Resistance of bread wheat (Triticum aestivum L.) varieties to fusarium head blight (Fusarium graminearum) in Ethiopia

Abstract

Fusarium head blight (FHB) is a devastating disease reduces wheat yield and quality. This study was aimed to evaluate wheat varieties' responses to spray inoculation with Fusarium graminearum mixture isolates under greenhouse conditions. The treatments were laid out in a randomized completely block design with three replications using 24 bread wheat varieties registered in Ethiopia. The varieties inoculated with mixed inoculum derived from four pathogenic isolates of F. graminearum that had been isolated from wheat kernels. Disease severity was evaluated using a 1–9 scale based on the proportion of bleached spikelets, and the area under disease progress curve (AUDPC) was determined from the disease severity data. At harvest, the kernel weight reduction was determined in comparison to the control. The evaluated traits were significantly interrelated and showed high and significant variation among (p < 0.0001) wheat varieties. Disease severity index among the varieties varied from 29% to 72%, while AUDPC varied from 326%- to 1010%-days. The disease progress rate of the Kingbird variety inoculated with F. graminearum was the slowest (0.0191 units day⁻¹), whereas Ogolcho had the fastest disease progression rate (0.0581 units day⁻¹). Kingbird, Wane, and Limu were moderately resistant, with lower disease severity, AUDPC, and a reduction in 1000-grain weight and grain weight per spike. Dereselgne, Dambal, and Ogolcho varieties were highly susceptible, with the greatest grain weight per spike reduction (53.2%, 41.4%, and 37.4%) and 1000-grain weight reduction (41.5%, 42.8%, and 37.5%), respectively. The results implied that there were different levels of FHB resistance in Ethiopian bread wheat varieties. Although current greenhouse evaluation of varieties gives encouraging results, field testing is required to confirm the current findings.

Abbreviations
• AUDPC
• area under disease progress curve
• DAI
• days after inoculation
• FDK
• fusarium damaged kernels
• FHB
• fusarium head blight
• GWS
• grain weight per spike
• SAS
• statistical analysis system
• SI
• severity index
• TGW
• 1000-grain weight

INTRODUCTION

Bread wheat (Triticum aestivum L) is one of the world's most important cereal crops production and nutritional value (Gómez et al., 2021). In Ethiopia, the area under wheat production is estimated 1.89 million ha (CSA, 2021), making the country the largest wheat producer in sub-Saharan Africa (5.78 million tonnes). Wheat is one of Ethiopia's most important crops, as it is also a source of food for farmers too (Minot et al., 2015). It is the preferred diet of Ethiopians, accounting for 14% of the caloric intake for over 90 million people (FAO, 2015), placing wheat second only to maize (Zea mays) (19%) and ahead of teff (Eragrostis tef), sorghum (Sorghum bicolor), and enset (Ensete ventricosum) (10%–12% each) (Anteneh & Asrat, 2020). However, wheat productivity in Ethiopia is low, which is 3.05 t ha⁻¹ (CSA, 2021) below world productivity (3.5 t ha⁻¹). Low productivity is attributed to a variety of factors, including biotic (diseases, insect pests, and weeds), abiotic (moisture stress and soil fertility), and lack of or slow adoption of new agricultural technologies (Zegeye et al., 2001). Among those biotic factors, Fusarium head blight (FHB) is among the most devastating wheat production problem in Ethiopia (Getahun et al., 2022).

FHB is caused by Fusarium species, which comprises 17 different species. Crop residues on the soil surface are the primary reservoir of FHB pathogens (Shaner, 2003). Inoculum can be ascospores, macroconidia, chlamydospores, or hyphal fragments (Bai & Shaner, 1994), but ascospores released from soil surface debris are the primary inoculum that initiates epidemics (Bai & Shaner, 1994; Shaner, 2003). Spores are transported by rain and wind to host floral tissues, where, at anthesis, they proliferate and spread rapidly intracellularly throughout the host spikelets, down into the rachial nodes, and ultimately up and down the rachides until FHB symptoms are clear, involving necrosis and bleaching of heads causing shriveled kernels (Dweba et al., 2017). The disease infection and fungal growth is favored by high temperature (25–30°C) and higher humidity (>90%), and abundant rain during the anthesis stage results in shriveled, lightweight kernels, reduced seed germination, seedling blight, and poor stands and can lead to significant yield losses (Khaledi et al., 2017; Parry et al., 1995). This disease can completely destroy the crop production of a potentially high-yielding crop within few weeks (McMullen et al., 1997). Furthermore, the disease produces various mycotoxins, including deoxynivalenol (DON) and zearalenone, which are toxic to humans and animals (Darwish et al., 2014). FHB severe epidemic outbreaks caused quantitative losses in yield of up to 50%–75% (Parry et al., 1995). In Ethiopia, yield losses due to FHB have been reported to be up to 50% depending on the susceptibility of the variety and environmental conditions (Mengesha et al., 2021). The cultivated Ethiopian wheat varieties have suffered from recurrent epidemics (Getahun et al., 2022). FHB on wheat has occurred in Ethiopia, with varied levels of incidence and severity depending on the cropping year, district, and wheat variety (Getahun et al., 2022; Kebede et al., 2021). The disease has been reported in the Oromia and South regions of Ethiopia, and there is evidence indicating that the disease is prevalent in many parts of the country (Getahun et al., 2022). In some districts, the mean disease incidence ranged from 22.28% to 53.35%, while the disease severity index ranged from 16.57% to 37.73% (Getahun et al., 2022).

