sequencing on an Illumina platform. Sequenced libraries were mapped to the cotton reference genome, and the resulting sequences were de novo assembled. A metatranscriptomics analysis revealed complete genome contigs of cotton leafroll dwarf virus in all tested samples. Additionally, 29 putative families of RNA and DNA plant viruses co-infecting cotton were found. Seven families of RNA viruses were more prevalent across all locations. These families included Botourmiaviridae, Hypoviridae, Mitoviridae, Narnaviridae, Partitiviridae, Solemoviridae, and Totiviridae. The information obtained in this investigation will help develop a broader perspective on cotton virus diversity and whether co-infections of viruses can influence (negatively or positively) plant physiology, product quality, and yield.

Amrasca biguttula is a pest of cotton, ornamentals and vegetables and is invading the southeastern United States. We addressed five questions: (1) Are US populations genetically identical to those in the putative native range? (2) Does the southeastern US outbreak extend the recent Puerto Rico and Florida invasion or represent separate introductions? (3) Does the lineage infesting ornamentals match that on other crops? (4) What is the mitochondrial cytochrome oxidase subunit 1 (mtCOI) diversity of A. biguttula? (5) Which regions constitute its native versus invaded range? Adults were collected from Florida, Georgia, South Carolina, Texas and Puerto Rico. Species identity was confirmed morphologically and by sequencing mtCOI, where possible, from five individuals per site, yielding consensus haplotypes. We analysed 11 US, 20 Puerto Rican and 342 sequences originating from publicly available databases (n = 373). Phylogenetic analysis confirmed monophyly of this species. A haplotype network recovered 70 unique haplotypes, including a star-like core of 68 radiating from Hap01 and two divergent singletons. The mainland US and Chinese populations were fixed for Hap01, indicating recent introductions. Puerto Rico exhibited Hap01, Hap52 and Hap70, reflecting multiple introductions but could also result from haplotype diversity carried within a single introduction involving many individuals. Furthermore, southeastern US sites were fixed for Hap01, indicating extension from Puerto Rico and Florida. No haplotype was host-specific, confirming a lineage across crops. Diversity metrics indicate recent expansion in Iraq, Hong Kong and Puerto Rico. South Asia, particularly Bangladesh, India and Pakistan, showed the highest haplotype diversity, supporting its role as the putative native range. In contrast, low diversity in China and the United States is consistent with recent invasion. Curation yielded 373 high-quality sequences, providing a unified database to support and management strategies.

The cotton aphid Aphis gossypii Glover is a pest of cotton Gossypium hirsutum L. in the southeastern United States, capable of direct feeding damage and as a vector of plant viruses. Dispersal activity in agroecosystems can be monitored passively with the use of colored pan traps. The unintended bycatch of Hymenoptera was noted in these traps during an approximately 1-mo span during the peak blooming period in 2021 and 2022. While native pollinators have been receiving recent attention due to their benefits to crops, native bees in Alabama have not yet been investigated on cotton. Adjacent states such as Georgia, Mississippi, and Texas have recorded the benefits of Hymenoptera species to cotton pollination, showing increased lint and seed yield and projecting potential gains to profits for growers. We identified the Hymenopteran species captured as bycatch in yellow pan traps, recording 66 species/species complexes in 4 families. Twenty-two of these species were recorded as cotton pollinators in other states, and 4 species were new or possibly new species records for the state. Our results show that bycatch can be a feasible method to document native bee pollinators and support future research into the idea that cotton may have the potential to provide resources to native bees in agroecosystems.

Infection by aphid-transmitted poleroviruses modulates gene expression associated with plant development and defense. This study assessed the gene expression patterns following cotton leafroll dwarf virus (CLRDV) infection in primary and alternate hosts. Two comparisons (CLRDV-infected versus noninfested and mock-inoculated versus noninfested) were evaluated to identify differentially expressed genes (DEGs) and to tease out differences in gene expression profiles between aphid feeding and aphid-mediated CLRDV infection in each host. CLRDV infection was characterized by 2,079, 1,238, 1,484, and 1,773 DEGs in the primary host cotton and in alternate hosts hibiscus, okra, and prickly sida, respectively. The number of DEGs upon aphid feeding was lower than in CLRDV infection in all hosts except okra. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes terms identified DEGs associated with development, defense, and vector fitness influencing compounds in CLRDV-infected plants. Genes associated with phytohormones, photosynthesis, salicylic acid, jasmonic acid, pathogenesis-related proteins, heat shock proteins, transcription factors, membrane transporters, terpenoids, carbohydrates, and amino acids were differentially expressed in CLRDV-infected plants and varied between hosts. Few overlapping and numerous unique genes in the above-stated categories were differentially expressed upon aphid feeding and varied between hosts. DEGs associated with signaling pathways, transcription factors, systemic resistance, pathogenesis-related proteins, and carbohydrate and amino acid biosynthesis were common between aphid-mediated CLRDV infection and aphid feeding alone. The observed gene expression patterns reiterate that differences in host susceptibility to the virus and/or the vector could differentially influence host defense and development, as well as vector fitness.

