Induced Expression of Xerophyta Viscosa Xvsap1 Gene Greatly Impacts Tolerance to Drought Stress in Transgenic Sweetpotato

bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1 Induced expression of Xerophyta viscosa XvSap1 gene greatly impacts tolerance to 2 drought stress in transgenic sweetpotato 3

4 Wilton Mbinda1*, Christina Dixelius2, Richard Oduor3

8 1Department of Biochemistry and Biotechnology, Pwani University, Kilifi, Kenya 9 10 2Department of Plant Biology, Uppsala BioCenter, Linnéan Center for Plant Biology, 11 Swedish University of Agricultural Sciences, Uppsala, Sweden 12 13 3Department of Biochemistry and Biotechnology, Kenyatta University, Nairobi, Kenya 14 15 *Corresponding author: Wilton Mbinda; Email: [email protected]; Phone: 16 254722723950 17 18

28 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

29 Abstract 30 Key message Drought stress in sweetpotato could be overcome by introducing XvSap1 31 gene from Xerophyta viscosa. 32 33 Abstract. Drought stress often leads to reduced yields and is perilous delimiter for expanded 34 cultivation and increased productivity of sweetpotato. Cell wall stabilization proteins have 35 been identified to play a pivotal role in mechanical stabilization during desiccation stress 36 mitigation. They are involved in myriad cellular processes that modify the cell wall properties 37 to tolerate the mechanical stress during dehydration in plants. This provides a possible 38 approach to engineer crops for enhanced stable yields under adverse climatic conditions. In 39 this study, we introduced the XvSap1 gene isolated from Xerophyta viscosa, a resurrection 40 plant into sweetpotato by Agrobacterium-mediated transformation. Detection of the transgene 41 by PCR coupled with Southern blot revealed the integration of XvSap1 in the three 42 independent events. Sweetpotato plants expressing the XvSap1 gene exhibited superior 43 growth performance such as shoot length, number of leaves and yield than the wild type 44 plants under drought stress. Quantitative real time-PCR results confirmed higher expression 45 of the XvSap1 gene in XSP1 transgenic plants imposed with drought stress. In addition, the 46 transgenic plants had increased levels of chlorophyll, free proline and relative water content 47 but malonaldehyde content was decreased under drought stress compared to wild type plants. 48 Conjointly, our findings show that XvSap1 can enhance drought resilience without causing 49 deleterious phenotypic and yield changes, thus providing a promising candidate target for 50 improving the drought tolerance of sweetpotato cultivars through genetic engineering. The 51 transgenic drought tolerant sweetpotato line provides a valuable resource as drought tolerant 52 crop on arid lands of the world. 53 54 Keywords Drought Stress, Gene expression, Sweetpotato, Xerophyta viscosa, XvSap1 gene 55 56 Introduction 57 Plants survival under adverse environmental conditions depends on integration of stress adaptive physiological 58 and metabolic changes into their endogenous developmental systems. The world’s seventh important crop, 59 sweetpotato [Ipomoea batatas (L.) Lam.] plays a significant role in food security and nutritional requirements, 60 for millions of people in Asia and Africa (Bovell-Benjamin 2007). This crop also has an enormous potential to 61 be commercially exploited as an industrial raw material (Kasran et al. 2015). Sweetpotato is widely cultivated 62 on marginal lands due to its relatively high tolerance to abiotic stress. The orange-fleshed sweetpotato cultivars 63 contain portentous β-carotene quantities which is a viable solution to combat vitamin A deficiency in sub- bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

64 Saharan Africa (Laurie et al. 2018). The crop’s effective clonal propagation, its cultivation simplicity and high 65 biomass generation make sweetpotato conceivably appropriate for molecular farming of valuable and novel 66 products (Rukundo et al. 2017). Like other crops, sweetpotato is adversely affected by drought stress, which 67 seriously hampers the crop productivity and negatively influence the expansion of sweetpotato cultivation (Jin et 68 al. 2017). Additionally, as source of bio-energy, sweetpotato will mainly be grown on marginal land in the 69 future. Improving the drought tolerance is essential for the crop’s adaption to the environmental stress and 70 increasing productivity. Sweetpotato trait improvement through conventional breeding is constrained by its 71 laborious and time-consuming genetics. These limitations principally arise owing to the crop’s high 72 heterozygosity, high levels of male infertility, complicated polyploidy and production of few seeds because of 73 its self-incompatibility, resulting to sturdy segregation of hybrid progenies and the loss of numerous valuable 74 traits (Martin 1965, Mwanga et al. 2017). The fast advancement in plant biotechnology has unlocked new 75 potentials for increasing tolerance to abiotic and biotic stresses to sweetpotato as well as improving its 76 nutritional quality by identifying key genes and introducing them through genetic engineering Additionally, 77 genetic engineering technology has the capability of introgressing genes from incompatible plant species or 78 other organisms, an imperative phenomenon for crop improvement. In sweetpotato, little progress has been 79 made in the generation and evaluation of transgenic events against drought tolerance. 80 81 Plant drought stress involve intricate regulatory processes that control water flux and cellular osmotic 82 modification through the biosynthesis of osmoprotectants (Golldack et al. 2014). ‘Resurrection plants’ survive 83 dehydration (losing over 90% of their water content) of their vegetative tissues to air dry state for extended 84 periods and recover complete metabolic state after rehydration (Challabathula et al. 2016). These resurrection 85 plants could serve as ideal models for ultimate design of important crops with enhanced stress tolerance. A 86 representative of this special category of plants is Xerophyta viscosa. This species is a southern African native 87 monocot species that has been extensively investigated to understand the genetic mechanisms of desiccation 88 tolerance and may serve as potential plethora of superior genes that could introgressed into important crops 89 (Costa et al. 2017). Of particular interest in this study is XvSap1, a stress-regulated protein, that was isolated 90 from a cDNA library constructed from dehydrated X. viscosa leaves. A gene that has been associated with 91 desiccation stress tolerance (Garwe et al. 2003). The XvSap1 protein has 49% identity to a cold acclimation 92 protein WCOR413, WCOR413, from Triticum aestivum (Mohammadi et al. 2007). Earlier work on utilizing 93 genes from X. viscosa have resulted in promising outcomes on drought stress improvement in important crops 94 such as maize (Seth et al. 2016) and tobacco (Kumar et al. 2013) which encouraged us to exploit XvSap1 for 95 genetic engineering of sweetpotato. 96 97 The XvSap1 gene is a highly hydrophobic protein. It encodes two membrane lipoprotein -lipid domains which 98 are activated in leaves of X. viscosa during dehydration stress (Garwe et al. 2003). Although the function of 99 XvSap1 in X. viscosa is yet to be fully determined, there are three proposed hypotheses. First, XvSap1 may be 100 involved in stabilizing the plasma membrane; second, it may be involved in maintaining ion homeostasis; and 101 last, XvSap1 may be a G-protein-coupled receptors (GPCR) associated with signal transduction in osmotic stress 102 (Iyer et al. 2008). G-protein-coupled receptors form a large family of proteins whose principle function is to 103 transduce extracellular stimuli into intracellular signals (Kroeze et al. 2003). Arabidopsis thaliana and Nicotiana bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