Core Ideas

• Widely grown wheat varieties in Ethiopia were susceptible to fusarium head blight.
• Some wheat varieties were found to be potential sources of fusarium head blight resistance and improved agronomic performance.
• The study will provide wheat producers in Ethiopia with information on variety selection during planting.

Management practices, such as use of chemical method with spray frequency, are implemented to manage FHB in Ethiopia (Mengesha et al., 2021). However, no fully effective FHB fungicide is currently available (Haidukowski et al., 2012), and the application window is very narrow, lasting only a few days around host anthesis (Mesterhazy et al., 2003). The use of synthetic pesticides has disadvantages, according to Avis (2007), including a lack of long-term efficacy due to the development of plant pathogen resistance. Due to the limited application window, the potential of environmental contamination, and the higher cost of producing wheat, fungicide use for FHB control is still challenging (McMullen et al., 2012). As a result, there is a need for alternative disease management options that are environmentally friendly, accessible, and affordable to local farmers. Managing FHB through resistant varieties is the most effective, cost-efficient strategy, and sustainable means to mitigate against losses caused by FHB (Bai et al., 2018; Rudd et al., 2001).

The mechanisms of plant resistance to FHB are very complex, and now it is generally agreed that FHB resistance is controlled by a polygenic system (Wisniewska et al., 2004). Wisniewska et al. (2004) pointed to three different resistance components at least: resistance to the initial infection (Type I), pathogen spread (Type II), and kernel colonization (Type III). Furthermore, Type I and Type II resistances vary independently among genotypes (Schroeder & Christensen, 1963). Resistance to primary infection, type I resistance, is determined by spraying spikes at mid-anthesis with a conidial suspension and measuring the percentage of diseased spikes, whereas type II resistance is determined by inoculating single florets with conidial inoculum and measuring the number or percentage of diseased spikelets over time (Steed et al., 2022). Although type II resistance has been reported in some wheat lines, type I resistance appears to be more significant (Steffenson, 2003). Resistance of studied wheat varieties could be described by disease score on heads, percentage of Fusarium damaged grains, kernel weight per spike, and DON contents (Wisniewska et al., 2004).

While previous studies focused on inoculations with specific isolates, this study also described the use of selected isolates in mixture, which is used in some breeding programs as a response to pathogen populations with variation (Miedaner et al., 2008). The problem is even more worsened than anticipated, as released wheat varieties grown in Ethiopia infected by FHB disease. Furthermore, wheat varieties that are currently in production have not been evaluated for FHB resistance in Ethiopia.

Despite the importance of wheat FHB and its potential to cause significant yield losses, limited research works have been conducted in Ethiopia on pathogenicity test on wheat varieties of the disease. Kebede et al. (2020) used pathogenicity test through variety and fungal isolate interaction and demonstrated differences among fusarium isolates in their reaction to FHB in a wheat variety. However, an earlier pathogenicity study did not include the most popular, adapted, and cultivated varieties in Ethiopia and was done on only one variety. Furthermore, detailed research have not been conducted on FHB, Fusarium damaged kernels (FDKs), and grain weight per spike reduction (GWS-R) of using wheat varieties. Therefore, evaluating wheat varieties for resistance to FHB would help to identify economically important traits. Thus, the objective of this study was to evaluate the response wheat varieties to spray inoculation with a mixture of F. graminearum under greenhouse condition.

MATERIALS AND METHODS Plant growth condition

Twenty four bread wheat varieties that are under production were collected from Werabe Agricultral Research Center and planted in July 2021 at Haramaya University research station greenhouse site. Characteristic features of the wheat varieties are presented in Table 1. The experiments were undertaken in randomized complete block design with three replications. Seeds of wheat varieties were surface-sterilized in 1% (v/v) sodium hypochlorite, rinsed three times in sterile distilled water, and placed on sterile moist filter paper in a Petri dish. Seeds were vernalized for 3 weeks at 4°C prior to germination. After vernalization, seedlings were planted into sunshine in 15-cm round plastic pots containing a mixture of loam soil, sand, and composted manure (2:1:1 by volume) in a greenhouse (Shin et al., 2014). Pots were prepared as above using soil autoclaved at 148°C at 102 kPa for 45 min. Plants were thinned to five plants per pot (there were three to five tillers from each plant, which yielded at least 10 panicles per pot at the time of inoculation). During planting, 100 kg ha⁻¹ of NPS fertilizer was applied. While an N-fertilizer of 200 kg ha⁻¹ was applied, one-third of it was applied during planting and the other two-thirds was applied 35 days later.