Cotton leafroll dwarf virus (CLRDV) is a polerovirus transmitted by Aphis gossypii Glover. Factors contributing to cotton (Gossypium hirsutum L.) yield losses caused by CLRDV infection remain unclear, but results from previous studies indicate that the environmental component of the disease triangle may significantly influence yield loss outcomes. This 3-year study was conducted to compare yield and yield components, fiber quality, root weight, and other morphological parameters of CLRDV-infected and noninfected plants grown under high heat conditions. Potted plants were grown in an insect-proof screen house covered with plastic, and data were collected on a per-plant basis. Plants did not exhibit obvious symptoms during any year of the study. CLRDV infection significantly reduced lint yield, number of seeds, number of bolls in the first fruiting position, seed index, and root dry weight. Fiber quality analysis showed that there was a reduction in the content of short fibers in CLRDV-infected plants. Possible interference of CLRDV on translocation of photoassimilates in the phloem, along with changes in net photosynthesis and other physiological processes, may explain how this virus affected some of the parameters evaluated, especially lint yield, yield components, and root dry weight.Copyright (c) 2026 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

Previous studies investigating the relationship between landscape composition surrounding a crop, and either plant pathogen vectors or vector-borne plant pathogens in the crop, have not observed consistent relationships, and no quantitative synthesis of the current literature has been conducted. We conducted a meta-analysis of 289 published effect sizes from 28 studies to better understand the relationships between landscape composition and plant pathogen vectors or vector-borne pathogens, while considering a series of biological, ecological, and environmental factors that may contribute to variation in the relationships. We found that vector and vector-borne pathogens were positively associated with landscape composition metrics, including the percentages of the landscape surrounding the study areas that were crop, non-crop, and natural habitat but not with habitat diversity. Vector host plant specialists exhibited a significant positive association with various landscape composition metrics, while host plant generalists did not, supporting the idea that specialists are more dependent upon specific host landscape features, whereas generalists have hosts that are generally more abundant across various landscapes. Measurements of both abundance and incidence of pests exhibited significant positive associations with landscape composition metrics. For both persistent and non-persistent pathogens, the positive association between pathogens and landscape composition became stronger at larger spatial scales (500 to 10,000 m). This meta-analysis demonstrated that increasing the percentage of crop, non-crop, or natural habitat within a landscape can increase vectors or vector-borne pathogens in the crop. However, the specific landscape feature that contributes to increased pest abundance or incidence will depend on the crop-pest system.

Cotton leafroll dwarf virus (CLRDV) is an aphid-transmitted virus recently identified across the Cotton Belt region of the U.S. Yield responses to CLRDV have been inconsistent, and asymptomatic infections can occur in which no losses are apparent. Conditions underlying the observed variation in yield responses are not understood. A three-year field study was conducted to examine whether yield loss caused by CLRDV is influenced by the age of cotton plants at the time of infection. Timing of infection was investigated by infesting caged plots of cotton with viruliferous Aphis gossypii to transmit CLRDV at different seedling growth stages, beginning after seedling emergence and continuing weekly for three to four weeks. Timing of infection impacted yield only in one out of three years suggesting that plant age is not a primary factor contributing to CLRDV-related yield loss. Future studies are needed to identify the environmental factors influencing disease severity and yield loss.