104 tabacum transgenic plants overexpressing XvSap1 exhibited high tolerance to osmotic, salt, heat and 105 dehydration stress (Garwe et al. 2006). In the present study, we generated and assessed transgenic sweetpotato 106 plants expressing XvSapI under the control of its stress inducible endogenous promoter. The results demonstrate 107 that transgenic sweetpotato plants expressing XvSapI had significantly improved tolerance to drought stress 108 compared with the wild type plants. 109 110 Materials and methods

111 Vector construction for plant transformation 112 The full-length cDNA of XvSap1 (accession no. CB330588.1), and its stress-inducible promoter was provided 113 by Prof. Jennifer A. Thomson (University of Cape Town, South Africa), was cloned into the binary vector 114 pNOV2819 at the BamHI and HindIII restriction sites. The binary vector pNOV2819 contains the 115 phosphomannose isomerase (pmi) gene as selectable marker that confers resistance to mannose. Since 116 sweetpotato is not sensitive to mannose selection (Mbinda et al. 2015), the pmi was substituted with the 117 neomycin phosphotransferase (nptII) gene flanked by the nos promoter and nos terminator at HindIII and KpnI 118 restriction sites. The binary vector harboring the transgene (pSAP1-XvSap1) and the selectable marker nptII, 119 conferring kanamycin resistance, was mobilized into the disarmed Agrobacterium tumefaciens strain EHA105 120 via the freeze–thaw method (Weigel and Glazebrook 2006) and used for sweetpotato genetic transformation. 121 122 Sweetpotato genetic transformation 123 Sweetpotato (Ipomoea batatas [L.] Lam., cv. Jewel (donated by International Potato Centre, Nairobi, Kenya) 124 plants were used in this study. The plants were maintained in vitro by sub-culturing every four weeks at 27±1 ºC 125 under 16 h/8 h light/dark conditions (Mbinda et al. 2016). Explants were prepared from stem segments from 3 to 126 4-week-old in vitro plants by transversely cutting them into 6–10 mm segments before diving the segments 127 along the axis. Genetic transformation, somatic embryogenesis and regeneration of putative transgenic plants 128 were performed (Mbinda et al. 2018). Cefotaxime (250 mg/l) and kanamycin (50 mg/l) were supplemented to 129 the medium to induce the development of transgenic cell clusters. The regenerated sweetpotato plantlets were 130 transferred to soil and grown in a greenhouse at 27±1 °C. 131 132 DNA extraction, PCR and southern blot analysis 133 Total genomic DNA was extracted from leaves of both wild type and putative transgenic sweetpotato lines using 134 the CTAB method (Borges et al. 2009). The extracted genomic DNA was amplified using gene specific primers 135 (Supp. Table S1) to evaluate to the presence or absence of XvSap1. Southern blotting was performed to evaluate 136 the copy number and stable integration of transgene in the transgenic sweetpotato lines using standard 137 techniques (Sambrook et al. 1989). Genomic DNA (20 µg) from PCR positive and wild type plant lines was 138 digested with EcoRI. Agarose (0.8 %) gel electrophoresis was used to separate the digested DNA, before 139 blotting onto a Hybond-N+ membrane, followed by UV cross-linking. A probe, amplified from the XvSap1 140 plasmid (Supp. Table S2), was labeled with dCTP α-32P using Amersham Rediprime II DNA Labelling kit (GE 141 Healthcare, Buckinghamshire, England) and used for hybridization at 42 °C. Blots were washed for 10 min at 42 142 °C and more stringently twice for 30 min at 55 °C. A third wash was done if the bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

