Natural Albino Mutant of Daylily ( Hemerocallis spp.) Reveals a Link between Drought Sensitivity and Photosynthetic Pigments Metabolism

Background : Mutant analysis remains one of the main genetic tools for characterising unclarified gene functions in plants, especially in non-model plants. Daylily ( Hemerocallis spp.) is a popular perennial ornamental plant grown worldwide. Analysis of daylily mutants can enhance understanding of genes regulating the albino phenotype and improve the cultivar quality of daylily. Methods : The natural albino mutant (Alb − / − ) was isolated by screening a self-pollinated progeny of daylily cultivar ‘black-eyed stella’. Transmission electron microscopy was used in analysing the structure of plastids between mutant and wild-type seedlings. The content of chlorophyll, carotenoids and chlorophyll precursors in plants was measured by ultraviolet spectrophotometry. RNA sequencing and physiological measurements were performed to explore the association between drought tolerance and mutation. Results : All the seedlings of the daylily albino mutants died spontaneously within fifteen days after germination when grown in soil. The carotenoid and chlorophyll content in the leaves of the mutant plants significantly decreased compared with those of the wild-type control. The mutant plants displayed stunted growth, and their leaves were white or light yellow in color. Abnormal plastids such as those showing endomembrane vesiculation and lacking stacking were discovered in the leaves of mutant plants. Furthermore, genetic analysis revealed that a single recessive nuclear gene mutation led to the albino trait, RNA sequencing and real-time quantitative PCR validation showed extensive differences in gene expression between the mutant plants and the wild-type control, and most of the genes related to chlorophyll metabolism were down-regulated, with foldchange ranging from 0.20–0.49. Additionally, the surviving homozygous plants (Alb +/+ ), which do not contain this mutation, were also isolated by analysing the phenotype of their self-pollinated progeny. The net photosynthesis rate and light saturation point of Alb +/+ were higher than those of heterozygous (Alb +/ − ) plants. Additionally, the Alb +/+ plants were more tolerant to drought conditions than the Alb +/ − plants, suggesting that a heterozygous Alb − mutation is sufficient to negatively affect photosynthetic efficiency and drought tolerance. Conclusions : The albino mutation negatively affects photosynthetic efficiency and drought tolerance, and homozygous mutation is required for the characteristic albino phenotype. This work highlights the link between albino mutation, photosynthetic pigment metabolism and drought sensitivity in daylily.