In 2017, cotton (Gossypium hirsutum L.) leafroll dwarf virus (CLRDV) was first reported in the United States. One CLRDV inoculum source includes the previous year's cotton stalks; hence, destroying cotton stalks could be effective for CLRDV management. However, tillage-intensive stalk destruction methods (SDMs) can degrade southeastern soils, but a cover crop may provide short-term benefits and reduce CLRDV incidence. Therefore, we examined three SDMs (Tillage, Pull, Mow) across two cover crop levels (no cover and rye [Secale cereale L.]/clover [Trifolium incarnatum L.] mixture) and two cotton varieties to determine how cotton growth, soil penetration resistance (PR), and two CLRDV incidence sample times (pre-harvest and post-harvest) were affected across six environments during the 2021 and 2022 growing seasons. None of the SDMs affected any factors examined in this experiment, except soil PR and cotton yield. The Pull and Mow SDMs both increased soil PR compared to the Tillage SDM. An 8% yield increase (Pull > Mow) was observed, but the Tillage SDM yield did not differ from Pull or Mow SDMs. The rye/clover mixture also increased soil PR. Although cotton stands were 15% greater with no cover crop, subsequent cotton yield and fiber quality were minimally affected by cover crops. The rye/clover mixture increased post-harvest CLRDV incidence, and cotton yields were equal between cover crops. Pre-harvest CLRDV incidence probability was 0.23, but post-harvest CLRDV incidence probability was 0.71. Continuing to identify and evaluate cultural practices that reduce CLRDV incidence is imperative to prevent negative impacts.

Colonizing aphids play an important role in the transmission of RNA phytoviruses in the family Solemoviridae. According to "host manipulation hypothesis," phloem limited and persistently transmitted Solemoviridae viruses modulate host physiology that positively affects vector behavior and fitness and facilitates virus spread. However, it is unclear if virus-modulated host effects on vectors across pathosystems involving Solemoviridae members are always positive. Cotton leafroll dwarf virus (CLRDV) is a recently introduced Solemoviridae member in the United States, and it is transmitted by the cotton aphid (Aphis gossypii). Effects of CLRDV infection on vector behavior and fitness were evaluated on its primary host plant, cotton (Gossypium hirsutum), and an alternate host plant, hibiscus (Hibiscus acetosella). In this study, changes to viruliferous and non-viruliferous aphid preference and aphid fitness on virus-infected and non-infected hosts were examined. In contrast to the hypothesized preference of non-viruliferous aphids for infected plants and vice-versa, both viruliferous and non-viruliferous A. gossypii preferred non-infected cotton and hibiscus plants over CLRDV-infected plants. This suggested that the preference of non-viruliferous vectors to non-infected plants might negatively impact virus acquisition, whereas the preference of viruliferous vectors toward non-infected plants could positively facilitate virus inoculation. The total fecundity and intrinsic rate of increase of aphids were higher on non-infected plants compared with CLRDV-infected plants. The lack of enhanced fitness benefits on CLRDV-infected hosts also could negatively impact virus spread. Overall, this study suggested that "host manipulation hypothesis" favoring vector attraction and enhanced fitness on infected plants does not apply to all pathosystems involving Solemoviridae members.

Cotton leafroll dwarf virus (CLRDV) is a plant viral pathogen first reported in the United States in 2017 that causes yield loss in certain cotton (Gossypium hirsutum L) cultivars but is asymptomatic in others. In Argentina, an infectious clone of the typical and atypical strains of CLRDV were developed to facilitate biological studies and aid in the development of resistant cotton cultivars. Two infectious clones derived from the original Alabama isolate (Macon 1) were created using Agrobacterium tumefaciens for plant infection. The first one, CLRDV-Macon1 (CLRDV-Ma1) was constructed by inserting the full-length sequence of CLRDV into the pJL89 plasmid, followed by transformation of A. tumefaciens strain LBA4404. This clone systemically infects cotton and Nicotiana benthamiana and was detectable in transfected plants by PCR amplification. A second clone, CLRDV-M-GFP, was constructed by replacing a fragment of the open reading frame 3-5 region of CLRDV-Ma1 with enhanced green fluorescent protein (EGFP). This clone expresses EGFP, making visualization possible by fluorescence microscopy in infiltrated cotton cotyledons, and is detectable using commercial antibodies. However, during the infection cycle, EGFP was found to be truncated, resulting from unintended mutations. This study reports the first infectious clones of a U.S. isolate of CLRDV and contributes new knowledge about the biology of poleroviruses.Copyright (c) 2025 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

