YIELD PENALTY ASSOCIATED WITH STACKING RESISTANCE TO LATE LEAF SPOT, ROSETTE DISEASES AND DROUGHT STRESS IN GROUNDNUT (ARACHIS HYPOGAEA L.)

Khalid Elsiddig Mohammed 1 , Eric E. Agoyi 4 , Thomas L. Odong 1 , Belay Miesho 3 , David K. Okello 2 , Olupot Grigon 1 , Patrick R. Rubaihayo 1 and Patrick Okori 1 . 1. College of Agriculture and Environmental Sciences, Makerere University, P. O. Box 7062, Kampala, Uganda. 2. Department of Groundnut Breeding, National Semi-Arid Research Resources Institute, P. O. Box Soroti, Uganda. 3. College of Agriculture , Aksum University-Shire campass, P.O.Box 314, Shire, Ethiopia. 4. Non Timber Forest Products & Orphan Crop UNIT, Laboratory of Applied Ecology, University of Abomeycalavi, 05 P.O Box 1752 Cotonou, Bénin. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History Received: xxxxxxxxxxxxxxxx Final Accepted: xxxxxxxxxxxx Published: xxxxxxxxxxxxxxxx

Groundnut production is constrained by late leaf spot, groundnut rosette disease and drought which are responsible for up to 100% yield loss. This study was conducted to determine yield penalty associated with stacking resistance to late leaf spot, rosette and drought stress in groundnut genotypes. Twenty eight genotypes comprising single, double and multiple resistances for the diseases and tolerant to drought were evaluated in screen houses at Namulonge and Kabanyolo, in 2017. Completely randomized design was used in two replications. Three watering regimes were applied. Diseases severity was scored at harvest based on 1-9 scale. Drought was assessed with relative water content and leaf membrane stability index. Yield penalty was calculated as yield of the resistant genotype under optimum conditions minus yield of the same genotype under stress. The results of analysis of variance showed significant difference (P < 0.001) from one location to another and genotype-by-location effect was significant for most traits. The highest pod yield was observed at optimum conditions and under disease control, while the lowest was observed under the combination of diseases and drought. The highest yield penalty (19.98) was recorded by rosette resistant genotype SGV 0007. Low yield penalty (6.6), due to leaf spot and rosette diseases, was observed for genotype Abutalata. Penalty was positively correlated (r = 0.24) with relative water content and rosette and negatively correlated with leaf spot. Genotypes which showed low yield penalty could be valuable genetic materials for breeding of groundnut resistance to multiple stresses in Uganda or similar environments.

ISSN: 2320-5407
Int. J. Adv. Res. 7 (6), XX-XX Introduction:-Groundnut (Arachis hypogaea L.) production is constrained by several biotic stresses (Maiti, 2002). These include foliar fungal diseases such as early leaf spot (ELS) caused by Cercospora arachidicola Hori, late leaf spot (LLS) caused by Cercospora personata Berk & Curt and rust caused by Puccinia arachidis Speg (Subrahmanyam et al., 1985); viral diseases such as groundnut rosette disease (GRD), peanut bud necrosis and stem necrosis (Lynch, 1990) and soil-borne diseases such as stem rot, collar rot and pod rot complexes are the most prevailing diseases in East and Central Africa (Okello et al., 2010). In Uganda, fungal diseases, particularly late leaf spot is the key constraint which affects the production of groundnut and results in up to 70% yield reduction (Okello et al., 2013). Also GRD is the most destructive disease to the groundnut production, and it can cause yield losses of up to 100% depending on the growth stage at which infection occurs (Okello et al., 2010). In addition to biotic stresses, groundnut is mainly grown under rain-fed conditions and its production depends on rainfall and rain distribution that are usually unpredictable (Reddy et al., 2003). The unpredictability of drought implies that improved groundnut genotypes should perform well not only under water limited conditions, but also when rainfall is adequate. Drought stress has adverse influence on water relations (Babu and Rao, 1983), photosynthesis (Bhagsari et al., 1976), growth and yield of groundnut (Suther and Patel, 1992) and relative water content and rate of transpiration of the groundnut (Kambiranda et al., 2011). This results in a drastic reduction of crop yield, depending on groundnut cultivars. Varieties producing high yield under drought conditions are thus required.
Breeding for high-yielding, foliar disease resistant genotypes requires identification of resistant cultivars with good breeding potential. It is possible to combine resistance to more than one disease in the same genetic background (Anderson et al., 1986), since the resistance to the different diseases is under different genes action. Unfortunately, many traits that are associated with resistance to pathogens reduce plant fitness, although others do not (Bergelsen, 1996). In a case study carried out in Germany, the resistant varieties possessed a lower yield potential than susceptible ones under disease-free conditions (BSA, 2012). Less susceptible varieties yielded about 5-7% lower than the highest-yielding susceptible varieties under disease-free conditions (Gummert et al., 2015). Mechelke, (2000) reported that resistant varieties presented 10-18% yield penalty in the absence of the pathogen, and thus too low to gain acceptance for cultivation on a large scale on commercial farms.
When genetic resistance is incorporated into cultivars against prevailing biotic and abiotic stresses, there is usually cost on yield potential (Knauft and Wynne, 1995) as cultivars that maximize stress tolerance tend to pay a penalty in yield production (Zavaleta et al., 2010). The costs of resistance contain some of the negative effects on plant fitness that may be caused by a resistance trait under natural growing conditions (Heil, 2002). Moderate to high resistance to LLS, GRD and drought have been developed in numerous groundnut genotypes. There is lack of understanding of the effect/cost of these diseases resistance genes and drought tolerance genes on groundnut yield. Therefore, this study was conducted to determine yield penalty associated with stacking resistance to LLS, GRD and drought stress in groundnut.

