Effect of Soil Moisture Regimes on Seed Iron and Zinc Concentration of Biofortified Bean Genotypes against Malnutrition in Sud-Kivu Highlands

This study investigated the influence of three soil moisture irrigation regimes on concentration of seed iron and zinc content of four biofortified bean varieties promoted for eradication of malnutrition in Sud-Kivu highlands. A field experiment was conducted in the Hogola marsh highlands during two cultural seasons B2013 and B2014. The experiment design was a RCBD with a split plot arrangement where the main plots were 110 m and split plots 20 m. A strategic application of homogenisation of the experimental site’s soil fertility by chemical fertilizers of the type: CaCO3, KCl and DAP was conducted out. Four biofortified varieties (CODMLB001, RWR2245, HM21-7 and RWK10) constituted main factor, while water regimes respectively [bottom of the slope: R1 = 48% soil moisture, at the middle of the slope R2 = 37% soil moisture and at the top of the slope: R3 = 29% soil moisture according to the gradient of humidity] represented secondary factor and seasonality, tertiary factor. The study showed that the concentrations of iron and zinc were highly correlated with soil moisture regimes. The variety HM21-7 demonstrated better adaptability because it showed a low rate of reduction of iron and zinc concentration under the three soil moisture regimes and was therefore best suited to fight malnutrition in the Sud-Kivu province.


Introduction
Bean (Phaseolus vulgaris) is one of the oldest world cultures preferentially consumed in the human diet due to its gastronomic (Broughton et al., 2002;Beebe, 2010;Silva et al., 2012;Casinga et al., 2016aCasinga et al., , 2016b)), culinary (Dinste, 2012;Garden-Robinson & McNeal, 2013) and socio-cultural (Dinste, 2012) specificities.Although faced with several environmental constraints throughout its vital life cycle, it is cultivated in more than 20 countries in Southern, Central and Eastern Africa owing to its adaptability to several ecological niches and so occupies more than four million hectares (Broughton et al., 2002;Hacisalihoglu et al., 2005;Yasar et al., 2008;Casinga et al., 2016a).Under swamped conditions, beans stomata remain closed for long periods, resulting in reduced respiration, transpiration and photosynthesis (Boru et al., 2003;Casinga et al., 2015a).Plants may be slow to recover when water recedes.Long-term impacts on the crop are often related to disease infection and retarded root development that limits access to available subsoil moisture later in the season (Naeve, 2002, Boru et al., 2003).Researchers noted that flooding during critical reproductive stages affect yield components including pod number and seed size (Casinga et al., 2015a(Casinga et al., , 2016a)).Despite its agro-ecological potential, Hogola enjoys a AW 3 climate type of Köppen's classification and its soil are classified as Ferralsol according to the FAO-UNESCO (Baert, 1995;Beernaert, 1999;Botula et al., 2012) while its texture is clay-silty (Casinga et al., 2015a(Casinga et al., , 2016b(Casinga et al., , 2016c)).Subsistence agriculture remains the main activity in our experimentation area in spite of land degradation correlated to the exponential population increase.