TABLE 1 Characteristic features of wheat varieties used for resistance reaction of wheat Fusarium head blight (FHB) at Haramaya University research station greenhouse site, Ethiopia.

Variety	Pedigree	Year of release	Institution released	Days to maturity	Yield (t ha−1) when released	Reaction to major diseases
						FHB	Sr	Yr
Balcha	–	2019	KARC/ EIAR	115–120	3.5– 4.5	NA	MR	MR
Biqa	PASTOR//HXL7573/2*BAU/3/WBLL1	2014	KARC/EIAR	141	3.2–5.4	NA	MR	MR
Bondena	D67.2/P66.270//AE.SQUARROSA(320)/3/CUNNIN	2018	AARC/SARI	121–127	3.8–5.5	NA	MR	MR
Bulluq	“UTQUE96/3/PYN/BAU//MILAN ”	2015	BARC/OARI	85–88	6–6.5	NA	MR	NA
Daka	ATTILA/3*BCN*2//BAV92/3/KIRITATI/WBLL1/4	2018	KARC/EIAR	125–135	4–6	NA	MR	MR
Dambal	“AGUILAL/3/PYN/BAU//MILAN ”	2015	SARC/OARI	142	5.6–6.3	NA	MR	MR
Danda'a	KIRITATI//2*PBW65/2*SERI.1B	2010	KARC/EIAR	110–145	3.5–5.5	S	MR	MR
Dereselgne	CI 8154//2*FR	1974	KARC/EIAR	75	NA	NA	NA	NA
Hidase	YANAC/3/PRL/SARA//TSI/VEE#5/4/CROC-1/AE	2012	KARC/EIAR	121	4.5–7.0	MS	MR	MR
Hoggana	PYN/BAU//MILAN	2011	KARC/EIAR	121–170	4.6–6	NA	R	MR
Honqolo	NJORO SD-7	2014	KARC/EIAR	102	3.5–6.3	NA	MR	MR
Hulluka	UTQUE96/3/PYN/BAU//MILAN	2012	KARC/EIAR	133	4.4–7.0	NA	MS	MR
Inseno-1	–	2009	AwARC/SARI	100	3.5	NA	MR	MR
Kekeba	KIRITATI/SERI/RAYO	2010	KARC/EIAR	90–120	3.3–5.2	NA	MR	MR
Kingbird	TAM200/TUI/6/PVN//CAR422/ANA/5/BOW/CRO	2015	KARC/EIAR	90–95	4–4.5	MR	MR	MR
Tay	ET12D4/4777(2)//FKN/GB/3/PVN"S	2005	ADARC/ARARI	104–130	2.5–6.1	NA	MR	MR
Lemu	WAXWING*2/HEILO	2016	KARC/EIAR	140	5.5–6.5	NA	MR	MR
Liben	“UKULA/KAUZ/6/PSN/BOW/4/MAYA/NACO/3 ”	2015	BARC/OARI	122–125	5.5–6.5	NA	MR	MR
Mandoyu	WORRAKATTA/PASTOR	2014	SARC/OARI	139	4.9–5.9	NA	MS	MR
Mekele 01	–	2012	KARC/EIAR	113-125	3.4-4.7	NA	MR	MR
Obora	UTIQUE96/FLAG-1	2015	SARC/OARI	144	4.6-6.3	NA	MR	MR
Ogolcho	WORRAKATTA/2*PASTOR	2012	KARC/EIAR	102	3.3-5.0	MS	MR	MR
Shorima	UTQUE96/3/PYN/BAU//MILAN	2011	KARC/EIAR	105-150	4.4-6.3	MR	MR	MR
Wane	SOKOLL/EXCALIBUR	2016	KARC/EIAR	125	5-6	NA	MR	MR

Abbreviations: AARC, Areka Agricultural Research Center; ADARC, Adet Agricultural Research Center; ARARI, Amhara Regional Agricultural Research Institute; AwARC, Awassa Agricultural Research Center; BARC, Bako Agricultural Research Center; EIAR, Ethiopian Institute of Agricultural Research; KARC, Kulumsa Agricultural Research Center; MR, Moderately resistance; MS, Moderately susceptible; NA, Not available; OARI, Oromia Agricultural Research Institute; S, Susceptible; SARC, Sinana Agricultural Research Center; SARI, Southern Agricultural Research Institute; Sr, steam rust; Yr, yellow rust;

