Peng Wu , Xiaolan Xu , Jingwen Li, Jun Zhang, Siyuan Chang, Xiyan Yang , Xiaoping Guo

National Key Laboratory of Crop Genetic Improvement, Huazhong Agricultural University, Wuhan 430070, Hubei, China

Keywords:

ABSTRACT As the main byproduct of cotton fiber,the cotton seed yields about 1.6 times that of fiber,with its oil rich in unsaturated fatty acids, mainly linoleic acid.It is desirable for breeders to increase the oil content of cottonseed without affecting the yield and quality of cotton fiber.In this study, a seed-specific promoter- (alpha-globulin gene promoter-) driven GhDGAT1 overexpression vector (PαGlob-GhDGAT1)was constructed and used to transform an upland cotton line YZ1 (Gossypium hirsutum).Overexpression of the cotton gene GhDGAT1 in cotton seeds increased its total oil content from 4.7% to 13.9%in different transgenic lines and different generations.With the increase of oil content,the composition and contents of the main fatty acids in cotton seed also changed,as reflected by the contents of the main saturated fatty acids and unsaturated oleic acid.GhDGAT1 could be used to increase oil content and improve oil composition in cottonseed.

1.Introduction

As the main byproduct of cotton fiber production, the yield of cottonseed oil in China ranks fifth after soybean, palm, linseed,and rapeseeds oil [1].Cottonseed oil consists mainly of saturated and unsaturated fatty acids including palmitic, oleic, and linoleic acids,accounting for about 24%,22%,and 52%respectively.Cottonseed oil is also rich in the natural antioxidant tocopherol, which has vitamin E activity and contributes to its stability, giving the product a long shelf life[2].Plant seed storage oil is present mainly in the form of triacylglycerols(TAG).Diacylglycerol acyltransferase(DGAT), which catalyzes the formation of TAG by diacylglycerol(DAG) plus acyl-CoA, is a rate-limiting enzyme in the synthesis of TAGin vivo[3].At least five types of DGAT have been identified:DGAT1, DGAT2, WS/DGAT, soluble DGAT and DacT [4].

DGAT1,a major type ofDGAT,has been studied in most eukaryotes.The EMS-inducedDGAT1mutation line AS11 inArabidopsisshowed lowerDGATactivity, lower oil content, higher linolenic acid content, and later seed maturation than wild-type plants [5].Overexpression ofAtDGAT1inArabidopsisincreasedDGATactivity and oil content in seeds [6].Similarly, oil content was increased up to 114% whenDGAT1was overexpressed in rapeseed [7].The accumulation of TAG in transgenic tobacco leaves increased 20-fold and fatty acid increased about 2-fold whenAtDGAT1was introduced into tobacco [8].A homolog,GhDof1, which belongs to a large family of plant-specific transcription factors,DOF, was isolated from cotton (Gossypium hirsutum) [9].InGhDof1overexpression lines, the oil content in cotton seeds increased to 29.4%compared to 25.4% in WT, while the total protein in mature seeds decreased to 39.8% compared to 43.1% in transgenic lines [9].

Increasing oil content in seeds without influencing fiber quality is one of the breeding targets in cotton.Here we describe the construction of aGhDGAT1transgenic line driven by a seed-specific promoter, and analysis the oil content and the composition and contents of the main fatty acids in cotton seed.

2.Materials and methods

2.1.Plant materials and growth conditions

YZ1,an upland cotton line developed by the Henan Academy of Agricultural Sciences, Zhengzhou, China, is characterized by okra leaf, early maturity, non-glandlessness, high combining ability,and high resistance to Fusarium wilt.We have described the transgenic transformation system of YZ1 previously [10,11].TheGhDGAT1transgenic lines and the wild type(control)were planted in Wuhan,Hubei in summer in 2017–2019.The area used for each line was 5.4 m2(3.0 m×1.8 m)with four rows,and three replications were planted in plots of 21.6 m2for each replication.Field management followed local standard practice.Fertilizer application rates varied across years with total amounts of pure nitrogen 225–270 kg ha-1,P2O5275–344 kg ha-1,and K2O 217–272 kg ha-1.Phosphorous, potassium, and 50% nitrogen fertilizer were applied before sowing and the remaining nitrogen fertilizer was topdressed at flowering stage.