Plant materials and experimental design
Twenty eight groundnut genotypes including parents and crosses showing single, double and multiple resistance to LLS, GRD and drought were exposed to artificial infestations with the diseases and drought stress according to their resistance capacity ( Table 1). The experiments were carried out in the screen houses at National Crop Research Resources Institute (NaCRRI) and Makerere University Agricultural Research Institute Kabanyolo (MUARIK) between 11 th January and 10 th may 2017. Three watering regimes: 80% (T1), 60% (T2) and 40% (T3) of soil field capacity (FC) were imposed based on moisture meter reading. The experiments were conducted in two replications using completely randomized design. The controls were irrigated with 80% FC (T1) with no disease infestation.

Infestation with Groundnut Rosette Disease (GRD)
Greenhouse infestation with virulent aphids was done following the method developed by Kayondo, et al (2014). Aphids were obtained from groundnut plants infested with GRD from the field as evidenced by green and chlorotic rosette symptoms and were transferred on to disease free plants of susceptible genotypes JL 24 and Acholi white, 14 days after planting in the cages for mass rearing and for maintenance of large stocks of virulent aphids needed for artificial infestation during the experiment. Two weeks after germination of the experimental plants, infector rows (Acholi-white and JL-24) were placed between each two rows of the tested genotypes and aphids were free to move and find suitable plant hosts.

Infestation with late leaf spot (LLS)
Infested leaf debris with Cercospora personata was collected from the field and stored in cloth bags in farm shed for the use in the experiments. Spores were prepared from the stored debris (Hemantkumar, 2005) and sprayed uniformly 15 days after planting following the method developed by Ibrahim, (2010).
Drought was assessed in terms of relative water content and leaf membrane stability index as follows: The relative water content (RWC) was recorded from four leaflets of the third fully expanded leaf from the top of the main stem for each pot two weeks after flowering according to Iqbal and Bano, (2009). The leaves were picked in the morning (9-11 a.m) and taken to the laboratory and leaf fresh weight (FW) recorded. Each leaf sample was soaked in distilled water for 8 hrs and blotted for surface drying and water saturated leaf weight (TW) was recorded. The samples were then oven-dried at 60 0 C for 6 hrs to a constant weight and leaf dry weight (DW) recorded. RWC was calculated based on the formula suggested by Gonzalez and Gonzalez (2001) as follows.
Where: FW is the sample fresh weight, TW is the sample turgid weight and DW is the sample oven dry weight. The leaf membrane stability index (LMSI) was determined according to Sairam, (1994). Leaf discs (0.5g) from the third fully expanded leaflet collected at pod filling stage of uniform diameter were put in test tubes containing 10 ml of double distilled water in two sets. Test tubes in one set were kept at 40 0 C in a water bath for 30 min and electrical conductivity of the sample was recorded (C 1 ) using an electric conductivity (EC) meter and the test tubes in the other set were incubated at 100 0 C in a water bath for 15 min and their EC recorded (C 2 ). LMSI was calculated using the formula proposed by Sairam, (1994).