Stink bug species emerged as major insect pests of cotton in the mid-southern United States following the eradication of the boll weevil and the introduction of genetically modified Bt cotton for lepidopteran pests. Considering the limited number of chemical classes available for insect control, further insights into other chemistries are necessary to inform management strategies with the overall goal of establishing and maintaining the most cost efficient and efficacious control programs for stink bugs in Alabama. The insect growth regulator, novaluron, has shown control of tarnished plant bugs, but little research has been done on its effect in stink bugs. The objective of our study is to evaluate the effects of novaluron, in a laboratory setting, on adult fecundity, nymphal mortality, and yield and damage in the field. We hypothesized that novaluron would have a direct effect on mortality in nymphs and could decrease fecundity in adult stink bugs. Although the effect on fecundity was counter to our hypothesis, this study shows effective control of nymphs in our model insect, the southern green stink bug, Nezara viridula (Hemiptera: Pentatomidae). Future evaluation of proper timing of novaluron applications could make this a valuable tool for residual control of stink bugs in cotton.

Cover crops are increasingly adopted to suppress weed growth and reduce reliance on chemical herbicides. A greenhouse experiment was conducted to evaluate the emergence and growth response of troublesome southeastern weeds to various cereal rye (Secale cereale L.) residue levels. Trays planted with Palmer amaranth (Amaranthus palmeri S. Watson), sicklepod (Senna obtusifolia L.), ivyleaf morningglory (Ipomoea hederacea), and large crabgrass (Digitaria sanguinalis L.) seeds mixed with soil were covered uniformly by four different levels of cereal rye biomass. The following field experiment was conducted at two locations in Alabama in a split-plot design, with the main plot factor being four seeding rates of cereal rye to obtain various cereal rye biomass. Subplot factors were preemergence herbicide flumioxazin and non-treated (NT) check. The greenhouse results demonstrated reduced seed emergence and lower weed biomass for Palmer amaranth, sicklepod, and large crabgrass in plots with higher cereal rye residue biomass compared to those with lower biomass. In both greenhouse and field conditions, the emergence of ivyleaf morningglory was not affected by the increasing biomass of cereal rye residue. Palmer amaranth seed emergence was the most sensitive to increasing biomass residue due to its small seed size. Cereal rye biomass and Palmer amaranth counts were strongly negatively correlated with a Pearson’s coefficient (r) of 0.83 while weakly negatively correlated for ivyleaf morningglory with 0.49. In conclusion, increasing the biomass of cereal rye residue is effective in suppressing Palmer amaranth seed emergence but not ivyleaf morningglory. The flumioxazin treatment demonstrated 95%–90% control for Palmer amaranth and ivyleaf morningglory, while the NT check exhibited 50% control of Palmer amaranth and 30% control of ivyleaf morningglory by cereal rye biomass alone. In conclusion, a sufficient amount of cereal rye biomass can effectively suppress the emergence and growth of weeds, particularly Palmer amaranth.

The two-spot cotton leafhopper, Amrasca biguttula (Ishida, 1913) (Hemiptera: Cicadellidae), is a polyphagous pest native to the Indian subcontinent, where it is a significant pest of cotton (Gossypium spp.), okra (Abelmoschus esculentus, Moench), and other crops. At adult and immature stages, they feed on leaf cell contents, causing characteristic “hopperburn” symptoms and yield losses of up to 40% in cotton in its native range. In July 2025, A. biguttula was detected for the first time during the cotton growing season in commercial and experimental fields across multiple counties in Alabama, Florida, Georgia, and South Carolina. Identification was confirmed through morphological examination of diagnostic external features and male genitalia. Within four weeks, the pest was found in 101 counties, with injury symptoms ranging from mild chlorosis to severe necrosis and defoliation. Given the southeastern United States’ average of 979,339 hectares of cotton valued at USD 1.7–USD 2.8 billion annually, the potential for economic impact is considerable. Rapid symptom development, broad host range, and multiple generations per year underscore the need for coordinated monitoring and research to determine preliminary control measures and to identify potential alternative hosts. This report documents the first confirmed occurrence of A. biguttula in U.S. cotton and highlights its potential as an established recurring pest threat in the region.

