Final Report for GS14-139

Project Type: Graduate Student
Funds awarded in 2014: $9,500.00
Projected End Date: 12/31/2015
Grant Recipient: University of Georgia
Region: Southern
State: Georgia
Graduate Student:
Major Professor:
Dr. Ron Walcott
University of Georgia
Expand All

Project Information


Watermelon seed treatments with biocontrol agents that are antagonistic against Acidovorax citrulli significantly reduced bacterial fruit blotch (BFB) seedling transmission. 108 CFU/flower was the optimal inoculation concentration for seed treatment. When watermelon seeds treated with biocontrol agents and PBS (negative control) were challenged with 107 CFU/ml of A. citrulli the mean BFB seedling transmission was 80%. By comparison, seedlots from flowers treated with Bacillus mojavensis RRC101 and Bacillus spp. #24 and #35, had significantly lower BFB seedling transmission percentages of 19.4%, 53.2%, and 41.7%, respectively. These results demonstrate the potential for seed treatments with biocontrol agents to improve BFB management.


The main goal of this project was to optimize a technique for applying biocontrol agents to watermelon seeds to reduce the seed-to-seedling transmission of bacterial fruit blotch of cucurbits (BFB). Acidovorax citrulli is the causal agent of BFB, a sporadic but devastating threat to watermelon fruit production worldwide. Outbreaks of this disease can result in severe fruit rot and significant economic losses (Burdman and Walcott, 2012). Infested cucurbit seeds are the primary source of inoculum for BFB outbreaks that develop in seedling transplant facilities and fruit production fields. Currently there are no commercially available sources of BFB resistance (Bahar and Burdman, 2010). Therefore disease management relies heavily on foliar-applied chemicals (e.g. Kocide, Mankocide, Nu-Cop) and pathogen exclusion. However, the efficacy of foliar applied chemicals can be limited under conditions of high rainfall and relative humidity.

A. citrulli exclusion relies on production of pathogen-free seeds in regions of countries with cool and dry climates that are unfavorable for BFB development. Despite these efforts, infested watermelon seed remain the most important source of inoculum for BFB outbreaks. One reason for the lack of effectiveness of seed treatments is that A. citrulli cells can be located in the embryos of watermelon seeds where externally applied chemicals cannot penetrate (Dutta et al., 2012). Additionally, seeds containing as little as 103 colony forming units (CFU) can result in BFB seed-to-seedling transmission under transplant house conditions (Dutta et al., 2012). Hence, an effective treatment must penetrate deep into the seed to eliminate resident bacterial cells.

One solution to this problem is to deposit biocontrol agents into internal tissues of watermelon seeds. On seed germination, the antagonists would rapidly colonize seed tissues and prevent the establishment of A. citrulli populations. Ideally, this would reduce BFB seed-to-seedling transmission and limit BFB outbreaks. Using recombinant DNA techniques, we generated a non-pathogenic strain of A. citrulli (AAC00-1ΔhrcC) that colonized germinating watermelon seedlings at wild-type levels (Johnson et al., 2011). When applied externally to naturally infested watermelon seeds, AAC00-1ΔhrcC significantly reduced BFB seed-to-seedling transmission (Johnson et al., 2011). To improve the efficacy of biocontrol seed treatment, we propse to deposit biocontrol bacteria within seeds by applying cell suspensions to the stigmas of female watermelon flowers. A. citrulli can penetrate the pistil of the female flowers and become deposited in the seed endosperm without inducing BFB symptoms on fruits (Walcott et al., 2003). By exploiting the “pistil invasion pathway”, we hypothesize that biocontrol agents will be more effective in limiting BFB seedling transmission.

Project Objectives:

The main objective of this study was to optimize a flower inoculation technique for efficiently delivering biocontrol bacteria into watermelon seeds. The specific objectives were:


  1. To optimize the concentration of bacteria applied to watermelon flowers to maximize seed infestation.


  1. To determine the efficacy of the flower treatments with biocontrol agents in reducing seed-to-seedling transmission of BFB.


Click linked name(s) to expand
  • Dr. Ron Walcott


Materials and methods:

  1. Optimizing the concentration of biocontrol bacteria inoculum required for seed treatment by flower inoculation


1a) Forty watermelon plants (cv. Crimson Sweet) were established in three-gallon pots under standard greenhouse conditions. At anthesis, female watermelon stigmas were hand pollinated and subsequently inoculated with a non-pathogenic strain, AAC00-1ΔhrcC. A micropipette was used to deliver 10 µl of cell suspension to each stigma. Bacterial concentrations were 0, 107, 108 and 109 CFU/Flower. Flowers were allowed to develop into mature fruit (~35-40 days after pollination); harvested and the seeds were extracted. Fruits were surface sterilized and seeds were extracted manually. Seed from each fruit were maintained as separate lots without fermentation; rinsed, air-dried for 24 h at room temperature and stored in paper bags. A. citrulli-specific quantitative real-time PCR assay was used to determine the percentage of infested seeds per lot and the concentration of bacteria per seed. Subsequently, seed germination percentage was determined.