143 residual radioactivity on the membrane was considered high. Finally, the membrane was monitored using a 144 Kodak Flexible Phosphor Imager (Eastman Kodak Co. New York, US). 145 146 Drought tolerance analysis 147 Wild type and transgenic sweetpotato stem cutting (12 cm) were planted into pots containing mixed soil in 148 greenhouse under 750-1000 µmol Em-2S-1 photosynthetic photon flux density with 28±2 °C, 16/8 day 149 photoperiod and 80±5 % relative humidity for 4 weeks. Thereafter, and in a completely randomized design, two 150 groups of plant treatments: normal conditions (61.7 % soil water content) and simulated water deficit stress (0.3 151 % soil water content) were set for 12 days. Morphometric parameters (shoot length and number of leaves), 152 biochemical assays (chlorophyll, proline, malondialdehyde and relative water contents) and the XvSap1 153 expression analysis of the different materials were evaluated. Three biological replications were prepared for 154 each treatment. Further, transgenic and wild type sweetpotato plants were planted in pots and regularly irrigated 155 for 30 days. Thereafter, the plants were exposed to drought stress for 90 days. Yield attributes in terms of fresh 156 weight and dry matter were measured for comparative analysis. Control plants (wild type and transgenic plant 157 lines) were irrigated throughout the experiment. 158 159 Assessment of soil water content 160 Soil samples collected at 0, 3, 6, 9 and 12 days after withholding water were dried at 105 °C. The soil water 161 content (SWC), as a measure of 'drought' severity was calculated using the weight fraction described by Coombs 162 et al. (1987). SWC (%) = [(FW − DW)/DW] × 100; where FW was the fresh weight for portion soil from the 163 internal area of sampled pots and DW was the dry weight for the soil portion after drying at 105 ºC for 4 days. 164 165 qRT-PCR analysis 166 Leaves from water stressed and control plants were collected at 0, 3, 6, 9 and 12 days after simulated drought 167 stress. Total RNA was isolated using Spectrum Plant Total RNA Kit (Sigma-Aldrich, St. Louis, US). First 168 strand cDNA was synthesized using gene specific primers (Supp. Table S3). Quantitative real-time PCR (qRT- 169 PCR) was performed with Bio-Rad iQ5 l System Software 1.0. The sweetpotato ubiquitin gene was used as the 170 internal control (Park et al. 2012). Relative expression values were analyzed with the comparative 2-ΔΔCT 171 formulae as described by Livak and Schmittgen (2001). 172 173 Measurement of photosynthetic pigment 174 Chlorophyll contents, under normal and drought conditions, were measured in attached leaves of both wild type 175 and transgenic plant lines with, SPAD chlorophyll meter (Minolta Co., Osaka, Japan), a non-destructive portable 176 device. The youngest fully expanded leaf flag from the top of each plant were used. Measurement started 9 days 177 prior to commencement of water deficit stress and continued for 12 days. The mean of the five values was taken 178 per plant for each of three replications. 179 180 Estimation of proline content 181 Colorimetric assay was used to analyse free proline content. Fresh leaf tissue (100 mg was homogenized in 10 182 ml of 3% (w/v) sulphosalicylic acid. The homogenate was filtered through number 1 Whatman filter paper. A bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

183 0.1 ml aliquot of the filtrate was mixed with 0.5 ml reaction solution [acidic ninhydrin [40% (w/v) acidic 184 ninhydrin (8.8 µM ninhydrin, 10.5 M glacial acetic acid, 2.4 M orthophosphoric acid), 40% (v/v) glacial acetic 185 acid and 20% (v/v) of 3%(v/v) sulphosalicylic acid]. The samples were incubated for 60 min at 100 ºC and the 186 reaction terminated by incubating the samples on ice for 5 minutes. Free proline from the samples was extracted 187 by adding 1 ml of toluene and vortexing for 20 s. The absorbance at 520 nm was measured with UVmini-1240 188 UV-Vis Spectrophotometer (Shimadzu, Kyoto, Japan) using toluene as a reference. Proline content [µmol/g 189 fresh weight (F. WT)] in leaf tissues was computed from a standard curve using 0-100 µg L-proline according to 190 formula by Bates et al. (1973). 191 192 Analysis of lipid peroxidation 193 Lipid peroxidation was measured in terms of malondialdehyde content, a product of lipid peroxidation, is an 194 indicator of oxidative damage, caused by stress such as drought and salinity in plants (Del Rio et al. 2005). 100 195 mg of leaf tissue was homogenized by adding 0.5 ml of 0.1 % (w/v) trichloroacetic acid. The homogenate was 196 centrifuged at full speed for 10 min at 4 °C. The collected supernatant was mixed with 0.5 ml of (1.5 ml 0.5% 197 thiobarbituric acid diluted in 20 % trichloroacetic acid). The mixture was incubated 95 °C in water bath at for 25 198 min before ending the reaction by incubating on ice for 10 min. The absorbance was measured at 532 and 600 199 nm using spectrophotometer (UVmini-1240 UV-Vis) with 1% thiobarbituric acid in 20% trichloroacetic acid as 200 control. The concentration of malondialdehyde (µmol/g FW) calculated as a measure of lipid peroxidation, was 201 determined according to Health and Packer (1968). 202 203 Determination of relative water content 204 Water loss in plants by transpiration was measured form a detached-leaf assay of the excised third leaf from 205 wild type and transgenic sweetpotato plants. Fresh weight of the leaves was recorded immediately after excision 206 and further incubated in a Petri plate at 28 °C in deionized water. The weight of these leaves was recorded after 207 1, 2, 3, 5 and 17 hrs. The rate of water loss was determined according to the formula described by Turner 208 (1981). 209 210 Statistical analysis 211 Analysis in all the experiments was carried out in three replicates. The data collected was analyzed using one- 212 way analysis of variance and post hoc Fisher’s least significant difference test to determine significant 213 differences between the means of each treatment. A value of at confidence level of 95% (p ≤ 0.05) was 214 considered to be statistically significant. Minitab statistical computer software was used for data analysis. All 215 graphs were drawn using SigmaPlot software. 216 217 Results 218 Sweetpotato transformation 219 A total of 167 sweetpotato stem explants were co-cultivated with Agrobacterium strain EHA101 carrying 220 pNOV2819-XvSap1 binary vector (Supp. Fig. S1). The transformed stem segments gave rise to 133 kanamycin 221 resistant calli. Three putative transgenic sweetpotato plants were regenerated from 117 surviving resistant calli 222 (Fig 1A). The regenerated sweetpotato plantlets (Fig. 1B) were hardened for propagation in a biosafety bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