Thrips are one of the major pests of seedling cotton (Gossypium hirsutum L.) in the U.S. Whereas previous studies have investigated the effects of neonicotinoids on reducing adult and immature thrips on cotton, their effect on oviposition in the field is understudied. In this study the effect of three neonicotinoid insecticide seed treatments on the number of thrips adults, immatures, eggs, plant injury, and plant biomass were evaluated. The experiment was replicated six times over weekly planting dates. Insecticides reduced thrips oviposition and immature thrips on seedlings. Seedlings from plantings in April had lower oviposition and number of immature thrips than seedlings planted in May, and the efficacy of insecticides against oviposition and immature thrips was reduced as seedlings reached four to five true leaves. Oviposition was not reduced by insecticides in seedlings planted after 2 May that were exposed to larger numbers of adult thrips. Insecticides reduced plant injury, which was variable across planting dates and growth stages. Differences in biomass were observed only on the last planting date and were likely due to more favorable growing conditions. The efficacy of these insecticide seed treatments to reduce thrips oviposition and numbers of immature insects depends upon pest population pressure, growth stage of the cotton, and growing conditions of the crop.

Aphids are among the most destructive insect pests to crops. Based on the degree of their host specialization, aphids, like other herbivorous insects, have been grouped into three categories: monophagous, oligophagous, and polyphagous [1]. Monophagous aphids feed on only one or a few closely related plant species, often of a single genus, oligophagous aphids feed on several plant species of the same family, and polyphagous aphids feed on plants that belong to more than one family. Polyphagous aphids are considered generalist herbivores, comprising less than half of the total aphid species [2; 3]. However, this polyphagous nature allows generalist aphids to disseminate plant pathogens to a wide range of host plants [3; 4].The cotton aphid (or melon aphid), Aphis gossypii Glover, is a highly polyphagous aphid species that can feed on at least 700 plant species in numerous families including Asteraceae, Cucurbitaceae, Malvaceae, Rutaceae, Solanaceae, and Fabaceae [5; 6; 7]. Population studies of A. gossypii have shown that diversity is mainly associated with differences in host plant preference. Moreover, several plant host-specialized biotypes have been documented [8; 9; 10].Other factors including geography, climate, and pesticide use can also contribute to shaping its population structure [11; 12]. Interestingly, profiles of microbial symbionts, on which aphids are dependent in numerous physiological processes, may vary in different A. gossypii biotypes and populations, suggesting specialized interactions evolved between A. gossypii and its microbial symbionts under selection pressure exerted by a variety of environmental factors [13; 14; 15]. Hence, a population-specific microbiome analysis is crucial to understanding aphid-microbe interactions in locally adapted A. gossypii.As a worldwide distributed agricultural pest, A. gossypii is responsible for severe yield losses of many economically important crops such as cotton, cucumber, and citrus [5; 7]. Besides injuring plants directly by sucking the sap, while feeding, it secretes honeydew which fosters growth of sooty mold that can block sunlight and decrease photosynthesis processes within the plant [16]. Moreover, A. gossypii is important for its ability to transmit over 75 plant viruses [17] and was ranked the second most competent aphid species in terms of number of potyviruses it vectors [3]. In cotton, A. gossypii transmits several viruses including cotton leaf roll dwarf virus (CLRDV), cotton anthocyanosis virus, and cotton bunchy top virus, posing a severe threat to cotton production [18; 19; 20].In the Southeast USA, cotton is one of the most economically important crops, and A. gossypii is a major insect pest of cotton and the only known vector of CLRDV. As a primary cotton-growing region in the Southeast, Alabama also reported the first occurrence of CLRDV in 2017 [21]. This virus was later detected throughout the Southeast [22; 23]. Given the significant economic impact of A. gossypii in Alabama, we performed metatranscriptomic and metagenomic analyses on locally collected cotton aphids to decipher their microbiota.These sequencing datasets provide genetic information, at both RNA and DNA levels, of symbiont microbes and their overall community composition in a local A. gossypii population from Alabama, USA. The microbiome data can be used to identify A. gossypii-associated and transmitted plant pathogens and discover insect-infecting microbes for aphid biocontrol. In addition, plant species identified in the sequencing data from the whole aphid, will provide insights into the plant host range of A. gossypii in the locality tested. Library preparation and Illumina sequencing were conducted at Novogene Corp. Inc.