Fungal isolates and inoculum preparation

Twenty F. graminearum isolates from infected wheat from different area of surveyed districts were tested for aggressiveness based on inoculating spikes of Ogolcho, a susceptible wheat variety (Getahun et al., 2022), with a macroconidial suspension in the greenhouse. Fusarium isolates were selected with respect to the results of pathogenicity tests in greenhouse conditions on susceptible variety. Finally, four isolates that caused 100% disease severity were mixed together for inoculation tests (Engle et al., 2003). Conidial suspensions were produced in mung beans liquor medium (Bai & Shaner, 1996). To prepare this medium, 40 g of mung beans was placed in l-L Erlenmeyer flask containing 1 L boiling water. The beans in different flasks were boiled for 20 min and filtered through cheesecloth. The mung bean filtrate in each flask was equally subdivided into five 400-mL Erlenmeyer flasks, autoclaved, and inoculated with a 4-mm² disk from culture of F. graminearum grown on potato dextrose agar for 4 days and placed on shaker at 24°C for 4 days. After the incubation period, the liquid culture was stored in refrigerator. Inoculum was stored at −20°C and amended by addition of 0.05% Tween 20 prior to inoculation. For inoculation, spore suspensions were prepared by filtering the aqueous culture through two layer of cheesecloth and adding sterile water to adjust the inoculum concentration (Gosman et al., 2005). Conidia were counted under microscope with the aid of a hemacytometer. The concentration of macroconidia and ascospores in suspension for each isolate was determined using a hemacytometer and adjusted with sterile, distilled water to 5 × 10⁵ mL⁻¹. The suspensions of the four isolates were then combined to create the macroconidial or ascospore suspensions used to inoculate plants. Suspensions were prepared fresh daily, or 1-day-old suspensions were used that had been stored at room temperature (Engle et al., 2003).

Inoculation

The wheat plants in each pot were inoculated at mid-anthesis (growth stage 59; Zadoks et al., 1974), 5 mL of spore suspension (5 × 10⁵ conidia mL⁻¹) amended with 0.05% Tween 20 was sprayed onto the spikes of each plant with mixes of four aggressive isolates of F. graminearum. One-term spraying of inoculum (conidial suspension 5 × 10⁵ ML⁻¹) was applied. The spikes were sprayed uniformly with a 1-L hand sprayer from all sides, which is used to evaluate resistance to primary infection. As the varieties had different flowering times, the inoculation period lasted for 8–10 days. Following inoculation, the plants were placed in a moist chamber for three successive nights, and the plants were mist irrigated and covered in a plastic bag. The misting schedule once every 40 min for 3 days post inoculation to maintain high humidity. Temperature with the moist chamber was 23–25°C and relative humidity was 90%–95%. On the fourth day after inoculation, plants were returned to their original position on greenhouse benches. After incubation, plants were kept until maturity under the same light and temperature conditions used prior to inoculation. Control plants were sprayed with 5 mL of sterile water amended with 0.05% Tween 20 and covered with a cellophane bag for use in the calculation of relative spikelet weight. The average greenhouse temperature was 25°C during the day, with a range of 19–30°C, and 19°C at night, with a range of 17–21°C and an average relative humidity of 85%.

Disease assessment

Symptoms of FHB ranged from dark brown, water-soaked spot on the glumes to belched spikelets. Spikelets with any of these disease symptoms were recorded as diseased. Each head was visually rated for percent surface area with FHB symptoms at 10, 15, 20, and 25 days after inoculation (DAI). Head blight symptoms were evaluated on a 1–9 scale, 1 = no symptom, 2 = <5%, 3 = 5%–15%, 4 = 16%–25%, 5 = 25%–44%, 6 = 46%–65%, 7 = 66%–85%, 8 = 86%–95%, 9 = 96%–100% of the spikelets with FHB symptoms (Miedaner et al., 1996). FHB ratings on each assessment date per pot were converted to severity index in percentage as follows: [Image Omitted. See PDF]

Final disease severity was calculated as the percentage of disease spikelet per spike on 25 DAI. From the calculated severity index, area under disease progress curve (AUDPC) was computed and expressed in %-days (Campbell & Madden, 1990) using the formula: [Image Omitted. See PDF]where X_i is the disease severity at the ith assessment, ti is the time of the ith assessment in days from the first measurement date, and n is the total number of disease assessments made during the epidemic period.

The disease progress rate was calculated from SI using linear regression using the monomolecular ln [(1/1−y)] model (Van der Plank, 1963) to estimate the disease progression in each wheat variety (Campbell & Madden, 1990).

Grain assessment

After ripening, spikes from both inoculated and control plants were manually harvested in early October from each pot at a moisture content of 12.5% and threshed at a low wind speed to avoid the loss of low-weight infected kernels at the Zadoks growth stage of 90. Grain data, such as 1000-grain weight and grain weight per spike for inoculated and control plant, were collected. Tolerance to the infection was expressed as percent reduction (R) for thousand grain weight (TGW-R) and grain weight per spike (GWS-R) for each variety were calculated as ([C − I]/C) × 100, where C is the control treatment value and I is the inoculated treatment value (Chrpova et al., 2013). FDKs were calculated as the percentage of infected grains in each pot divided by the total number of grains harvested in each pot. Each pots seed weight was measured in thousands. Finally, each inoculated spike was separately taken to laboratory and re-isolation was performed to confirm the identity of the test pathogen.