223 greenhouse. Under normal conditions (25-27 °C and 61.7 % SWC), all transgenic sweetpotato lines were 224 phenotypically and morphologically indistinguishable from the wild type lines (Fig. 1C and Fig. 1D). 225 226 Molecular characterization of transgenic sweetpotato plants 227 PCR analysis showed the presence of the 464 bp XvSap1 fragment in the transformed plants but was absent in 228 the non-transformed wild-type plants (Supp. Fig. 2). The results of Southern hybridization assay clearly showed 229 all transgenic lines (XSP1-XSP3) contained a single copy insertion (Supp. Fig. 3). Further, results of reverse 230 transcription-PCR (RT-PCR) analysis of the three transgenic independent events showed XvSap1 mRNA 231 transcripts, which was absent in the nontransformed control plants (Supp. Fig. 4). 232 233 Phenotypic Characteristics 234 Under normal conditions, 4 weeks old transgenic and non-transformed controls sweetpotato plant lines did not 235 have visible phenotypic differences (Fig. 1A and B). Our results therefore indicate that the introgression of 236 XvSap1 into sweetpotato did not lead to any major observable effects in plant architecture and growth 237 properties. Upon exposure of the plants to severe water deficit stress (0.3 % SWC), shoot length of wild type 238 control plants reduced by 55.2 % whereas that transgenic sweetpotato plant line was slightly higher (Fig. 2A). 239 Similarly, after drought stress treatment, the average number of leaves in transgenic sweetpotato plants slightly 240 declined by 15.7 % as compared to the wild type plants which had a drastic decline of 44.3% from an average of 241 8.7 leaves to 4.7 leaves (Fig. 2B). After 6 days of water stress (27.8 % soil water content), severe wilting was 242 evidently observed in the leaves of the wild type plants (Fig. 1E) but of leaves of transgenic plants showed only 243 minor damage (Fig. 1F). Severe dehydration symptoms were observed in the transgenic plants following 12 244 days of water deficit stress (0.3 % soil water content). The severe symptoms included wilted leaves, drying up of 245 leaf edge and tip, and brown leaves. In both the wild type and transgenic plant lines, older leaves evidently 246 displayed leaf wilting and chlorosis when the plants were exposed to severe drought stress (0.3 % soil water 247 content). Overall, the wild type plants displayed much more severe symptoms compared to the transgenic plants. 248 249 Under normal conditions, there was no phenotypic difference in the tuber formation between the wild type (Fig. 250 1G) and transgenic plants (Fig. 1H). Following 90 days of drought stress, the yield of wild type plants (Fig. 1I) 251 drastically reduced compared to transgenic plants that produced bigger, and higher numbers of tubers (Fig. 1J). 252 When plants were regularly watered, the fresh weight yield and dry weight yield of both wild type and 253 transgenic plants did not differ (Fig. 3A). Upon subjecting the plants to drought stress for 90 days, both the fresh 254 weight and dry weight of transgenic and wild type plants were significantly changed. The fresh and dry weight 255 yield per plant of transgenic plants under drought conditions was 341.5 g and 215.2 g respectively whereas wild 256 type plants were 291.8 and 153.6 respectively (Fig. 2B). Further, we investigated the transcript expression levels 257 in leaves of wild type and transgenic XSP-1 line under normal and drought stress conditions by qRT-PCR. The 258 expression levels of XvSap1 in transgenic plants under normal condition was low but was induced under stress 259 to 50-fold after 6 days of withholding water. Thereafter, expression started to decline, which is expected when 260 using a stress-induced promoter (Fig. 4). 261 262 Biochemical characteristics bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