(Sacramento, CA, USA).For RNA sequencing, ribosomal RNAs from both eukaryotes and prokaryotes were first depleted from total RNA samples using the Ribo-Zero rRNA removal kit (Illumina, USA). The remaining RNAs were fragmented into ~250 to 300 bp and then reverse-transcribed into doublestranded cDNAs. For metagenomic sequencing, 1 µg of genomic DNA was randomly sheared into short fragments of approximately 350 bp. The double-stranded cDNAs (for RNA sequencing) and sheared genomic DNA fragments (for DNA sequencing) were subsequently end repaired to produce blunt ends, added with a single 'A' nucleotide at the 3' ends, and further ligated with Illumina adapters. After fragment size selection and PCR amplification, the prepared metatranscriptomic and metagenomic libraries were sequenced on the Illumina NovaSeq platform (Illumina, CA, USA) with pair-end 150 mode.To generate a metatranscriptome, RNA raw reads were first preprocessed by trimming adaptors and removing low-quality reads using Trimmomatic (v0.39) in paired end mode [24].Parameters for Illumina clip were seed mismatches = 2, palindrome clip threshold = 30, and simple clip threshold = 10. Other parameters included the sliding window trimming with a window size = 5, required quality = 20, and minimum read length = 50. Clean reads were then aligned to the A. gossypii genome (NCBI accession GCF_020184175.1) [25] using the BWA-MEM mapping tool (v0.7.17) with its default parameters [26]. Unmapped paired reads were assembled to create metatranscriptomic contigs using SPAdes (v3.15.5) in meta mode [27].A metagenomic assembly was similarly generated following these three steps: 1) preprocessing of DNA raw reads, 2) mapping of clean reads to the reference genome, and 3) assembling of unmapped paired reads. For Step 1, Readfq (v8; https://github.com/cjfields/readfq) was used to trim adaptors and remove the low-quality reads that have: a) more than 40 lowquality bases with Q-value < 38, b) more than 10 ambiguous nucleotides "N", or c) more than 15 bp's overlap with adaptors. Step 2 was conducted using BWA-MEM as described above. ForStep 3, Megahit (v1.2.9) was used at the default setting to generate metagenomic contigs [28].Contigs longer than 400 bp were retrieved for taxonomic analysis. Contig sequences were first aligned to a preformatted NCBI non-redundant (NR) reference database downloaded on August 28, 2023, with the BLASTX function by running DIAMOND (v2.1.8) [29]. The output was written in DAA (DIAMOND alignment archive) format, which was then used for Meganization, an approach of performing taxonomic and functional binning of the sequences [30]. The DAA file was run against the MEGAN database 'megan-map-Feb2022.db' in long read mode, using MEGANIZER, a program included in the MEGAN package (v6_25_3) [31]. Lastly, a taxonomic analysis was conducted using MEGAN, in interactive mode, to determine kingdom and genus level assignations for all contigs.Total RNA and DNA extracted from the A. gossypii sample, consisting of 10 fieldcollected alataes, had high purity (OD260/280 > 2.0) and high quality (RIN = 8.3). A total of 88,776,140 and 84,900,570 raw reads were obtained from the Illumina sequencing of RNA (AAL8R) and DNA (AAL8D) samples, respectively, which, after preprocessing to remove adaptors and low-quality reads, yielded 86,527,106 and 84,867,588 clean reads (Table S1). The GC content of DNA reads was lower (26.77%) than the RNA reads (39.00%) but similar to PacBio reads (27.26-27.99%) of the published A. gossypii genome used as reference [25].Mapping of RNA and DNA reads to the A. gossypii genome revealed 49.21% and 98.17% of genome coverage, respectively. left A total of 17,914,277 and 5,563,338 potentially non-host RNA and DNA reads were unmapped, accounting for 20.70% and 6.56% of their total clean read numbers , respectively (Table S1).Two de novo assemblies were generated: one from the AAL8R and the other from the AAL8D non-host reads not mapped to the A. gossypii genome. The AAL8R assembly consisted of 23,101 contigs with an average length of 365 bp and a medium (N50) length of 337 bp. The AAL8D assembly consisted of 11,415 contigs with an average length of 984 bp and a medium length of 1,390 bp (Table S1). Contigs longer than 400 bp, including 3,804 AAL8R contigs and 8,454 AAL8D contigs, were finally selected for taxonomic annotation.Taxonomic analysis of the non-host reads using Kraken2 [32] The acquired metatranscriptomic and metagenomic contigs were annotated at the kingdom and genus levels using the DIAMOND+MEGAN taxonomic analysis approach [30].Over half of the contigs in both RNA and DNA datasets assembled from non-host reads were classified into specific kingdoms. This included 2704 (71%) AAL8R and 4504 (53%) AAL8D contigs (Fig. 1). "Bacteria", "Metazoa", and "Fungi" were the three most abundant kingdoms for AAL8R. "Metazoa", "Bacteria", and "Naldaviricetes" were most abundant for AAL8D. In both the RNA and DNA datasets, a high proportion of sequences received a "Metazoa" assignation. This is likely the result of reads that did not map to the reference genome due to the presence of sequence gaps and as a result were designated as non-host reads [25]. Genetic variation between the reference genome, obtained with aphids collected in China [25], and those used in this experiment, collected in Alabama, may be another factor that led to the designation of some reads as non-host. Contigs assembled from these "non-host" reads consequently received the "Metazoa" assignation.Previous studies showed that the microbiome of A. gossypii can be determined by a variety of factors, including plant host, geography, and life stage [13; 14; 15; 33; 34]. Our genuslevel taxonomic analysis on bacterial contigs indicated that the genus Arsenophonus was the most dominant group of symbionts in both AAL8R and AAL8D samples (Fig. 2A).Arsenophonus species are known as male-killing facultative symbionts found in a broad range of arthropod hosts [35; 36]. Aside from acting as son killers to benefit female offspring [37], some Arsenophonus species were recognized as insect-vectored plant pathogens [38]. In