38
SEED TESTING INTERNATIONAL   www.seedtest.org
RULES DEVELOPMENT •
and other organisms or samples which do not 
contain the target.
Sixty-three isolates (44 target and 19 non-
target isolates) were selected and plated on 
two non-selective media (PDA and MA). When 
necessary, isolates were subcultured on SNA and 
CLA to enhance sporulation. Based on colony 
morphology, analytical specificity was validated 
and reached 100% performance for both 
inclusivity (morphology found as expected for the 
target pathogens) and exclusivity (morphology 
different to the prescribed morphology for the 
target pathogens) on both media.
The analytical sensitivity is the lowest quantity 
or concentration of a pest that can be reliably 
detected with a given analytical method.
The performance of the analytical sensitivity was 
based on the ability to detect one infected seed in 
a sample of 399 seeds (0.25% of contamination). 
Thirty-eight seed samples of wheat, barley or 
oat containing 399 seeds free from F. culmorum 
and F. tricinctum were spiked with one seed 
contaminated by one of these Fusarium 
(culmorum or tricinctum). Examination resulted 
in 100% detection of each contaminated seed 
in all the samples. Analytical sensitivity was 
therefore validated.
Robustness is the ability of a method to produce 
results that do not vary, even if there are small 
parameter variations in the method.
To evaluate the robustness, three parameters 
were studied (light conditions, duration of 
incubation and plating design). For the light 
conditions, incubation under darkness and  
NUV were compared on both media (PDA  
and MA) and resulted in having better 
sporulation most of the time under NUV.  
The results on the duration of incubation  
(6, 7 and 9 d), led us to propose that the 
examination is best to be executed after 7 d 
incubation. Plating design was validated with  
no differences for both conditions: five and ten 
seeds per plate.
Comparative Test
To demonstrate the repeatability, reproducibility 
and diagnostic performance of the method a CT 
was performed. In a CT it is extremely important 
that participants execute the protocol on 
provided samples, adhering as strictly as possible 
to the protocol.
Selection and Quality Monitoring of 
Test Sample Sets
The CT was conducted in seven globally 
dispersed laboratories between March and 
April 2023. Each laboratory received eight coded 
Fusarium species isolates for identification 
and ten samples consisting of three highly 
contaminated oat seed samples positive for  
F. poae (A); four medium-contaminated wheat 
seed samples positive for F. poae and  
F. avenaceum (B); and three wheat seed samples 
negative for F. poae and F. avenaceum (D). Prior 
to shipment of the samples, the level of seed 
lot contamination and the homogeneity were 
evaluated. All selected seed lots were proven to 
be homogeneous.
A stability test was conducted after the last 
participant had indicated having begun the 
experiment, and was validated with stable 
contamination percentages over time.
CT Results
Isolate identification
The media, the incubation conditions and 
the morphological criteria allowed the seven 
laboratories to identify all the different species of 
Fusarium.
Detection on seed samples
For each of the seven participating laboratories, 
all negative samples were detected as negative 
(0+/3) as expected, and all the expected Fusarium 
species were found to be positive in the samples 
(F. poae: 7+/7; F. avenaceum: 4+/4). Therefore, 
accordance (repeatability) and concordance 
(reproducibility) of results for the negative and 
positive seed lots were calculated according to 
Langton et al. (2002) and the result was 100% for 
both performance criteria.
The results of the seven participating laboratories 
were analysed with Hampel’s method and 
boxplot (Fig. 3) for medium and high levels.
Conclusion
All the performance criteria were validated 
by this study. The CT organised with seven 
participants allowed the reproducibility of the 
method to be evaluated. Nevertheless, the CT 
results on naturally infected seeds showed that 
the identification of the species by morphological 
criteria is difficult. The difficulty is related to 
the presence of several different species and 
not due to the lack of laboratories’ expertise. 
All laboratories identified the species by 
morphological criteria for the target isolates.
This conclusion led us to propose that the 
expression of results will:
• Specify the identified Fusarium species
and/or
• Be given in complex following Crous et al. 
(2021), O’Donnel et al. (2022)
and/or
• Be given as Fusarium sp. when the identification 
is not possible
• Or be a mix of these expressions depending on 
the Fusarium species present
 
To help with the identification of Fusarium 
species, this method will be enriched in the 
coming years through an ongoing project, by an 
optional identification using barcoding.
References
1.	Crous, P.W., Lombard, L., Sandoval-Denis, M., 
Seifert, K.A., Schroers, H.-J., Chaverri, P. et al. 
(2021). Fusarium: more than a node or a foot cell. 
Studies in Mycology, 98, 1–184.
2.	Dean, R., Van Kan, J.A.L., Pretorius, Z.A., 
Hammond-Kosack, K.E., Di Pietro, A., Spanu, 
P.D. et al. (2012). The top ten fungal pathogens 
in molecular plant pathology. Molecular Plant 
Pathology, 13, 414–430.
3.	Desjardins, A.E. (2006). Fusarium Mycotoxins: 
Chemistry, Genetics, and Biology. American 
Phytopathological Society (APS Press), St. Paul, 
MN, USA. 260 pp.
4.	ISTA (2020). 7-022: Detection of Microdochium 
nivale and Microdochium majus in Triticum 
spp. (wheat) seed. International Rules for Seed 
Testing, 2020: Validated Seed Health Testing 
Methods. International Seed Testing Association, 
Wallisellen, Switzerland.
5.	Langton, S.D., Chevennement, R., 
Nagelkerke, N. and Lombard, B. (2002). 
Analysing collaborative trials for qualitative 
microbiological methods: accordance and 
concordance. International Journal of Food 
Microbiology, 79(3), 175–181.
6.	Nelson, P.E., Toussoun, T.A. and Marasas, 
W.F.O. (1983). Fusarium Species: An Illustrated 
Manual for Identification. Pennsylvania State 
University Press, USA. 183 pp.
7.	O’Donnell, K., Whitaker, B.K., Laraba, I., 
Proctor, R.H., Brown, D.W., Broders, K. et al. 
(2022). DNA sequence-based identification of 
Fusarium: a work in progress. Plant Disease, 
106(6), 1597–1609.
Figure 3. Boxplot comparison of homogeneity, participating laboratories and stability test at high 
level (Fusarium poae)

View this content as a flipbook by clicking here.