263 At the start of the experiment, chlorophyll content of both wild type and transgenic sweetpotato lines were the 264 same and increased as the plants grew (Fig. 5A). Chlorophyll concentration in leaves of both wild type and 265 transgenic plant lines progressively dropped with increasing drought stress intensity resulting in lower SPAD 266 values compared to control treatment; plants grown under normal conditions throughout the experiment. In 267 contrast, water deficit stress had mild effect on chlorophyll content in leaves of transgenic line compared to 268 control, as shown in Fig. 5A. 269 270 To evaluate the changes brought about by drought stress, we measured the accumulation of free proline in the 271 shoot leaves of wildtype and the transgenic XSP1 line. Because level of free proline is known to be an indicator 272 of drought stress resistance capacity in plants (Zandalinas et al., 2018). No significant difference in free proline 273 concentration was observed in both wild type and transgenic leaves under normal conditions (Fig. 6B). After 12 274 days of water deficit stress experiments, the proline content in wild type plants increased 3.8-fold to 16.4 µg/g 275 FW (Fig. 5B). In contrast, the XSP-1 line exhibited higher accumulation of proline x-fold??? compared to that 276 of the wild type plants (Fig. 5B). 277 278 Further validation for the effect of XvSap1 expression on lipid peroxidation was carried out by estimation of 279 malondialdehyde (MDA) in the leaves. Under normal conditions, the levels found in the wild type line and the 280 XSP1 line was similar (Fig. 5C). In contrast, imposition of water deficit stress to the plants resulted in about 3.4- 281 fold increase in the malondialdehyde content in leaves of wild type plant line and much less malondialdehyde 282 content in leaves of the transgenic sweetpotato plants was significantly lower than in the wild type plants (Fig. 283 5C). 284 285 Relative Water Content 286 Under normal conditions, there were no significant differences in the leaf relative water content between control 287 and transgenic sweetpotato plant XSP1 lines, and the relative water content of both lines was about 82 % (Fig. 288 6D). After subjecting the plants to drought stress for 6 days, the relative water content of the control wildtype 289 leaves rapidly reduced by a 37.0 % to 51.8 %, while the relative water content of the transgenic line declined 290 very gradually from 82. to 68.6 %, a 16.4 % decline (Fig. 5D). After 12 days of water deficit stress, the leaf 291 relative water content of the transgenic plant had decreased by just 38.4 % as compared to 56.7 % in the wild 292 type plants (Fig. 6D). These results demonstrate that the wildtype sweetpotato line was more drought sensitive 293 than the transgenic line. 294 295 Discussion 296 As drought frequency and severity are projected to increase in the future due to climate change perturbations, 297 the ensuing deficit in water, especially in arid and semi-arid regions, will greatly reduce agronomic productivity. 298 Increasing tolerance to these stresses through the development of crop varieties with improved tolerance is the 299 only probable solution to extremely reduce the impacts of drought stress on important crops such as sweet 300 potato crops. Through several studies, various stress tolerance genes that have profound effects in enhancing 301 drought tolerance in plants have been identified and among them are the members of G-protein-coupled bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

302 receptors (GPCRs) genes playing a crucial role in regulating plant water status under normal and adverse 303 circumstances (Bai et al. 2018; Lu et al. 2018; Wong et al. 2018). 304 305 Numerous strategies have been applied to engineer plants with enhanced stress tolerance with limited success. 306 One major problem is due to the undesirable or unfavorable phenotypic changes such as growth retardation and 307 decrease in yield. Compared with wild type plants, we did not observe any phenotypic changes in transgenic 308 sweetpotato plants. Similar results were also observed in our previous study when sweetpotato was genetically 309 engineered with an aldose reductase, XvAld1, isolated from X. viscosa (Mbinda et al. 2018). Other studies using 310 genes from X. viscosa and transferred into other crops; maize (Seth et al. 2016) and tobacco (Kumar et al. 311 20113) have also been successful, reaffirming the proposition that X. viscosa has prodigious potential for mining 312 of valuable genes for biotechnological utilization with no nocuous variations to the accruing transgenic plants. 313 314 Under water deficit conditions, transgenic sweetpotato plants were identified as drought tolerant based on the 315 reduction proportion of vine length and number of leaves compared to wild type. Upon re-watering after 316 simulated drought treatment, transgenic plants resulted in rapid recovery with higher shoot height. Results from 317 previous work corroborate our finding that dehydration stress in plant causes hydraulic restriction, with the 318 ensuing high tension in the xylem water column and the stomatal closure of (Powell et al. 2017). 319 320 In plants, suitable amounts of reactive oxygen species (ROS) aggregation are vital for tolerance to abiotic stress 321 (Das and Roychoudhury 2014). Nevertheless, excess ROS generation causes degradation of chlorophyll (Dietz 322 et al. 2016). The degradation of photosynthetic pigments, which vital to plants principally for light harvesting 323 during photosynthesis process and production of reducing power, is therefore one of the most sensitive indices 324 and diagnostic tool for evaluation of plant responses to water deﬁcit stress tolerance identification, genotypic 325 variation, altitudinal variation of many crops such as potato (Rolando et al. 2015) and cassava (Orek et 326 al.,2016). Plants can overcome drought stress by increasing the accumulation of chlorophyll which protects 327 them by scavenging the excessive energy by thermal dissipation (Reddy et al. 2004). Waning of leaf chlorophyll 328 concentration in response to drought stress is a common phenomenon in plants, occasioned by disordering 329 chlorophyll synthesis and which results to changes in thylakoid membrane structure and plant chlorosis (Kalaji 330 et al. 2016). When plants are subjected to environmental stresses, leaf chloroplasts are injured which leads to 331 disrupted photosynthesis. Expression of XvSap1 enhanced drought tolerance by stabilizing antioxidant status in 332 leaf tissues during drought stress. We hypothesize that XvSap1 enhanced drought tolerance by increasing 333 stabilization of cell membranes of the desiccation stressed transgenic plants,The enhanced chlorophyll content 334 exhibited by the transgenic line could be a result of stabilized antioxidant status in leaf tissues drought stress. 335 Accumulation of proline and polyamines is a common response to various abiotic stresses and results in our 336 study corroborate with responses of other crops to water deﬁcit, such as kiwi fruit (Chartzoulakis et al. 1993), 337 and cotton (Massacci et al. 2008), whose leaf chlorophyll content was higher in plants grown in normal 338 conditions than under drought conditions. 339 340 Desiccation stress on plants induces osmotic and oxidative stress in plants, leading to cellular adaptive responses 341 such as accumulation of compatible solutes (Chen and Murata 2011). Free proline is the most important bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