Data analyses

FHB disease severity index (SI), AUDPC, FDK, TGW-R, and GWS-R components for each treatment were subjected to analysis of variance (ANOVA) using the PROC GLM procedure (SAS, 2019) to determine the response of wheat varieties to F. graminearum (cause of wheat FHB) inoculations. Duncan's multiple range test was used to separate mean differences between treatments at a 5% probability level (Gomez & Gomez, 1984). The correlation analysis based on Pearson's correlation coefficient was used to investigate the relationship between and among wheat FHB disease parameters and grain components. Linear regression was used to examine the association between disease severity and AUDPC with GWS-R. It estimates how GWS-R changes as the AUDPC and severity change. Linear regression analyses were appraised using Excel 2013.

RESULTS Fusarium head blight severity

Wheat varieties reaction to FHB assessed after inoculation with four selected F. graminearum isolates, used in mixture, are presented in Table 2. The severity index of FHB on the varieties differed significantly (p < 0.0001). On susceptible varieties, symptoms began to appear on 4 DAI and the disease development was faster. The final FHB severity index assessed at 25 DAI ranged from 29% to 71.7% (Table 2). The first FHB symptoms, dark brown symptoms usually extend into the rachis, even down into the stem tissue as the fungus spreads within a spike, were observed on the varieties Dereselgne, Dambal and Ogolcho at 4 DAI. The highest disease severity (71.7%) was recorded from Dereselgne variety, followed by Dambal (68.6%) and Ogolcho (68%). Conversely, the lowest (29%) FHB severity was indicated on the variety Kingbird, followed by the variety Wane (35.7%) and Limu (36.7%), which was better than Bulluk and Shorima. Planting varieties Kingbird,Wane, and Limu varieties reduced FHB severity by 58.7%, 50.2%, and 48.8%, respectively, when compared to variety Dereselgne.

TABLE 2 Severity index (SI) and area under disease progress curve (AUDPC) of fusarium head blight after inoculation, Fusarium-damaged kernels (FDKs), reduction of thousand grain weight (TGW-R), and reduction of grain weight per spike (GWS-R).

Variety	SI (%)	AUDPC (%-days)	FDK (%)	TGW-R (%)	GWS-R (%)	Rank (five traits)
Balcha	55.0	541.7	63.3	25.8	29.9	12
Biqa	65.9	780.5	71.7	32.3	35.9	20
Bondena	53.5	645.7	62.7	22.7	26.7	12
Bulluk	38.3	430.5	54.0	15.4	20.3	3.6
Daka	53.7	504.3	60.0	18.9	28.7	8.4
Dambal	68.6	899.3	81.0	42.8	41.4	23.2
Danda'a	65.0	696.7	68.3	28.6	34.8	19
Derselegn	71.7	1009.8	85.0	41.5	53.2	23.8
Hidase	66.4	865.3	78.0	33.6	37.3	21
Hoggana	51.9	544.2	63.0	18.8	31.1	10.2
Honqolo	40.1	487.2	57.3	17.6	22.9	6.2
Hulluka	61.8	632.7	66.7	20.6	33.6	14.6
Inseno-I	63.8	627.5	66.0	20.7	25.4	12.6
Kekeba	56.2	629.5	63.3	22.2	30.0	13.4
Kingbird	29.0	325.7	38.0	15.1	11.9	1
Liben	46.2	516.5	61.0	21.9	22.2	8.4
Limu	36.7	400.0	50.0	15.5	17.8	3.2
Mandoyu	55.6	588.7	65.0	21.8	29.7	12.2
Mekele 01	63.5	634.0	66.0	21.2	29.9	14
Obora	64.7	672.3	67.3	26.1	33.9	18
Ogolcho	68.0	870.0	79.7	37.4	37.5	22
Shorima	39.7	482.0	56.0	16.1	22.4	5.2
Tay	61.9	581.3	66.3	19.9	30.5	13.2
Wane	35.7	373.5	44.0	15.4	13.1	2
LSD (0.05)	12.5	111.8	9.1	10.5	5.6
CV (%)	13.8	11.1	8.7	26.8	11.7
p-value	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001

Note: Severity index was assessed at 25 days after inoculation.

Abbreviation: CV, coefficient of variation; LSD, least significant difference.