2.2.Yeast transformation, Nile red staining, and TAG quantification

The full-length CDS ofGhDGAT1gene (GH_A07G0132.1) was amplified by primersGhDGAT1-F andGhDGAT1-R (Table S1) and inserted in yeast expression vectorpYESby the Gateway method[12].pYES-GUSwas constructed as a control.

H1246, a mutant yeast strain (a tetrad ofdga1,lro1,are1, andare2) [13], was transformed withpYES-GUSandpYES-GhDGAT1.The transformed yeast lines were incubated in SD (–Ade, –His, –Leu, –Trp, –Ura) medium for 18 h at 30 °C until the OD600was 1.0.Yeast culture aliquots of 95 μL were then mixed with 5 μL of 0.8 mg mL-1Nile red in a 96-well plate and incubated at 30 °C for 5 min.Fluorescence signal was detected with a microplate reader at 485 nm and 538 nm [14].Three biological replications were performed.

The contents of triglyceride were also measured in the yeast cultures.ThepYES-GhDGAT1andpYES-GUSyeast cultures at logarithmic phase were refined by removal of supernatant after centrifuging at 1200×gfor 5 min and drying in a vacuum freeze dryer.Yeast cells (20 mg) were then washed with 1 mL PBS.The triglyceride content of each yeast line was determined with a triglyceride kit (E1013, APPLYGEN, Beijing, China, http://applygen.com/)according to the manufacturer’s instructions.Three biological replications were performed.

2.3.Construction of seed-specific over-expression vector and cotton transformation

The sequence of cotton alpha-globulin seed-specific promoter(PαGlob, GenBank accession number AX795651.1) was retrieved from NCBI(https://www.ncbi.nlm.nih.gov/).The full length ofPαGlobwas amplified with specific primers αGP-F and αGP-R(Table S1),withSacI andSpeI site.The PCR products were cloned into vectorpK2GW7.

The full length CDS of theGhDGAT1gene was inserted into αGPpK2GW7by the Gateway method.The resulting plasmids were designated asPαGlob-GhDGAT1, and introduced into theAgrobacterium tumefaciensstrains LBA4404 by electroporation and then used to transform cotton.A segregated wild-type plant harboring no transgenes, derived from the self-pollinatedAGP1overexpression line hemizygote, was used as null control.

2.4.Southern blotting, qRT-PCR, and DGAT enzyme activity assay

DNA was extracted from T0plants using a Plant Genomic DNA Kit (Tiangen, China, https://en.tiangen.com/).After confirmation by PCR with primers 35S-F andGhDGAT1-R (Table S1), Southern blotting [15]was performed for 8 positive T0plants with the digoxin-labeledNPT IIfragment as the DNA probe to detect insertion copies using a DIG High Prime DNA Labelling and Detection Starter Kit II (Roche, Basel, Switzerland, https://www.sigmaaldrich.com/).The single-insertion-copy transgenic lines were used for subsequent experiments.

Total RNA of 30-DPA (days post-anthesis) ovules and leaves of each transgenic line was isolated with a RNAprep Pure Plant Kit(DP441,Tiangen),and complementary DNA was synthesized using the SuperScript III reverse transcriptase(Invitrogen,Carlsbad,USA,https://www.thermofisher.com/).Quantitative real-time PCR(qRTPCR) was performed using the ABI Prism 7000 system (Applied Biosystems, Foster City, CA, USA).Gene-specific primer QGhDGAT1-F and reverse primer Q-GhDGAT1-R(Table S1)were used to amplify a 166-bp internal fragment ofGhDGAT1.GhUBQ7(Table S1)was used as internal control to normalize the expression levels of target genes and the relative expression levels were calculated using the modified 2-ΔΔCT method [16].Three biological replications were performed.