342 osmolyte and signaling molecule which accumulates generally in the cytoplasm without prejudicing normal 343 cellular physiological functions and also contributes in protection of membranes, proteins and enzymes against 344 various stresses (Golldack et al. 2014). Proline accumulation has therefore a positive connection with their 345 tolerance to various environmental stresses. The finger millet varieties with higher volume change when 346 imperilled to drought stress have a higher proline content, but it is not clear if it is an adaptive response or a 347 biochemical change as a result of the damage triggered by desiccation. Proline has been reported to contribute 348 approximately 10-15% of the osmotic adjustments in water deﬁcit stressed castor plants (Babita et al,.2010), pea 349 cultivars (Sánchez et al. 1998) and transgenic sweetpotato (Mbinda et al. 2018) grown under drought stress. The 350 biochemical relation between XvSap1 gene action and the high proline content in the cells seems to be a 351 sensitive balance 352 353 Enhanced reactive oxygen species (ROS) activity eventually leads to lipid peroxidation of the cell membranes 354 and subsequent membrane leakage, a common phenomenon during various environmental stresses and causes 355 severe damage to plant cells, leading programmed cell death. Malondialdehyde is a product of lipid peroxidation 356 and its quantification is generally employed as an index of oxidative damages in plant tissues. The observed 357 reduced lipid peroxidation in transgenic sweetpotato plants was probably due to lower reactive oxygen species 358 generation under water deficit stress. These results suggest that inducible expression of XvSap1 gene modulates 359 oxidative stress responses. Moreover, plant growth and development are constrained by sstress factors such as 360 scavenging and quenching of free radicals occasioned by ROS. These mechanisms are extensively regulated and 361 plant tolerance to drought stress is related to their antioxidant scavenging ability. Enhanced levels of the 362 antioxidant constituents impedes the stress damage before it develops to lethal status (Demidchik 2015; 363 Mickelbart et al. 2015). Our results on growth and development of transgenic and control sweetpotato plant 364 lines corroborate to what. 365 366 Ability to retain water during dehydration is an important strategy for plant tolerance to stress caused by 367 drought. Consequently, measurement of relative water content is a rapid approach to estimate plant water status 368 and provides projection levels of the cellular hydration levels after stress treatments (Sánchez-Rodríguez et al. 369 2010). Our study shows an enhanced water retention capacity in the transgenic sweetpotato plants. Decreasing 370 transpiration rate is significant for plant survival under drought conditions. 371 372 Our results provided clear evidence that the presence of X. viscosa gene in transgenic sweetpotato plants 373 displayed favorable traits associated with drought tolerance. From these results, we therefore believe that 374 XvSap1 gene may have great potential in sweetpotato crop improvements for drought stress. 375 376 Acknowledgments The authors thank Prof. Jennifer A. Thomson, University of Cape Town, South Africa for 377 XvSap1 cDNA and XvPSap1 promoter and Prof Christina Dixelius for nptII. This work was funded by DST, 378 India (Grant No. SR/SO/BB-37/2008). The authors acknowledge the Plant Transformation Laboratory, 379 Department of Biochemistry and Biotechnology, Kenyatta University, Nairobi, Kenya, and the Department of 380 Plant Biology, Swedish University of Agricultural Sciences, Uppsala, Sweden, for providing the laboratory bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

