Maize (Zea mays L.) is commonly grown worldwide and used as food (animals and humans) and as transportation fuel  in  developed  countries. In  Tanzania,  maize  is  the main staple as well as cash crop and is produced on approximately 4.0 million ha in almost all of the four official  agro ecological zones by small or limited resource farmers compared to rice, wheat, millet and sorghum and is their main source of income (DTMA, 2014). Maize contributes about 60 and 30% of dietary and utilizable protein, respectively (DMTA, 2014; Lyimo, et al., 2001).

Despite the increase in land acreage, production is still very low averaging 1.37 t ha^-1 grain yield due to limited moisture, while 5 t ha^-1 are required to satisfy the food demand (DTMA, 2014). Due to a fast growing population, it is estimated that maize production must increase by 3.0 to 3.5% annually to satisfy demand, which exceeded the 2.0% growth rate over the past two decades (Lyimo et al., 2001).

Climate change has caused a disruption in rainfall patterns and thus no longer a reliable source of moisture especially at flowering, pollination and embryo development stages to produce a harvestable crop. For example, most of the agro ecological regions (Magehema et al., 2014), that receive rains twice per year are now either receiving it once or only for short periods despite high maize production. Maize requires high amounts of moisture to produce a complete crop especially at tasseling, silking and blister kernel stages (Xu et al., 2014). The southern highlands receive the highest percent distribution in terms of area and amount of maize production of 26%, followed by Lake, Eastern, Northern, Western, Southern and Central zones of 25, 13, 12, 10, 8 and 6%, respectively (DMTA, 2014).

Drought is the most limiting factor for maize production especially if deficit occurs at grain filling and dough stages and is first among the criteria considered by farmers when selecting varieties (Lyimo et al., 2001). In maize production, moisture requirements fluctuate based on growth stages with greatest requirements. There are few drought tolerant maize varieties in the country that were developed through other on-going projects. These varieties have been used in some ecological zones although their genetic diversity is not well known.

Drought stress greatly reduces net photosynthesis rates in plants (Mafakheri et al., 2010), and as the intensity of drought stress increases, the photosynthetic rate decreases even further, eventually reducing yield (Liu et al., 2012). The synthesis of chlorophyll a and b is inhibited as well as a reduction in the content of chlorophyll a and b binding proteins, leading to a decline in the light-harvesting pigment proteins associated with photosystem II (Farooq et al., 2009). Plants under water deficit reduce transpirational water losses by reducing stomatal conductance causing partial or complete closure (Ghannoum, 2008). Stomata closure reduces CO₂ uptake thus reducing photosynthetic rates, transpirational cooling, water and nutrient uptake (Lu et al., 2011).  

Kamara et al. (2003) revealed that water deficit imposed at various developmental stages of maize reduced total biomass accumulation at silking by 37%, at grain filling by 34% and at maturity by 21%. Therefore, moisture stress can cause delay in silk emergence, hence increasing the anthesis-silking interval (ASI; the time in days between pollen shed and silk emergence) (Aylor et al., 2003) or silks may desiccate and become non-receptive to pollen germination. 

Delay in silk emergence is connected to a delay in silk elongation, a process, which takes place through cell division and expansion. Water deficit affects cell expansion, which minimizes silk elongation rates, sometimes to zero, which in turn affects ASI (Monneveux et al., 2006). Longer ASI causes asynchronies in maize flowering that result in incomplete pollination and kernel set, hence yielding losses. The first two weeks after silking are most critical for fertilization and embryo development both of which require high amounts of moisture. Water stress at this stage causes embryo abortion which in turn affects final grain yield (Hayano-Kanashiro et al., 2009; Wei et al., 2009), reduces silk length, inhibits embryo development after pollination and reduces ear size and number of kernels to be fertilized (Farooq et al., 2009).   

There are a set of drought tolerant genes that are known to control physiological and biochemical processes of plants during water stress and breeding practices such as inbreeding or breeding open pollinated varieties (OPVs) reduce genetic diversity among plant materials. Current research shows that most of maize varieties developed in Africa are OPVs and the genetic features of most of them are not known.

The objectives of this study were (1) to compare water use, grain yield and growth responses of ten maize cultivars under well-watered and water stressed conditions and (2) to assess the expression of drought tolerant genes at both vegetative and flowering stage to identify tolerance or susceptibility.

Greenhouse experiments were conducted in a Randomized Complete Block Design with a 10×3 factorial treatment arrangement and three replications at Sokoine University of Agriculture (SUA) in Morogoro, Tanzania. The treatment combinations were moisture at 100 (equivalent to field capacity), 75 and 50%, respectively, of field capacity and ten maize hybrid varieties eight of which (‘CML444’, ‘CML395’, ‘CML539’, ‘WE4107’, ‘WE2112’, ‘WE3102’, ‘WE4112’ and WE 4116) were obtained from the Water Efficiency Maize for Africa (WEMA) project located at the Ilonga Agricultural Research Institute-Morogoro, Tanzania and are classified as being drought tolerant.  The remaining two hybrid maize hybrid varieties, STAHA and  TMV1,  are  regularly grown by local farmers and are classified as being drought susceptible.

Planting and treatment initiation

The oven dry weight method was used to determine percent soil moisture that approximates field capacity, beginning with determining the gravimetric moisture content (Gardner, 1986). A 1000 ml graduated cylinder with a hole drilled in the bottom, to allow air to escape when water was added was filled with a random subsample of air-dried soil, which was leveled to a similar consistency. The surface of the soil was covered with paper towel and 100 ml of water were poured slowly onto the surface to obtain an even distribution through the column. The cylinder was covered to avoid evaporation and allowed to equilibrate stand for 24 h following which there was a sharp line where the water stopped moving in the soil column. Triplicate soil samples were collected from the cylinder about 5 cm above the wetting front and dried at 100°C to a constant weight after which dry weight was recorded and gravimetric moisture fraction was determined using the fresh to dry weight ratio.  This information was used to determine the amount of oven-dry soil per pot, and watering to field capacity (1 L) was done by weighing the pots.