aphids, members of Arsenophonus can also play a role in parasitoid defense [39] and plant host specification [40].Our analysis demonstrated that Pseudomonas was the second most dominant bacterial genus (569 contigs) in the AAL8R sample (Fig. 2A). Like Arsenophonus, Pseudomonas has been Formatted: Indent: First line: 0.5"shown to interact with its insect host in a multifaceted manner: while some species are entomopathogenic, others may be beneficial endosymbionts of insects or insect-vectored plant pathogens [41]. Other bacterial genera with ≥ 10 contigs in either AAL8R or AAL8D included Aureimonas, Buchnera, Hamiltonella, and Serratia (Fig. 2A). Among these, Aureimonas was found in a recent cotton microbiome study [42] but has not been reported as an aphid symbiont.Given that cotton components were likely present in the gut of A. gossypii collected in the cotton field, it is not possible to discriminate whether Aureimonas DNA reads originated from the aphid or the cotton host. By contrast, Buchnera is a well-studied primary endosymbiont present in almost all aphid species [43]. Despite the contig numbers not being the highest, Kraken2 taxonomic analysis indicated that the number of reads assigned to Buchnera comprised 78.79% (7,056,145 reads) and 56.7% (1,577,306 reads) of the non-host RNA and DNA reads, respectively. This suggests that using contig numbers to infer the abundance of a taxon could be inaccurate, as it does not take into account many factors, such as genome size, contig length, and sequencing depth. However, the number of contigs represents a useful metric for initial assessments, providing a general overview of the taxonomic composition within a sample, especially when combined with other analytical methods. Previous studies using 16S rRNA sequencing have confirmed the presence of several bacterial genera in A. gossypii, including Buchnera, Arsenophonus, Pseudomonas, Hamiltonella, and Serratia [13; 14; 15]. Furthermore, research progress on Hamiltonella and Serratia has been made in recent years, and both genera have been identified in the microbial community of A. gossypii [13; 15]. While Hamiltonella was shown to mainly play a role in stress tolerance and parasitoid defense in insects, Serratia has been shown to be symbiotic or pathogenic to its insect host [44; 45; 46]. Two genera of DNA viruses, Alphabaculovirus and Aplhanudivirus, were detected in both RNA and DNA samples (Fig. 2B). Detection of DNA viruses in the RNA sample suggested that they were actively replicating in the host cells. Alphabaculovirus and Aplhanudivirus are double-stranded DNA (dsDNA) viruses that infect insects [47; 48]. Therefore, we suspect that the contigs of these two genera in the samples could correspond to DNA viruses of A. gossypii.Further analysis is needed to determine their biological and molecular properties. We also found sequences of two genera of putative RNA viruses, Goukovirus and Cripavirus (Fig. 2B), whose members are known to infect insects [49; 50]. Although the aphid-transmitted cotton virus, CLRDV, is widely distributed in cotton fields [23], we did not find CLRDV contigs in this study, possibly due to the relatively small aphid sampling size.Taxonomic analysis on the fungal contigs revealed that Conidiobolus, Fibularhizoctonia, Basidiobolus, and Beauveria were the most abundant genera with at least 5 contigs each (Fig. 2C). Notably, Conidiobolus and Beauveria encompass significant entomopathogenic species like B. bassiana, which has been employed as a biological insecticide to manage a diverse array of insect pests [51; 52]. Despite the prevalence of Neozygites fresenii, a naturally occurring fungal pathogen of A. gossypii in the Southeast USA [53; 54], we did not identify any contigs assigned to the Neozygites genus. This may be the result of the small aphid sample size used in this study.Formatted: Font: Italic Formatted: Font: Italic Formatted: Font: Italic Our analysis, mostly from the RNA sample, showed that the highest number of plant contigs were assigned to the genus Gossypium. This is not surprising as the aphids were collected in cotton fields. However, we also detected at least one contig for twelve other plant genera belonging to several families (Fig 2D). We concludeAlthough these spurious hits in the database are not conclusive, that thesethey might represent remains of plant hosts fed upon by the aphids. Should this be true, tThis finding supports the well-established fact that A. gossypii is polyphagous but also suggests that the aphids we collected moved in and out of the cotton field and had fed on a variety of plants. This is of importance because with this movement A. gossypii can potentially transmit viruses, fungi and bacteria to cotton from plants adjacent to commercial cotton fields. The methods presented in this study, on an expanded scale, can be used to monitor pathogens and plant use of A. gossypii in cotton fields.In conclusion, through sequencing A. gossypii alataes collected in Alabama, USA, we generated two de novo assemblies: a metatranscriptome and a metagenome. The DIAMOND+MEGAN taxonomic analyses on these assemblies uncovered putative sequences of a variety of organisms that may form complex interaction networks associated with A. gossypii. These protocols can be applied not only for microbiome analysis but also to investigate the host range of herbivorous insect species. Additionally, the DNA reads can be used for population genomics research, the RNA reads can be used to enhance gene annotation, and both RNA and DNA reads can contribute to refining the assembly of the A. gossypii genome.