Data analysis
Analysis of variance (ANOVA) were performed in GENSTAT statistical package 16th edition (Payne et al., 2013).

Phenotypic Variability
The results of analysis of variance among the 28 groundnut genotypes for late leaf spot (LLS), groundnut rosette disease (GRD), Drought (D) indexes, pod yield and yield penalty evaluated in two locations are presented in Table  2. The results showed highly significant difference (P < 0.001) among genotype for all the traits studied except LLS score. High significant differences (P < 0.01) was recorded by the locations except LLS and GRD. These differences indicated the presence of high genetic variability in the genotypes (Wambi, et al., 2014 andMugisa et al., 2016) and locations for these traits. The highly significant (P < 0.001) difference of genotype-by-location interactions for all traits except LLS and GRD and significant (P < 0.01) variance due to genotype-by-treatment for RWC and pod weight this confirmed existence of wide variation among genotypes, locations and treatments. Similar findings were reported by Mugisa et al. (2016) who studied Determinants of groundnut rosette virus disease occurrence in Uganda and found that the disease severity and groundnut yields were significantly affected by location and genotype and their three way interactions. This variation implying that these genotypes consisted of a source of high yielding and resistance to LLS and GRD and drought tolerance with low yield penalty to be used for improvement of existing low yielding and susceptible groundnut varieties currently in use. The results also suggested that for the purpose of breeding, cultivars could be developed for different disease resistances and drought tolerance.

Means of disease severity and drought tolerance of groundnut genotypes
Means of resistance to late leaf spot (LLS) and groundnut rosette disease (GRD) severity at harvest and drought (D) tolerance for groundnut genotypes grown in the screen house at NaCRRI and MUARIK, 2017 are presented in Table  3. LLS severity at harvest for the studied resistant genotypes ranged between 2.25 for SGV 0005 and SGV 0804 x SGV 03590-C10 and 3.5 for SGV 0802 x Abutalata-C2 indicating that these genotypes were resistant (Subrahmanyam et al., 1995

Means of pod yield of resistant genotypes to late leaf spot, rosette diseases and tolerant to drought
High pod weight (12.8 g plant -1 ) was observed by Serenut 1 single resistance to late leaf spot under normal conditions (Fig 1.A). The high pod weight 15.55 and 12.85 observed by double resistance to LLS+GRD genotypes viz, Serenut.2 and SGV 0005 respectively under 80% water field capacity (Fig 2.A). High pod weight (18.87) was observed by the double resistant genotypes (SGV 01510 x SGV 03590-C12) to LLS+D under control (Fig 2.C). The lowest pod weight (0.95) observed under LLS +GRD+T2 was recorded by SGV 0801 x Abutalata-C1 (Fig 6.B) followed by Serenut.2 (1.43) under LLS+GRD infestation (Fig 2.A). This high yield under control and non waterstressed (T1) and low yield observed under the high stress level of three combinations of stresses indicate the effect of diseases and drought on groundnut yield. Similar finding was reported by Mohammed et al. (2018) who studied sources of resistance to LLS, GRD and yield potential and found low yield performance under severe late leaf spot and groundnut rosette disease. On the other hand, Hamidou et al. (2012) evaluated groundnut under drought stress for yield component and found 72% decreased pod yield due to drought. This suggest that these resistant genotypes with low pod yield under stress spent more energy in resistance than yield, similar conclusion was drawn by Nigam et al. (1991). This yield loss due to resistance against biotic and abiotic stresses is called yield penalty (Mechelke, 2000). 2.9 2 LLS Score: Late leaf spot severity at harvest; GRD Score: groundnut rosette disease severity at harvest; T1: 80% of soil field capacity; T2: 60% of soil field capacity; T3: 40% of soil field capacity; RWC: Relative water content; LMSI: Leaf membrane stability index; Pod wt: Pod weight (g).