Ten liters pots were filled with 10 kg of air-dried soil into which diammonium phosphate ((NH₄)₂HPO₄; 18-46-0) fertilizer was thoroughly incorporated based on soil test recommendations for the Morogoro region. The drought stress treatments of 100 (field capacity), 75 and 50% of field capacity were initiated at vegetative stage (three weeks after emergence) for thirty days watered again for one week, and then imposed one week before flowering stage to two weeks after. Field capacity moisture (well-watered) was kept at 10 kpa and drought stress kept at 30 kpa using tensiometers.

Phenological measurements taken included plant height, stem diameter, leaf length, width and chlorophyll content. Leaf chlorophyll content was measured for fully expanded leaves using an Apogee CCM 200+ chlorophyll content meter (Apogee Instruments, Logan UT, USA) beginning one week after emergence to harvest, while leaf length and width were assessed on one leaf per plant up to maximum growth (no further increases in either length or width), corresponding to about six to eight weeks after emergence.

Gene expression profiling

Mature disease free maize leaves were collected at the early vegetative and flowering stages of plant development from all ten genotypes cleaned with RNAlater and stored at -80°C. RNA was isolated according to the CTAB method (Hilario and MacKay, 2007). A 0.3 g leaf sample was ground manually using a mortar and pestle in 300 µl of buffer followed by an additional 700 µl and further ground and then centrifuged at 12000 g for 5 min at 4°C. The supernatant was transferred to new Eppendorf tubes, containing 200 µl of acid water to which extracted chloroform was added and shaken vigorously for 15 s and incubated at room temperature for 3 min. The sample was centrifuged at 12000 g for 15 min at 4°C, and the upper aqueous phase transferred to a new tube followed by the addition of 0.5 ml of isopropanol, mixed, and incubated at room temperature for 10 min, and centrifuged at 12000 g for 10 min at 4°C. The supernatant was discarded and 1.5 ml of 75% ethanol was added to the pellet, vortexed briefly and centrifuged at 12000 g for 5 min at 4°C. The supernatant was discarded and the pellets allowed to dry after which they were dissolved in 20 µl of RNase free water, briefly centrifuged and then transferred to a new tube. The quality and quantity of total RNA was checked using Nano drop and native 1% agarose gel in TBE.

Riverse transcriptase polymerase chain reaction (RT-PCR) and real time PCR (qPCR)

Total cDNA was synthesized from total RNA utilizing NotI-OligodT bifunctional primer with Access Quick RT-PCR system (Promega Corp., Fitchburg WI) as per the manufacture’s guidelines. Total cDNA was utilized in downstream quantitative expression profilings in RT-PCR with SYBR-Green. Primers utilized for expression profiling included ProDH, UBQ7, Ubiquitin, P5CS, P5CR and Maize actin (internal reference). The conditions for amplifications were an initial denaturation at 95°C for 2 min, a final denaturation at 95°Cfor 30 s, annealing at 53°C for 30 s for (ProDH) and 50°C for other primers, extension at 72°C for 30 s, a final extension at 72°C for 5 min and a 4°C hold. Of the five primers, only ProDH, UBQ7 and Ubiquitin showed good amplification bands and therefore they were used for further qPCR analysis. Quantitative qPCR was performed using a Thermo scientific maxima SYBR Green/Fluorescein qPCR master mix (2×) as per the manufacturer’s instructions.

Biomass production

The analysis of variance to cultivar, drought stress, and the cultivar×drought stress interactions for plant height, stem diameter, leaf length and width, chlorophyll content, anthesis-silking interval (ASI) and dry seed weight of the ten maize varieties are shown in Table 1. Drought stress had no significant impact on plant height or stem diameter, but enhanced chlorophyll content, ASI and weight of dry kernels, while cultivar significantly influenced all variables.

While the main effect of drought stress on plant height was not significant, certain trends were evident. Plant height and leaf chlorophyll content tended to decrease with increased water stress while ASI and dry seed weight decreased significantly with increased drought stress (Table 2). 

For cultivar main effects, hybrids ‘TMV1’ and ‘STAHA’ responded differently to drought stress as compared to the drought tolerant hybrids (Table 3). ‘STAHA’ had bigger stems than all other cultivars studied and leaf chlorophyll concentration was similar for ‘STAHA’, ‘TMV1’, ‘WE4116’, ‘WE4107’ and ‘WE210’.

Drought tolerant and susceptible maize cultivars were used to assess growth, development and grain yield, drought gene expression level and genetic diversity. Drought had no significant effect on vegetative growth but exerted significant impacts on ASI and grain yield. Drought tolerant genes were expressed at different levels and only a few cultivars expressed all three genes at the vegetative and flowering stages. These genes were mostly down regulated at flowering than at the vegetative stage. Therefore, based on a short ASI, high dry mass of kernels and drought gene expression at both vegetative and flowering stages, cultivars ‘WE2112’, ‘WE3102’ and ‘WE4107’ appeared to be tolerant of the drought stress levels evaluated in this study. Thus, these cultivars which have shown a degree of drought tolerance, could be tested again through on-farm field trials in regions of the country where drought is severe, followed by differential gene expression using many other drought tolerant genes which were not involved in this study, and could form the basis for developing other drought tolerant varieties.

The authors have not declared any conflict of interests.

Contribution of the Washington Carver Agricultural Experiment Station, Tuskegee University is acknowledged. This research was supported by funds from Innovative Agricultural Research Initiative, USAID.