Introduction
Photosynthesis converts solar energy into chemical energy and relies on photosynthetic pigments that absorb light energy.Albino mutations in plants affect photosynthesis rate by reducing the content of chlorophyll and carotenoids, which are the main photosynthetic pigments.Chlorophyll deficiency is generally caused by the mutations of genes encoding enzymes that catalyse biochemical reactions involved in chlorophyll metabolism or genes responsible for changes in chloroplast ultrastructure [1][2][3].Carotenoids belong to a large pigment group and are categorised into carotenes, like α-carotene and β-carotene [4] or xanthophylls, such as zeaxanthins, violaxanthins, indicaxanthins, and lutein, etc. [5].Most carotenoids are in-volved not only in plant photosynthesis, but also in photoprotection of photosynthetic organs [6].Chloroplasts, as the primary organelles involved in photosynthesis, play an important role in plant stress responses [7].Furthermore, it was suggested that high levels of photosynthetic pigments confer tolerance to drought conditions in plants [8].Albino mutants lack plant photosynthetic pigments.They have been identified and characterised in Arabidopsis [9,10], cotton [11], maize [12], pepper [13], tomato [14], and tobacco [15].Moreover, the photosynthetic pigment content in apple and potato was negatively affected by drought stress [16,17].The visible phenotype of albino mutants is leaf chlorosis, which is often accompanied by reduced yield or even death.Therefore, the albino pheno-type is considered a non-adaptive mutation [11].Mutant analysis remains one of the main genetic tools for characterising gene functions in plants, especially non-model plants, such as daylily (Hemerocallis spp.) [18].Great progress has been made by studies on the metabolism of chlorophyll and carotenoids in the analysed albino mutants in recent years.For example, the precursors of carotenoid biosynthesis are isopentenyl pyrophosphate and its isomer dimethylallyl diphosphate [19].These compounds are not only the precursors of carotenoids, but also of chlorophyll, abscisic acid, cytokinin, and gibberellin [20].Thus, the intermediate products of the methyl-D-erythritol 4-phosphate (MEP) pathway play an important role in chlorophyll and carotenoid biosynthesis.In fact, the disruption of enzymes encoded by genes in the MEP pathway may reduce the chlorophyll and carotenoid content of plants.For example, Arabidopsis loss-of-function mutations in genes involved in the MEP pathway (dxs, dxr, IspD, IspE, IspF, IspH and gps1) result in an albino leaf phenotype with reduced chlorophyll and carotenoid levels [9,10,[21][22][23].The disruption of the pds3 gene impairs chlorophyll, carotenoid and gibberellin biosynthesis in Arabidopsis [24].Furthermore, silencing the expression of IspG and IspH in tobacco results in an albino leaf phenotype, whereas silencing the idi gene in tobacco results in mottled white-pale green leaves [25].The Arabidopsis double mutant of the ipi genes exhibited reduced chlorophyll and lutein content in leaves [26].These data indicate that genes involved in the MEP pathway are important factors in the regulation of chlorophyll and carotenoid biosynthesis.In addition to the obstruction of chloroplast synthesis and metabolism, mutations in genes related to chloroplast development lead to an albino phenotype [27][28][29].
Daylilies (Hemerocallis spp.) are popular perennial ornamental plants.Most daylily cultivars are developed by sexual hybridisation [30], and their population amplification depends on division propagation.The American Daylily Association (https://www.daylilies.org/DaylilyDB/) has collected approximately 90,000 daylily cultivars.Mutant analysis is a widely used tool for characterising gene functions in plants including non-model plants, such as daylily, and great progress has been made in cloning genes from plant mutants in recent decades.Several studies have reported gene cloning using daylily [31][32][33][34][35][36][37].Mappingbased cloning is an effective strategy for identifying causal mutations responsible for variations in a trait.In addition to mapping-based cloning, microarray [38], RNA sequencing (RNA-seq) [39], and mapping by sequencing [18,40] are used in analysing the gene functions of mutants.Giving that the vegetative growth period of daylily is longer than 2 years, mapping-based cloning is not an ideal method for analysing the natural mutants of daylily.Thus, RNA-seq was performed to compare gene expression between wildtype and albino mutant plants of daylily in the present study.
To understand the genetic and physiological mechanisms underlying daylily albino mutations, we performed its functional characterization and determined the most likely genetic and physiological causes.Moreover, impacts on the physiological functions of the plant, such as photosynthesis, metabolism and tolerance to drought, were explored.

Plant Materials and Growth Conditions
Hemerocallis Middebdorffii Trautv.& C. A. Mey and Hemerocallis 'black-eyed stella' and 'stella de oro' were cultured in experimental fields (Shanghai Institute of Technology, N: 30  .Plants were grown in soil and cultured in a medium containing half-strength Murashige and Skoog (MS) salts [41] and 3% sucrose (v/v), MS medium, sucrose, and other drugs were purchased at Shanghai Titan Technology Co., Ltd.(Shanghai, China).

Microscopic Analysis
The sections of daylily leaf tissue from albino mutants and wild-type plants were observed and recorded under a light microscope (ECLIPSE E200, NIKON, Japan).
Transmission electron microscopy analysis was performed according to the methods described by Hashimoto et al. [42] with some modifications.In detail, the leaf tissues of mutant and wild-type plant leaves were sampled, fixed and pumped with 4% glutaraldehyde at 4 °C, stored at 4 °C overnight and washed three times with 0.1 mol•L −1 phosphate buffer (pH = 7).Then, the leaf samples were fixed with 1% osmic acid for 2 h and washed three times with phosphate buffer.The leaf samples were then dehydrated in a series of ethyl alcohol solution gradually followed by acetone, embedded in resin, and sectioned by slicing in an ultrathin slicing machine (EM UC7, LEICA, German).The samples were stained and then observed by transmission electron microscope.