Means for yield penalty associated with resistance to late leaf spot, rosette and drought tolerance
The highest yield penalty of pod weight (19.98) among single resistance was recorded by SGV 0007 with GRD (Fig  1.B). It was observed that the GRD alone was responsible for higher yield penalty compared to the combinations, this could be due to GRD being an active viral disease which reduced vegetative growth of the plant, thus decreased the yield potential while, the diseases combinations are contradicted and reduced the severity of each other which did not affect yield as much as in single infestation, such a mechanism was described in Agrios (2005). Reduction in pod dry weight by GRD infested has been widely reported (Wilson, 2014). This meant that as the severity of the disease increased, the yield decreased significantly through the significant negative effects of GRD on both the morphological and reproductive growth of groundnut (Usman, 2013).
In the double resistance to LLS+GRD the high yield penalty (14.1) observed in Serenut.2 (Fig 2.A)  several agronomic traits including disease resistance and yield in inbred and hybrid materials under infected and disease free conditions and reported that both introgressions might confer a yield cost even in the absence of SLB, and introgression of 6A gene was associated with a statistically significant reduction in yield. This confirmed that the yield cost is associated with the resistance phenotype rather than with linkage drag. In a barley experiment, that was heavily infected with an avirulent B graminis hordei had 7% lower grain yield and 4% smaller grains than uninoculated control plants (Smedegaard and Stølen, 1981).
The low yield penalty (2.3) of single resistance was recorded by drought tolerant genotype (SGV AWI.0802) (Fig  1.C) under drought stress indicating the low effect of drought tolerance on yield. The low yield penalty (2.17) of double resistance to GRD+D was observed for SGV ER 10009 at GRD (Fig 3.A), (2.8) under drought (Fig 3.B) and (3.8) under GRD+T2 (Fig 3.C). As predicted yield penalty was high for multiple resistance under the combination of LLS+GRD+D. C2 showed the lowest (8.3) yield penalty under combination (Fig 6.B). Indicating the ability of genotypes to balance the allocating of the nutrition between resistance and yield and confirming the hypothesis that not all resistant genes are costly (Kolster et al., 1987). Similarly, Jorgensen and Jensen (1990) reported that no cost was associated with ten genes of resistances to barley powdery mildew.

Interrelationships among diseases indexes, drought, yield and yield penalty
The results of correlation analysis among the traits studied are presented in Table 4. Relative water content (RWC) showed positive weak correlation with yield penalty (r = 0.18) indicating that the tolerance to drought with the increasing RWC had negative effect on yield which confirmed that yield penalty by drought tolerance on groundnut as genotypes spend a lot of energy in the resistance/tolerance to drought instead of yield. Similarly, Mafakheri et al. (2010) studied the effect of drought stress on chlorophyll content and yield characteristics in three varieties of chickpea with four watering regimes and reported up to 66% yield penalty under drought stress conditions. Yield penalty showed positive weak correlation (r = 0.01) with groundnut rosette disease indicating that the resistance to GRD was costly to the yield.
The highly significant (P < 0.01) negative association of LLS at harvest with yield penalty (r = -0.45) indicated that LLS resistance was not costly to the yield suggesting that "stacking" of LLS resistance within a variety may not increase yield costs. Similar results were obtained by James et al. (2010) who studied costs of disease resistance and found lines containing three resistance to septoria did not exhibit greater yield losses.

Conclusion:-
There was variation in the yield and yield penalty among the 28 groundnut genotypes which could be used in selecting parental lines for improving yield and resistance to the late leaf spot, rosette diseases and drought. There was, however, high yield penalty among single, double and multiple resistance recorded by some cultivars such as SGV 0007, Serenut.2, C12, C1 and SGV ER 10003. This high yield penalty under stress indicate the energy drain involved in defense but there was also low yield penalty among some cultivars indicating the ability to balance the sharing of the resources between resistance and yield. Yield penalty was positively associated with resistance to GRD and drought tolerance and negative associated with LLS resistance. The genotypes with low penalty could be used in multiple disease and drought affected areas.