1b) Seeds (n = 40 seeds/lot) from each lot were challenged with 106 CFU/ml of A. citrulli strain A. citrulli 213-60 by vacuum infiltration, air-dried for 24 h and planted individually into test tubes. Seeds were allowed to germinate and grow for 14 days at 85-90% RH and 28°C and the percentage of seedlings displaying BFB symptoms was recorded.


  1. Determining the efficacy of the optimal inoculum concentration of biocontrol agents in reducing BFB seedling transmission



2a) A watermelon (cv. Crimson Sweet) field plot was established at the Black Shank Farm in Tifton, GA. At anthesis, female watermelon flowers were hand pollinated and the stigmas were inoculated with 108 CFU of Bacillus mojavensis RRC101, Bacillus spp. #24, Bacillus spp. #35 and 0.1M phosphate buffer saline (PBS) as a control. Bacillus mojavensis RRC101 is an endophytic bacterium known for its antagonistic abilities against plant pathogenic fungi, in particular Fusarium moniliforme, a pathogen of maize (Bacon and Hinton, 2001). The other two Bacillus species (24 and 35) were isolated from germinating watermelon seeds. A micropipette was used to deliver 10 µl of cell suspension per stigma and flowers were allowed to develop into mature fruit. After harvest, fruits were stored at room temperature and then extracted as described above.


2b) Seeds (n = 40 seeds/lot) from each lot were challenged with 107 CFU/ml of A citrulli 213-60 by vacuum-infiltration, air dried and planted individually in test tubes with cotton balls saturated with water. Seedlings were incubated at 28°C and 80-90% RH for 14 days and the percentage of seedlings from each seedlot displaying BFB symptoms was recorded.

Research results and discussion:

In this study we confirmed that 108 CFU/flower is the optimal inoculum concentration required to maximize seed infestation with biocontrol agents. 69% of the seedlots produced in fruit from AAC00-1ΔhrcC-inoculated flowers were infested with the biocontrol agent. As expected, seeds produced from flowers inoculated with water had 0 CFU/g of seed. In comparison, AAC00-1ΔhrcC concentrations in seeds from flowers inoculated with 107, 108, and 109 CFU ranged from 0-44, 45-485, and 0-66 CFU/g of seed respectively. The mean bacterial concentration for flowers inoculated with 108 CFU/flower (159.3 CFU/g of seed) was significantly higher than all other concentrations (P = 0.0137), (Appendix A).


Johnson et al., 2011 showed that the nonpathogenic strain of A. citrulli, AAC00-1ΔhrcC colonized female watermelon flowers, infested watermelon seeds at wild-type levels, and reduced BFB seed-to-seedling transmission. Similar results were obtained in this study. Watermelon flowers inoculated with 107, 108, and 109 AAC00-1ΔhrcC CFU/flower yielded seeds that resulted in BFB seedling transmission percentages of 3, 23, and 23%, respectively, when challenged with A. citrulli 213-60. Conversely flowers inoculated with water had a BFB seedling transmission percentage of 77.5%. One advantage of AAC00-1ΔhrcC as a biocontrol agent is that it can infest watermelon seed and localize in similar tissues as wildtype AAC00-1. However, AAC00-1ΔhrcC was generated by genetic manipulation and might not be suitable for commercial application. Hence, we screened germinating watermelon seeds for naturally occurring bacteria that might serve as biocontrol agents against A. citrulli. Out of 300 candidate bacteria, three strains were selected based on their ability to reduce BFB seed-to-seedling transmission when applied as a seed treatment. These three strains, identified as Bacillus species were tested as watermelon flower treatments in field trials in 2015.



Seeds from flowers inoculated with Bacillus mojavensis RRC101, Bacillus species #24 and #35 were challenged in a seedling grow out assay with 107 CFU/ml of A. citrulli 213-60. Seedlots from watermelon flowers inoculated with PBS had a mean BFB seedling transmission of 79.7%. In comparison, seedlots from watermelon flowers treated with Bacillus mojavensis RRC101, Bacillus spp. #24 and #35 had mean BFB seedling transmission percentages of 19.4, 53.2 and 41.7% respectively (Appendix B). The effect of the flower treatments on mean BFB seedling transmission percentage was significant (P = 0.04) and B. mojavensis was the most effective bacterial strain (Fig 1). These data suggest great potential for flower treatments with biocontrol agents to aid management of BFB. However, further optimization and evaluation is needed. While it is unlikely that this approach alone would eliminate BFB seedling transmission, if compatible with other management practices, it could be an important part of an integrated disease management program.

Participation Summary

Educational & Outreach Activities

Participation Summary

Education/outreach description:

S. Sutton, R. Walcott. 2015. Biological Control Seed Treatments for a Scourge of Watermelons. UGA Extension Spring Newsletter

Project Outcomes

Project outcomes:

Results of this study will provide an environmentally sound and sustainable approach for managing seed borne plant diseases. We identified naturally occuring watermelon seed microflora that can be used as biocontrol seed treatments. Some of these bacteria have broadd spectrum activity that may also limit colonization of seeds by phytopathogenic fungi. These may be of potential use for seed companies.


Areas needing additional study

The field study must be repeated in order to obtain more precise information. This approach also must be evaluated under standard commercial watermelon seed production conditions to ensure its compatibility. Efforts are also needed to characterize the mode of action of the biocontrol agents and optimize the inoculum load per seed after flower treatment.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.