The 30-DPA ovules of T3plants were used for DGAT enzyme assay.The activity of the DGAT enzyme was determined with an ELISA kit (SU-B91387, Jining Shiye, Shanghai, China, http://www.shjning.com/).Three biological replications were performed.

2.5.Oil content, protein content, and fatty acid profiling analysis in cottonseed

The oil contents of the transgenic plants in T3–T5generations were assayed.The total oil content of dry seeds from each transgenic lines and control was measured by nuclear magnetic resonance spectroscopy (NIUMAG, Shanghai, China, https://www.niumag.com/).Three biological replications were performed.

Extraction of fatty acids from mature cotton seeds in T3–T5generations was performed as described previously[11].Briefly,about 200 mg of ground cotton seeds were added to 1.5 mL of 5% (m/v)sulfuric acid/methanol solution, 200 μL C19-nonadecanoic acid(2 mg mL-1) internal standard solution and incubated at 80 °C for 1 h.Then, 4 mL of NaCl buffer and 500 μL of n-hexane were added to the solution and mixed.After centrifugation at 1000×gfor 5 min, 150 μL of the supernatant was separated by gas chromatography–mass spectrometry (GCMS-QP2010 Ultra, Shimadzu,https://www.shimadzu.com/) according to previous publications[17].Five biological replications were performed.

Mature cotton seeds from the T3generation were ground in liquid nitrogen and 50 mg of powder was added to 1 mL PBS.The protein contents of all lines were determined with a BCA Protein Quantification kit (E112-01, Vazyme, China, http://www.vazyme.-com/) according to the manufacturer’s instructions.Three biological replications were performed.

2.6.Measurement of agronomic traits and fiber quality under field conditions

Agronomic traits were measured at the late stage of plant boll opening in T5generations,including first fruit branch position,fruit branch number, leaf branch number, effective boll number, plant height, average seed cotton weight (20 bolls were randomly selected), average lint weight (20 bolls were randomly selected)and lint score(The ratio of average lint weight to average seed cotton weight).Each strain was taken from the self-crossing middle boll, and a 10 g sample was prepared for fiber quality measurement using an HFT 9000 instrument (Premier, Coimbatore, India,https://www.gmdu.net), including four main indexes: upper half mean length, uniformity index, micronaire, and strength.Five plants were randomly selected for each strain.

In the cotton boll-opening period (about 120 days into the growing period), measurements were made from 10:00 to 12:00 AM at a mean daily temperature of 34 °C in clear weather.Five plants were randomly selected for each line in the T5generation.A Li-6400 Portable Photosynthesis System (LI-COR, Lincoln,NE, USA, https://www.licor.com/env/products/photosynthesis/)was used to determine intercellular CO2concentration, photosynthetic rate, conductance to H2O, and transpiration rate of the top three leaves of the plant.

2.7.Statistical analysis

Statistical significance was determined using Student’st-tests with GraphPad Prism7 (https://www.graphpad.com/), andPvalues ＜0.05 or ＜0.01 or ＜0.001 were considered statistically significant.

3.Results

The fluorescence of oil droplets in thepYES-GhDGAT1yeast was brighter than that in the controlpYES-GUS(Fig.1A and B), and the TAG content was different (P＜0.01) between thepYES-GhDGAT1yeast cells and the control (Fig.1C).