Determination of the Chlorophyll and Carotenoid Content
The contents of chlorophyll and carotenoids were determined according to the methods described by Chen [43] with some modifications.The leaf samples of 20d daylily seedlings were ground using a pestle and mortar.Chlorophyll and carotenoids were extracted using a prepared extraction solution containing ethanol, acetone, and water in a ratio of 5:4:1 (v/v/v).The absorbance was measured by a microplate reader (Spark readers, TECAN, Swiss) at wavelengths of 470, 645 and 663 nm.The content of total chlorophyll, chlorophyll a (Chla), chlorophyll b (Chlb) and carotenoids was calculated according to the following formula:

Determination of the 5-Aminolevulinic Acid
The content of 5-aminolevulinic acid (ALA) was determined according to the methods described by Dei [44] with some modifications [45].Leaf samples of 12-day-old daylily seedlings were ground and ALA was extracted by 4% trichloroacetic acid (m/v).Approximately 5 mL of extraction solution, 2.35 mL of sodium acetate (1 mol•L −1 ), 0.15 mL of acetylacetone, and 2.5 mL acetate buffer (1 mol•L −1 , pH 4.6) were mixed and placed in boiling water for 10 min.The coloration of the mixture was performed by adding an equal volume of Ehrlich-Hg reagent after cooling.The absorbance was then measured by a microplate reader at 553 nm.The ALA content was then calculated according to the formula 7.2 × 10 4 × D 553 .

Determination of Porphobilinogen
The content of porphobilinogen (PBG) was determined according to the methods described by Peng et al. [46] with some modifications.Leaf samples from 12 d old daylily seedlings were ground, and PBG was extracted by buffer containing 0.1 mol•L −1 EDTA and 0.6 mol•L −1 Tris-HCL (pH 8.2).Coloration of the extraction was performed by adding an equal volume of Ehrlich-Hg reagent.Then the absorbance was measured by a microplate reader at 553 nm.The content of PBG was calculated according to the formula: 6.1× 10 4 × D 553 .

Determination of Uroporphyrinogen III and Coproporphyrinogen III
The contents of uroporphyrinogen III (urogen Ⅲ) and coproporphyrinogen III (coprogen Ⅲ) were determined according to the methods described by Bogorad et al. [47] with some modifications.Leaf samples of 12 d old daylily seedlings were ground, and PBG was extracted by 0.067 mol•L −1 phosphoric acid buffer (pH 6.8).Then, 5 mL of extraction solution and 0.25 mL sodium thiosulfate were mixed and illuminated with intense light for 20 min.The pH of the mixture was adjusted to 3.5 by adding 1 mol•L −1 glacial acetic acid.The absorbance of the aqueous phase, which was combined after chloroform extraction, was measured by a microplate reader at 405.5 nm.The urogen Ⅲ content was calculated using the formula: 5.48× 10 5 × D 405.5 .The absorbance of the aqueous phase, which was combined after ether extraction, was measured by microplate reader at 399.5 nm.The content of coprogen Ⅲ was calculated according to the formula: 4.89× 10 5 × D 399 .

Determination of Protoporphyrin IX, Mg-Protoporphyrin IX and Pchlide
The content of protoporphyrin IX (proto Ⅸ), Mgprotoporphyrin IX (Mg-proto Ⅸ), and pchlide were determined according to the methods described by Hodgins et al. [48] with some modifications [46].Leaf samples from 12 d old daylily seedlings were ground after the addition of 25 mL of 80% alkaline acetone.The extraction was centrifuged at 15000 ×g and 4 °C for 15 min.The absorbance of the supernatant was measured by a microplate reader at wavelengths of 575, 590, and 628 nm.The content of proto Ⅸ, Mg-proto Ⅸ, and pchlide were calculated according to the following formula:

Determination of Photosynthetic Parameters
Photosynthetic parameters were determined according to the methods described by Iqbal et al. [45] with some modifications [49].A portable photosynthesis tester (CIRAS-3, PP SYSTEMS, USA) was used to determine photosynthetic parameters.The fifth leaves from apical meristem of daylily were selected for measurement in September and October 2022 (10:00-12:00 am).Photosynthetic parameters including net photosynthetic rate (Pn), water use efficiency (WUE), transpiration rate (Tr) and intercellular CO 2 concentration (Ci) were recorded after that the absolute value of the change amplitude of Pn was less than 0.2.