The cotton aphid, Aphis gossypii Glover, is an important plant disease vector and a highly polyphagous agricultural pest that feeds on a broad range of host plants. During feeding, its salivary glands serve as a route for the transmission of circulative plant viruses and produce a range of secretory proteins, called effectors, to modulate host cellular processes. To understand the molecular mechanisms underlying aphid-plant interactions, we developed a bioinformatics pipeline that incorporated the salivary gland transcriptome, genome, and head vs abdomen differential gene expression data to predict secretory protein-encoding genes enriched in the salivary glands of A. gossypii. Annotation of the 351 predicted genes showed that the most abundant functional categories were associated with cellular signaling and metabolism processes, and revealed that 98 genes were hemipteran-specific. Notably, 51 genes encode secretory proteins matching the putative saliva proteins identified in prior proteomics studies. Quantitative PCR analysis validated differential expression of 4 selected genes between heads and abdomens and indicated that alate adults exhibited the highest gene expression, suggesting these genes may play key roles in host colonization. Additionally, 25 genes showed sequence similarities to functionally characterized hemipteran effectors, with some appearing to form effector groups with distinct evolutionary patterns. Collectively, this study identified numerous putative plant-manipulating genes in A. gossypii and provided valuable insights into the mechanisms of aphid-plant interactions.

The introduction of cotton leafroll dwarf virus (CLRDV) (Family: Solemoviridae, Genus: Polerovirus) into the southeastern U.S. Cotton Belt has prompted research into understanding the epidemiology, including identification of potential insect vectors. Aphids (Hemiptera: Aphididae) are vectors of cotton-infecting poleroviruses. Of the eight aphid species reported to feed on cotton in the U.S., only Aphis gossypii Glover has been reported to transmit the phloem-limited virus, CLRDV, to cotton. Population dynamics of alatae aphids and additional information regarding CLRDV spread throughout the year may help identify other vector species. Using insect traps to monitor populations of these aphids is more effective for the seven species that are seldom observed, or difficult to sample because they feed underground on roots. However, current identification keys for aphids rely on using apterous colonies already infesting known host plants or are not specific to cotton fields. Here, we present a modified key for identifying cotton infesting aphids captured in pan traps around agricultural fields in the southeastern U.S., including A . gossypii, Protaphis middletonii Thomas, Aphis craccivora Koch, Rhopalosiphum rufiabdominale Sasaki , Macrosiphum euphorbiae Thomas , Myzus persicae Sulzer , Aphis fabae Scopoli, and Smythurodes betae Westwood. Pictures are provided to aid in the identification of these species.