To further confirm the effect of the pathway of invasion on A.
citrulli localization in watermelon seeds, we investigated the effect
of removing seed layers on BFB seedling transmission. Samples
(n = 100 seeds/treatment) from pistil-inoculated and pericarpinoculated
lots were dissected as described above. Treatments in
this study included (i) removal of the testae; (ii) removal of the
testae and PE layer tissues; and (iii) intact seeds (positive
control). Each seed was planted in a capped, sterile 10-ml test
tube containing cheese cloth saturated with sterile deionized
distilled water and maintained at 85% RH. Seeds were incubated
at 28°C and 85% RH with continuous fluorescent light, and seed
germination and BFB seedling transmission percentages were
determined after 14 days. Germination and BFB seedling transmission
percentages were compared by factorial analysis using
the Student’s t test (P < 0.05) in SAS (version 8.1 for Windows;
SAS Institute Inc.). To confirm that putative BFB seedling symptoms
were caused by A. citrulli, bacterial isolation was attempted
from at least five symptomatic seedlings from each treatment.
Small (2 to 4 mm2) pieces of symptomatic cotyledon tissue were
macerated in 200 μl of sterile PBS and a loopful (≈20 μl) of
macerate was streaked onto Nunhems medium. After incubation
for 5 days at 28°C, putative A. citrulli colonies (round, red colonies
with smooth margins) were tested by real-time PCR assay using
an A. citrulli-specific TaqMan assay (10).