381 facilities. This research work was funded by the National Council for Science and Technology, Kenya (Grant 382 No. NCST/5/003/3rdSTICALL/109). 383 384 Author contributions RO, WM conceived and designed experiments. WM performed all experiments. RO and 385 CD supervised the execution of the research, WM analyzed data and WM, CD and RO wrote the manuscript. 386 387 Conflict of interest Authors declare that they have no conflict of interest. 388 389 References 390 Babita M, Maheswari M, Rao LM, Shanker AK Rao DG (2010) Osmotic adjustment, drought tolerance and 391 yield in castor (Ricinus communis L.) hybrids. Environ Exper 69:243-249 392 Bai L., Liu, Y, Mu Y, Anwar A, He, C, Yan Y, Li Y, Yu X (2018). Heterotrimeric G-protein γ subunit CsGG3. 393 2 positively regulates the expression of CBF genes and chilling tolerance in cucumber. Frontiers in 394 Plant Science, 9: 488 395 Bates LS, Waldren RP, Teare ID (1973) Rapid determination of free proline for water-stress studies. Plant Soil 396 39: 205-207 397 Bovell BAC (2007) Sweetpotato: a review of its past, present, and future role in human nutrition. Adv Food 398 Nutr Res 52:1-59 399 Challabathula D, Puthur JT, Bartels D (2016) Surviving metabolic arrest: photosynthesis during desiccation and 400 rehydration in resurrection plants. Ann NY Acad Sci 1365:89-99 401 Chartzoulakis K, Noitsakis B, Therios I (1993) Photosynthesis, plant growth and dry matter distribution in 402 kiwifruit as influenced by water deficits. Irrigation Science 14:1-5 403 Chen TH, Murata N (2011) Glycinebetaine protects plants against abiotic stress: mechanisms and 404 biotechnological applications. Plant Cell Environ 34:1-20 405 Costa MCD, Artur MA, Maia J, Jonkheer E, Derks MF, Nijveen H, Williams B, Mundree SG, Jiménez-Gómez 406 JM, Hesselink T, Schijlen EG (2017) A footprint of desiccation tolerance in the genome of Xerophyta 407 viscosa. Nature Plants 3:1-10 408 Das K, Roychoudhury A (2014) Reactive oxygen species (ROS) and response of antioxidants as ROS- 409 scavengers during environmental stress in plants. Front Environ Sci 53: 1-13 410 Del Rio D, Stewart AJ, Pellegrini N (2005) A review of recent studies on malondialdehyde as toxic molecule 411 and biological marker of oxidative stress. Nutr Metab Cardiovasc Dis 15:316-328 412 Demidchik, V (2015) Mechanisms of oxidative stress in plants: from classical chemistry to cell biology. Environ 413 Exper Bot 109:212-228 414 Garwe D, Thomson JA, Mundree SG (2003) Molecular characterization of XVSAP1, a stressresponsive gene 415 from the resurrection plant Xerophyta viscosa Baker1. J Exp Bot 54:191-201 416 Garwe D, Thomson JA, Mundree SG (2006) XVSAP1 from Xerophyta viscosa improves osmotic, 417 salinityand hightemperaturestress tolerance in Arabidopsis. Biotechnol J 1:1137-1146 418 Golldack D, Li C, Mohan H, Probst N (2014) Tolerance to drought and salt stress in plants: unraveling the 419 signaling networks. Front Plant Sci 5:151 420 Health RL, Packer L (1968) Photoperoxidation in isolated chloroplasts: I. Kinetics and stoichiometry of fatty 421 acid peroxidation. Arch Biochem Biophys 125:189-198 422 Iyer R, Mundree SG, Rafudeen MS, Thomson JA (2008) XvSap1, a desiccation tolerance associated gene with 423 potential for crop improvement. Plant desiccation tolerance, 283.In: Jenks M, Wood AJ (Eds.) Plant 424 Desiccation Tolerance, Blackwell Publishing, Iowa, USA pp. 283-296 425 Jin R, Kim BH, Ji CY, Kim HS, Li HM, Kwak SS (2017) Overexpressing IbCBF3 increases low temperature 426 and drought stress tolerance in transgenic sweetpotato. Plant Physiol Biochem 118:45-54 427 Kalaji HM, Jajoo A, Oukarroum A, Brestic M, Zivcak M, Samborska IA, Cetner MD, Łukasik I, Goltsev V, 428 Ladle RJ (2016). Chlorophyll a fluorescence as a tool to monitor physiological status of plants under 429 abiotic stress conditions. Acta Physiol Plant 38: 102 430 Kasran M, Hashim H, Sarmin S, Nezuri DM (2015) Extraction of starch and enzymatic production of high 431 amylose starch from sweetpotato (Ipomea batatas) var. Telong. J Trop Agric Food Sci 40:203-210 432 Kroeze WK, Sheffler DJ, Roth BL (2003) G-protein-coupled receptors at a glance. J Cell Sci 116:4867-4869 433 Kumar D, Singh P, Yusuf MA, Upadhyaya CP, Roy SD, Hohn T, Sarin NB (2013) The Xerophyta viscosa 434 aldose reductase (ALDRXV4) confers enhanced drought and salinity tolerance to transgenic tobacco 435 plants by scavenging methylglyoxal and reducing the membrane damage. Mol Biotechnol 54:292–303 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

436 Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and 437 the 2− ΔΔCT method. Methods 25:402-408 438 Lu P, Magwanga RO, Lu H, Kirungu JN, Wei Y, Dong Q, Wang X, Cai X, Zhou Z, Wang K Liu F (2018). A 439 novel G-protein-coupled receptors gene from upland cotton enhances salt stress tolerance in Transgenic 440 Arabidopsis. Genes, 9: 209 441 Martin FW (1965) Incompatibility in the sweetpotato. A review. Econ Bot Economic Botany 19:406-415 442 Massacci A, Nabiev SM, Pietrosanti L, Nematov SK, Chernikova TN, Thor K, Leipner J (2008) Response of the 443 photosynthetic apparatus of cotton (Gossypium hirsutum) to the onset of drought stress under field 444 conditions studied by gas-exchange analysis and chlorophyll fluorescence imaging. Plant Physiol 445 Biochem 46:189-195 446 Mbinda W, Anami S, Ombori O, Dixelius C, Oduor R (2016) Efficient plant regeneration of selected Kenyan 447 sweetpotato (Ipomoea batatas (L.) Lam.) cultivars through somatic embryogenesis. J Tissue Sci Eng 448 7:176 449 Mbinda W, Nawiri S, Nzaro M, Kinyagia B, Mgutu A. Oduor R (2015) Evaluation of the phosphomannose 450 isomerase-based selection system for genetic transformation of sweetpotato. Int J Life Sci 9:46-53 451 Mbinda W, Ombori O, Dixelius C, Oduor R (2018) Xerophyta viscosa Aldose Reductase, XvAld1, enhances 452 drought tolerance in transgenic sweetpotato. Mol Biotechnol 60:203-214 453 Mickelbart MV, Hasegawa PM, Bailey-Serres J (2015) Genetic mechanisms of abiotic stress tolerance that 454 translate to crop yield stability. Nature Rev Genet 16:237-252 455 Mohammadi M, Kav NN, Deyholos MK (2007) Transcriptional profiling of hexaploid wheat (Triticum aestivum 456 L.) roots identifies novel, dehydrationresponsive genes. Plant Cell Environ 30:630-645 457 Morán R, Garcıa, R, López, A, Zaldúa, Z, Mena J, Garcıa, M, Armas, R, Somonte D, Rodrıguez J, Gómez M, 458 Pimentel E (1998). Transgenic sweet potato plants carrying the delta-endotoxin gene from Bacillus 459 thuringiensis var. tenebrionis. Plant Science, 139:175-184 460 Mwanga RO, Andrade MI, Carey EE, Low JW, Yencho GC, Grüneberg WJ (2017) Sweetpotato (Ipomoea 461 batatas L.). In Genetic Improvement of Tropical Crops (pp. 181-218). Springer, Cham. 462 Orek, C. O., Zhang, P., Gruissem, W., Ferguson, M., & Vanderschuren, H. (2016). Gene expression associated 463 with ‘staygreen’trait for drought tolerance in cassava (Manihot esculenta Crantz). 464 Powell TL, Wheeler JK, de Oliveira AA, da Costa ACL, Saleska SR, Meir P, Moorcroft PR (2017) Differences 465 in xylem and leaf hydraulic traits explain differences in drought tolerance among mature Amazon 466 rainforest trees. Global Change Biol 23:4280-4293 467 Reddy AR, Chaitanya KV, Vivekanandan M (2004). Drought-induced responses of photosynthesis and 468 antioxidant metabolism in higher plants. J. Plant Physiol 161(11), 1189-1202 469 Rolando JL, Ramírez DA, Yactayo W, Monneveux P, Quiroz R (2015). Leaf greenness as a drought tolerance 470 related trait in potato (Solanum tuberosum L.). Environ Exper Bot 110:27-35 471 Rukundo P, Shimelis H, Laing M, Gahakwa D (2017) Combining ability, maternal effects, and heritability of 472 drought tolerance, yield and yield components in sweetpotato. Front Plant Sci 7:1981 473 Sánchez FJ, Manzanares M, de Andres EF, Tenorio JL, Ayerbe L (1998) Turgor maintenance, osmotic 474 adjustment and soluble sugar and proline accumulation in 49 pea cultivars in response to water stress. 475 Field Crop Res 59:225-235 476 Sánchez-Rodríguez E, Rubio-Wilhelmi M, Cervilla LM, Blasco B, Rios JJ, Rosales MA, Romero L, Ruiz JM 477 (2010) Genotypic differences in some physiological parameters symptomatic for oxidative stress under 478 moderate drought in tomato plants. Plant Sci 178:30-40 479 Seth MS, Bedada LT, Runo SM, Ngugi MP, Iyer R, Thomson JA, Mneney EE, Oduor RO. (2016) Drought 480 tolerance in transgenic tropical maize (Zea mays L.) by heterologous expression of peroxiredoxin2 481 gene-XvPrx2. Afr J Biotechnol 15:1350-1362 482 Turner NC (1981) Techniques and experimental approaches for the measurement of plant water status. Plant 483 Soil 58:339-366 484 Weigel D, Glazebrook J (2006) Transformation of agrobacterium using the freeze-thaw method. Cold Spring 485 Harb Protoc, 2006:1031-1036 486 Wong GR, Mazumdar P, Lau SE, Harikrishna JA (2018). Ectopic expression of a Musa acuminata root hair 487 defective 3 (MaRHD3) in Arabidopsis enhances drought tolerance. J Plant Physiol 231: 219-233 488 Zandalinas SI, Mittler R, Balfagón D, Arbona V, Gómez CA (2018) Plant adaptations to the combination of 489 drought and high temperatures. Physiol Plant 162:2-12 490 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