Implementation of Tests
After clearing the land of weeds using a machete, the experimental site (1280 m 2 ) was plowed twice with a hoe to a depth of 30 cm in an interval of 10 days and then manually harrowed with a rake.The land was then subdivided into blocks, plots and sub-plots.Conducted under RCBD design with a split plot arrangement.The blocks were three and parallel on a slope.Their dimensions were 5.5 m × 62 m and separated from each other by a distance of 0.75 m.Three plots of 5.5 m × 20 m were delineated within each block at the bottom of the slope (R3 = 48% soil moisture), at the middle of the slope (R2 = 37% soil moisture) and at the top of the slope (R1 = 29% soil moisture) according to the gradient of humidity, so as to represent three water regimes within each block.Four sub-plots of 1 m × 20 m parallel to the slope were in turn demarcated within each plot.
At the physiological maturity stage, 60 well filled and non-soiled pods were collected on 30 plants located on three central lines of each experimental unit (excluding edging plants) and then shelled and sorted.The healthy seeds were then analyzed to determine the iron and zinc content by atomic absorption after extraction with Triamine Pentacetique acid according to Chauhan et al. (1981) and Okalebo et al. (2002).
Four soil samples were collected using a hole at 25 cm depth in each experiment plot.The latter were mixtured to form a composite sample on which the measurements were made before the sowing date.
Soil moisture was measured by the gravimetric method and in addition soil moisture characteristics θ (φ m ) were determined by the table methods of suction and pressure membrane, respectively for the range of matric potential superior to -0.01 MPa and lower to -0.01 MPa according to the procedure described by Okalebo et al. (2002).Two composite soil samples from each sub-plot taken 25 cm subterranean at the vegetative stage V 3 and harvest stages were analyzed according to Kjeldahl (1883) method' for Nitrogen, while Olsen et al. (1954) method' was used for the estimation of available phosphorus in soils.In addition, potassium was analyzed by Dognin's (1981) method.DAP (18-46-0), KCl with 60% concentration in K and CaCO 3 chemical fertilizers were applied strategically according to the level of the NPK elements in each experimental unit in order to standardize the fertility level of the experimental units.The development of the plots described above made it possible to conduct the experiment according to a factorial model in an experimental split-plot device.
In this study, biofortified bean varieties randomized and sowed in-line in sub-plots at 4 cm depth with two seeds per pooled at 20 × 20 cm were the main factor, while water regimes represented the secondary factor.
The experimentation was conducted in two seasons (February-June 2013 and February-June 2014).These constituted the tertiary factor.

The Observed Parameters
The concentration of iron and zinc content in dry plant biomass at the vegetative stage V3 and in seeds, micronutrient loss content, as well as water regimes were determined.

Statistical Analyses
The experimental results were evaluated by the analyses of variance, correlation and multiple regression whereas the means were separated by the LSD α=0.05 test.
Excel 2013, R 3.3.0and Assistat 9.5.1 software were used as a calculation tools.

Iron and Zinc Content in Dry Biomass and Seeds
Statistical analyzes revealed significant differences both between water regimes and among varieties within different water regimes on micronutrient content (P = 0.02; α = 0.05).Specifically, zinc showed significant difference between the two experimental seasons and a significant difference between different water regimes whereas for iron no significant difference was observed both seasonality and water regimes for all varieties.

Discussion
Plants respond differently to environmental stress by adopting and developing adaptive mechanisms according to Chaves et al. (2003) and Casinga et al. (2015cCasinga et al. ( , 2016b)), confirming the results of the different varieties of biofortified bean obtained in the study site.HM21-7 developed good adaptation to by maintaining high Iron and Zinc concentration under this water excess stress.The major effect of excess of moisture on plants is the alteration of the hormonal balance and dry matter concentration and inhibition of photosynthesis and micronutrient uptake (Nunez-Elisea et al., 1999;Pezeshki, 2001;Dat et al., 2004;Kreuzwieser et al., 2004) confirming the decrease of our four genotypes of the micronutrient content of Iron upstream [CODMLB001 (81 > < 40.5 mg/g); RWR2245 (52 > < 42.5 mg/g); HM21-7 (62 > < 40.7 mg/g) and RWK10 (55 > < 37.9 mg/g)] and of Zinc content downstream [CODMLB001 (34 > < 13 mg/g); RWR2245 (34 > < 16 mg/g); HM21-7 (33 > < 18 mg/g) and RWK10 (35 > < 11 mg/g)] during the two experimental seasons.Besides, it confirms the results of Pfeiffer and McClafferty's (2007) work, as well as that of Silva et al. (2012) attesting that biofortified bean's contents in Iron and Zinc is not only genotypical, but like any quantitative character, it is also influenced by environmental factors (agronomic, climatic and edaphic) and G×E interactions such as differential genotype responses to agronomic, climatic and edaphic factors.

Conclusion
The concentrations of Iron and Zinc contents are highly correlated with water regimes.It is up to the population living in the mountainous and marginal soils of mountainous Sud-Kivu to cultivate the variety HM21-7 under the three water regimes in case of inaccessibility to arable land.As to national and international organizations, they should popularize the variety HM 21-7 according to the agro-ecological zones because it is best suited to curb malnutrition in the province of Sud-Kivu.

Figure
Figure 2. Corre = The top of th