FHB disease progress rate

Disease progress rates, parameter estimates, and intercepts of FHB are presented in Table 3. Mean disease progress rates showed significant variations among the varieties evaluated for reaction to FHB (Table 3). Disease progress rates ranged from 0.0191 to 0.0581 units day⁻¹ were computed. Compared to other varieties, faster disease progress rates were calculated from varieties, Ogolcho, Dereselgne, Tay, Obra, and Mekele 01, which were 0.0581, 0.0575, 0.0571, 0.0565, and 0.0556 units day⁻¹, respectively. Variety Kingbird, Limu, Bulluk, Honqolo Shorima, and Wane showed slower disease progress rates which were 0.0191, 0.022, 0.0228, 0.0234, 0.0242, and 0.0252 units day⁻¹, respectively compared to other varieties (Table 3). The rate of disease development did not always correspond with the final disease severity level. For example, disease development rate on the variety Ogolcho was higher than Dereselgne. However, the final disease severity on Ogolcho was less than Dereselgne. Similarly, the rate of disease development on variety Tay was higher than Hidase, but the final disease severity (61.9%) on Tay was less than Hidase (66.4%). Variety Kingbird recorded 67.1%, 66.8%, and 66.5% reduction in disease progress rate compared with Ogolcho, Dereselgne, and Tay varieties, respectively. Likewise, planting Limu variety obtained 62.1%, 61.7%, and 61.4% disease progress rate reduction as compared to Ogolcho, Dereselgne, and Tay variety in that order.

TABLE 3 Effect of wheat varieties on Fusarium head blight disease progress rate.

Variety	Intercept	SE of intercept	Disease progress rate (units day−1)	SE of rate	R2 (%)
Balcha	−0.361	0.074	0.0450	0.0040	92.6
Biqa	−0.331	0.089	0.0555	0.0048	92.9
Bondena	−0.181	0.056	0.0377	0.0030	93.9
Bulluk	−0.107	0.048	0.0228	0.0026	88.5
Daka	−0.348	0.103	0.0423	0.0056	85.1
Dambal	−0.224	0.141	0.0551	0.0077	83.8
Danda'a	−0.361	0.131	0.0528	0.0072	84.5
Derselegn	−0.135	0.157	0.0575	0.0088	79.5
Hidase	−0.196	0.054	0.0515	0.0030	96.3
Hoggana	−0.279	0.094	0.0398	0.0053	83.9
Honqolo	−0.077	0.079	0.0234	0.0043	74.4
Hulluka	−0.419	0.121	0.0535	0.0066	86.9
Inseno-I	−0.451	0.135	0.0553	0.0074	84.9
Kekeba	−0.331	0.157	0.0472	0.0092	72.5
Kingbird	−0.116	0.037	0.0191	0.0020	89.9
Liben	−0.177	0.113	0.0316	0.0061	72.6
Limu	−0.112	0.031	0.0220	0.0017	94.6
Mandoyu	−0.329	0.178	0.0451	0.0097	68.3
Mekele 01	−0.449	0.165	0.0556	0.0089	79.4
Obora	−0.434	0.119	0.0565	0.0065	88.3
Ogolcho	−0.288	0.280	0.0581	0.0153	59.2
Shorima	−0.093	0.035	0.0242	0.0019	94.1
Tay	−0.506	0.100	0.0571	0.0055	91.6
Wane	−0.181	0.044	0.0252	0.0024	91.8

Note: Disease progress rate obtained from regression line of severity index (%) against time of disease assessment (days).

Abbreviations: R², coefficient of determination for the monomolecular epidemiological model; SE, standard error.

Area under disease progress curve

AUDPC values for the tested wheat varieties are presented in Table 2. The AUDPC was used to summarize FHB development on various wheat varieties. The AUDPC of FHB on the varieties differed significantly at (p < 0.0001). High AUDPC values of 1009.8%-, 899.3%-, and 870%-days were computed from severity assessments of Dereselgne, Dambal, and Ogolcho varieties, respectively. However, the lower values of 325.7%-days (Kingbird), 373%-days (Wane), 400%-days (Limu), and 430.5%-days (Bulluk) were recorded from pots of each wheat variety. Similarly, planting Kingbird Wane, Limu, and Bulluk lowered AUDPC by 67.7%-, 63%-, 60.4%-, and 57.4%-days, respectively as compared to variety Dereselgne.

TGW and GWS-R

The resistance level of host varieties accounted for the highest proportion of the variation in TGW reduction (TGW-R) and GWS-R. The tested wheat varieties differed significantly (p < 0.0001) in GWS-R and TGW-R indicating broad variation among the varieties (Table 2). The percentage of variation due to host genotype ranged from 11.9% to 53% (GWS-R). Examination of the most important characteristic, effect on grain yield, however, showed significantly lower reduction of GWS on the variety Kingbird (11.9%), Wane (13.1%), and Limu (17.8%), while average GWS-R values for the varieties (Shorima, Honqolo, Daka, and Liben) and susceptible varieties (Ogolcho, Dambal, and Derselgn) were 37.5%, 41.4%, and 53.2%, respectively. As shown in Table 2, variation in GWS-R was especially high in the moderately resistance groups and susceptible varieties. Planting Kingbird,Wane, Limu, and Bulluk lowered GWS-R by 77.6%, 75.4%, 66.5%, and 61.8%, respectively, as compared to variety Dereselgne.