491 Figures

492 493 Fig. 1. Transgenic sweetpotato lines obtained by somatic embryogenesis regeneration and responses of the 494 drought stress in transgenic and wild type sweet potato plants. A. Developing embryo. B. Transgenic shoots 495 obtained through somatic embryogenesis. C and D. morphological appearance of transgenic and wild type 496 plants respectively. E and F. Phenotypes of transgenic and wild type plants grown in soil after 12 days of 497 drought stress respectively. G and H. Yield performance of wild type and transgenic plants grown under normal 498 conditions respectively. I and J. Yield performance of wild type and transgenic plants grown under drought 499 conditions for 90 days respectively. 500 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

14 Wildtype 12 Transgenic

6 Shoot length (cm) length Shoot 4

0 Normal Drought A 501 Growth conditions 12 Wild type Transgenic 10

Number of leaves 4

0 Normal Drought B 502 Growth conditions 503 Fig. 2. Phenotypic index changes of wild type and XvSap1 transgenic sweetpotato plants under normal and 504 drought stress condition. A. Effect of drought stress on shoot length. B. Effect of drought stress on number of 505 leaves. Data are from three independent replicates of the same event. 506 507 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

508

509 510 Fig. 3. Yield attributes of tolerant transgenic plants and wild type plants grown in the drought stress. A. fresh 511 weight B. Dry matter * and ** indicate a significant difference at p < 0.05. 512 513 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

300

250

200

150

Relative expression Relative 100

0 36912

514 B Number of days of drought stress 515 Fig. 3. qRT-PCR analysis of XvSap1 transcript levels in leaves of transgenic sweetpotato XSP1 line during 12- 516 day water deficit stress. 517

50 Wildtype 45 Transgenic

Chlorophyll values conent (SPAD) 20

15 9DBS 6DBS 3DBS 0DBS 3DAS 6DAS 9DAS 12DAS

A Number of days of water defict stress 518 bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

30 Wild type Transgenic 25

10 Proline content (µmol/gProline FW) 5

0 B Normal Drought 519 Growth conditions

30 Wild type 25 Transgenic

10 MDA content (nmol/g FW) (nmol/g content MDA 5

0 Normal Drought 520 C Growth conditions 521

100 Wild type Transgenic 80

Relative water content (%) 20

0 Normal 6 days drought 9 days drought 522 D Growth conditions bioRxiv preprint doi: https://doi.org/10.1101/603910; this version posted April 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

523 Fig. 5. Biochemical index changes of wild type and XvSap1 transgenic sweetpotato plants under normal and 524 drought stress condition. A. Effect of drought stress on chlorophyll content. Chlorophyll content was measured 525 as SPAD values. B. Effect of drought stress on proline content. C. Effect of drought stress on malondialdehyde 526 content. D. Effect of drought stress on relative water content. Data are from three independent replicates of the 527 same event.