An overexpression vector was constructed to verify the function ofGhDGAT1in cotton.Eleven regenerated plants were obtained,eight of which were transgenic positive plants(Fig.S1).The transgenic copy numbers of eight T0generation materials were determined, and two single T-DNA copy lines (AGP1andAGP2) were obtained (Fig.1D).The expression levels ofGhDGAT1were higher(P＜0.01) in 30-DPA ovules in the transgenic lines than in a null line (a negative plant line segregating from theAGP1overexpression line)and wild types by qRT-PCR(Fig.1E),and the fold changes ofAGP1andAGP2compared to the null line were 10.40 and 4.12,respectively.However, the expression ofGhDGAT1in leaves did not differ among the transgenic plants,the null,and the wild type(Fig.S2A).The activity of the DGAT enzyme in 30-DPA ovules of all lines was determined by ELISA, and the activities were higher(P＜0.01) inAGP1andAGP2than in the null line in the T3generation (Fig.1F).

The oil contents of the transgenic plants in T3–T5generations were detected.The oil contents were higher(P＜0.01)in theAGP1andAGP2lines than in the null line across the T3–T5generations(Fig.1G).In the T3generation, the seed oil contents ofAGP1andAGP2lines were respectively 26.9% and 29.2%, values 4.7% and 13.9% higher than that of the null line (25.7%).The oil contents ofAGP1andAGP2seeds were respectively 28.6% and 27.7%, values 12.3%and 7.9%higher than that in the null line(25.7%)in the T4generation.In the T5generation,the seed oil contents ofAGP1andAGP2lines were respectively 31.9% and 31.6%, values 12.5% and 11.4%higher than that of the null line (28.4%).The protein contents in matureseeds ofAGP1andAGP2lines werelower(P＜0.01)thanthose in the null line and wild type in the T3generation(Fig.S2B).

The changes in total oil content were accompanied by changes in the contents of fatty acids.Oleic acid (C18:1) contents ofAGP1andAGP2were higher (P＜0.01) than those in the null lines for each generation, and palmitic acid (C16:0) and stearic acid(C18:0) contents were also higher (P＜0.05 orP＜0.01).However,the contents of linoleic acid (C18:2) were only slightly increased(P＞0.05) in most years (Fig.1H).

Some important agronomic traits, photosynthetic index and fiber quality ofGhDGAT1transgenic plants and null plants were determined in the T5generations.There were no differences(P＞0.05) in those traits betweenGhDGAT1transgenic plants and null plants (Tables S2 and S3; Fig.S3).

Thus, overexpression of the cotton geneGhDGAT1in yeast and cotton increased total oil content in both yeast cells and cotton seeds.With the increased oil content, the composition and contents of the main fatty acids in cottonseeds also changed, as reflected in the contents of the main saturated fatty acids and unsaturated oleic acid.

4.Discussion

DGAT1is the mainDGATcontrolling seed oil content and can contribute 20%–30% increases in oil content.In the present study,the increase in total oil content from 4.7% to 13.9% in different transgenic lines and generations suggests the presence of other DGATs in the cotton genome that regulate TAG synthesis.The expression level ofGhDGAT1was more than twice higher inAGP1than inAGP2,however,the total oil content and fatty acids seemed to have no difference between these two transgenics, or even higher inAGP2than inAGP1.The oil content has basically been accumulated at 30 DPA.Perhaps the expression level of the ovule at 30 DPA cannot reflect the oil content specifically, but the overexpression ofGhDGAT1can eventually increase the oil content of the cottonseeds to a certain extent.

The present study provides evidence and materials for improving oil content and changing oil composition in cotton seed.We use seed-specific promoters to drive gene expression, which does increase the oil content of seeds, with no changes in agronomic traits, photosynthetic index, and fiber quality ofGhDGAT1transgenic plants and null plants.

CRediT authorship contribution statement

Peng Wu and Xiaolan Xu carried out all the experiments with the help of Jingwen Li, Jun Zhang, and Siyuan Chang; Xiaoping Guo conceived the project,Xiyan Yang and Xiaoping Guo designed the experiments.Peng Wu managed the main data analysis and wrote the manuscript; Xiyan Yang revised the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This work was supported by National Science and Technology Major Project of the Ministry of Science and Technology of China(2016ZX08005-005).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2020.10.003.