Concerning TGW-R, high TGW-R was recorded on variety Dambal (42.8%), Dereselgne (41.5%), and Ogolcho (37.4%), while low 15.1%, 15.4%, 15.4%, and 15.5% values of TGW-R were recorded from the wheat varieties Kingbird,Wane, Bulluk, and Limu, respectively. Planting Kingbird,Wane, and Limu reduced TGW-R by 64.7%, 64%, and 63.8%, respectively, when compared to Dambal.

Fusarium damaged kernels

ANOVA of FDK showed significant (p < 0.0001) variations among the varieties (Table 2). The varieties Kingbird, Wane, and Limu showed small shriveled kernels. Mean FDKs ranged from 38% to 85%. The highest (85%) FDK was recorded on the variety Dereselgne, followed by Dambal (81%), Ogolcho (79.7%), and Hidase (78%). The lowest FDK was recorded on the varieties, such as Kingbird (38%), Wane (44%), and Limu (50%). Planting Kingbird reduced FDK by 55.3%, 53.1%, 52.3%, and 51.3% compared to varieties Dereselgne, Dambal, Ogolcho, and Hidase, respectively.

Correlation among and between SI, AUDPC, FDK, TGW, and GWS-R

Coefficients of correlation between the examined traits obtained with different parameters are shown in Table 4. The majority of correlations between FHB assessments used to evaluate disease reactions were significant and positively related (r = 0.342 − 0.883). Among them, the correlation coefficient (r) between FDK and SI was the highest, and between FDK and AUDPC was second highest. Severity index correlated more closely with FDK (r = 0.883), AUDPC (r = 0.817***), and GWS-R (r = 0.741***) than with TGW-R (r = 0.342**) (Table 4). The AUDPC maintained positive association (r = 0.876***) and (r = 0.825***) with FDK and GWS-R, respectively. According to the regression equation indicated in Figure 1, for every unit increase in AUDPC and severity, there was a 0.0378 and 0.441 unit increase in grain weight per spike reduction, respectively.

TABLE 4 Pearson correlation coefficients for correlation of Fusarium head blight (FHB) severity index (SI), area under disease progress curve (AUDPC), thousand grain weight reduction (TSW-R), Fusarium-damaged kernels (FDKs), and grain weight per spike reduction (GRS-R) of FHB inoculated wheat under greenhouse conditions.

	SI	AUDPC	TGW-R	FDK	GWS-R
SI	–	0.817***	0.342**	0.883***	0.741***
FDK	0.883***	0.876***	0.365**	–	0.770***
AUDPC	0.817***	–	0.467***	0.876***	0.825***
TGW-R	0.342**	0.467***	–	0.365**	0.379**
GWS-R	0.741***	0.825***	0.379**	0.770***	–

** and *** denote correlation coefficients significant at p < 0.01 and p < 0.0001.

FIGURE 1. Linear regression relating wheat Fusarium head blight (FHB) severity index and area under disease progress curve (AUDPC) with grain weight per spike reduction (GWS-R) of in FHB infected wheat varieties under greenhouse conditions.

Kingbird demonstrated the highest resistance as determined on the basis of all examined traits (average ranking 1), followed by Wane, Limu, Bulluk, Shorima, Honqolo, Daka, and Liben (average rankings 2–8.4). Above average or medium performance in a majority of traits was characteristic of 10 varieties with rankings 10.2–18. The other tested materials (rankings 19–23.8) can be considered as moderately susceptible or susceptible to FHB (Table 2).

DISCUSSION

The most practical approach for minimizing economic losses from wheat FHB is to develop FHB-resistant wheat varieties. Wide ranges of mean values were also characteristic of the varieties, indicating their high differences in resistance due to the varieties inherent genetic resistance factors against the disease. Our results were consistent with findings by Muthomi et al. (2007) who tested eight different Kenyan wheat varieties and found all susceptible. However, the varieties differed in susceptibility disease severity and kernel weight reduction. In the current study, lower values for disease parameters were obtained for the moderately resistant varieties, such as Wane, Kingbird, and Limu. Correspondingly, Šíp et al. (2011) reported all isolates and the mixture of F. graminearum isolates detected the lowest values of all traits in the moderately resistant cultivar; however, there were obvious differences among isolates in the ability to detect significant differences among varieties that differ greatly in FHB resistance. On the other hand, Goswami and Kistler (2005) reported significant variation among the strains of F. graminearum species in their ability to cause FHB on wheat. The carefully selected mixture of isolates with varying properties is an alternative to the parallel use of specific isolates, which could bring tests closer to natural infections while lowering costs (Šíp et al., 2011). From our findings, no complete resistance reaction was observed in any of the varieties evaluated. In accordance to Shen and Ohme (2006), there are no wheat varieties with high levels of FHB resistance, though some varieties have usable levels of partial resistance that limit yield loss and mycotoxins accumulation. In contrast to our findings, Mesterhazy et al. (2005) found either no disease or only traces of infection in the most resistant varieties. In the current study, symptom expression, percentage of FDK, AUDPC, TGW-R, and GWS-R due to infection were evaluated as disease resistance indicators. According to Mesterhazy et al. (1999), in order to better understand the complex nature of FHB resistance, not only FHB visual symptoms, but also TGW-R, FDK, and yield loss traits should be investigated.

With regard to disease severity, significant (p < 0.0001) variations were indicated among the evaluated wheat varieties to disease reactions. Disease severity has been used to assess F. graminearum resistance in the spike (Engle et al., 2003). Lower percentage of infected spikelets in a spike during the whole examined period was characteristic of the moderately resistant variety Kingbird. The variety (Kingbird) had statistically less symptomatic inoculated spikelets than other varieties. This could imply that Kingbird was a variety that had sufficient type I resistance to be detected by these methods. Additionally, in the varieties tested, anthers apparently had little resistance to infection, whereas glume tissues may harbor some type of resistance that limits infection. This is in agreement with the findings of Engle et al. (2003), where Ernie (resistant) had statistically less symptomatic inoculated spikelets than the susceptible control. Similarly, Muthomi et al. (2007) in Kenya reported the mean disease severity ranged from 9% for the least susceptible to 51.6% for the most susceptible wheat cultivar. In other study, Engle et al. (2003) reported the reaction of genotypes to F. graminearum varied with the procedures used to inoculated spike tissues. Gosman et al. (2010) indicated for inoculations with F. culmorum, variation of disease development between the most resistant and the most susceptible varieties. In agreement with our findings, Khanal et al. (2021) found resistant genotypes to have low FHB severity, ranging from 24% to 39%. Wheat variety characteristics and environmental conditions are known to influence and cause variations in plant disease development, either by affecting the host, the pathogen, or even their interactions. According to Madden et al. (2008), other factors influencing the final disease level include host tissue availability, temperature, moisture, and crop plant resistance level.

In this study, the AUDPC for Kingbird was lower than the susceptible variety Derseslign. Our findings are supported by the findings of Cazal-Martinez et al. (2020), who found that resistant genotypes had lower AUDPC than susceptible genotypes. Similarly, Muthomi et al. (2007) found that the AUDPC ranged from 140.5%- to 956%-days for the least and most susceptible Kenyan wheat varieties, respectively. The current study was in agreement with Mendes et al. (2018), who found that resistant genotypes had lower AUDPC than susceptible genotypes. Our study has shown that Kingbird exhibited low levels of damage in kernels caused by Fusarium species, which might be due to a resistance mechanism. These findings are supported by other studies, including Mendes et al. (2018), who found a significant difference between genotypes, with the mean FDK ranging from 39% to 97%. It is recommended that in a breeding program, selecting for fewer scabbed spikelets and fewer scabbed kernels will result in the selection of genotypes with low DON levels.

There was high correlation of FHB disease severity index with FDK, AUDPC, and GWS-R at p < 0.0001. Therefore, it can be used as a criterion for indirect selection of disease resistance in breeding programs. In the current study, correlation coefficients ranged from 0.342 to 0.883. Similarly, as in the field experiments performed by Šíp et al., 2010, the correlation of FHB symptoms with yield losses (r = 0.56–0.78). All varieties under study presented positive and significant correlations between FHB severity and FDK (r = 0.883; p < 0.0001). Mesterhazy et al. (2005) found highly significant correlations between FHB, FDK, and yield loss, which is consistent with our findings. Nopsa (2010) also reported similar correlations for winter wheat. Gosman et al. (2010) reported that the coefficient of correlation for the relationship between pathogens was moderate (r = 0.52, p = 0.003).

CONCLUSIONS

There are significant differences between wheat varieties, regardless of the isolate complex's ability to cause FHB in wheat. Some wheat varieties were found to be potential sources of FHB resistance and improved agronomic performance; this could serve in the development of superior high-yielding and disease-resistant varieties, with high implications for the disease control. Furthermore, the study will provide wheat producers in Ethiopia with information on variety selection during planting. Although the current results from variety evaluated in greenhouses are encouraging, more studies on field evaluation are needed to confirm the current findings at the field level (out of greenhouse) and determine the type and nature of resistance in wheat varieties.

AUTHOR CONTRIBUTIONS

Muluken Bekele: Conceptualization; data curation; formal analysis; methodology; software; writing—original draft; writing—review and editing. Chemeda Fininsa: Conceptualization; investigation; methodology; supervision. Abdi Hassen: Conceptualization; investigation; supervision Zelalem Bekeko: Conceptualization; investigation; supervision.

ACKNOWLEDGMENTS

The authors would like to acknowledge the Southern Agricultural Research Institute (SARI) for funding this study. We thank Haramya University and Werabe Agricultural Research Center for providing laboratory facilities and wheat seed.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.