Determination of the Relative Water Content of Leaves
The relative water content of the leaves was determined according to the methods described by Chen et al. [50] with some modifications.Daylily leaves with the same physiological status were treated under drought conditions, and weighed.Fresh weight was recorded as the initial weight.The saturated weights (Wt) were recorded after the samples being placed in distilled water for 2 hours to absorb water.After that, the dry weights (Wd) were recorded after the samples being dried at 80 °C for 48 h.The water content (RWC) was calculated according to the following formula:

Determination of Soil Moisture Content
Soil moisture content was determined according to the methods described by Wang et al. [51] with some modifications.The wet weight (M1) of the soil was acquired by randomly weighing the soil from three different positions in the pot.Dry weight (M2) was recorded after the samples being dried at 80 °C for 48 h.The soil moisture (W) content was calculated according to the following formula: (M1 -M2)/M2 × 100%.

RNA Sequencing (RNA-seq)
Leaf samples from 12 d old albino mutant and wildtype plants were ground, and five sets of biological replicates of each sample were randomly sampled.Three sets with the good-quality samples were obtained for analysis.Total RNA was extracted after digestion by DNase.The mRNA was enriched by magnetic beads with oligo dT.Used mirVana™ miRNA ISOlation Kit, Ambion-1561 kit and Refrigerated centrifuge (ST16R, Thermo, USA), Gel imaging system (Tanon 2500, Biotanon Co., Ltd., Shanghai, China), UV spectrophotometer (NanoDrop 2000, Thermo, USA).
cDNA libraries were constructed using a VAHTS Universal V6 RNA-seq Library Prep Kit for Illumina sequencing according to the manufacturer's instructions.The libraries were finally sequenced using an Illumina Novaseq 6000 platform.A total of 150 bp paired-end reads were generated.The transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd.(Shanghai, China).Oligonucleotide adapters were removed using Trimmomatic software (version 0.36) [52], and clean reads were obtained after low quality bases and N bases were filtering out.Clean reads were then spliced through the pairedend method performed using Trinity software (version 2.4) to obtain transcript sequences [53].The most extended sequence was selected as a unigene.The unigenes were annotated to the NR, COG/KOG, and Swissprot databases.GO annotation was acquired by mapping Swissprot ID to the GO term, and pathway information was acquired by comparing unigenes with the KEGG database [54].
Fragments per kilobase of exon model per million mapped fragments (FPKM) and unigene count were analysed in bowtie2 (version 2.3.3.1)[55] and eXpress software (version 1.5.1)[56].The number of unigene reads in the samples were acquired by eXpress software.All the data were standardised by estimating SizeFactors, a function of DESeq package (version 1.18.0)[57].The p-value and foldchange of the difference in gene expression between mutant and wild-type plants were calculated using nbinomTest software.The unigenes with difference value greater than 2 and p value less than 0.05 were selected for further analysis.

Quantitative Real-Time PCR (qPCR)
Six differentially expressed genes (DEGs) from RNAseq data related to terpene, carotenoids, and chlorophyll metabolism were randomly selected for quantitative realtime PCR (qPCR) validation.Primers were designed using premier5 software (Supplementary Table 1).The Prime ScriptTM RT reagent kit with gDNA Eraser (TaKaRa) was applied to synthesise cDNA.qPCR was then performed according to the methods described by Zuo [58].HfUBQ and HfEF-1a were used as internal reference genes [59].The 2 −∆∆Ct method was used for data analysis [50].

Drought Stress
Two-week-old sterile cultured self-pollinated progeny of daylily 'Black-eyed stella' were transplanted to a medium containing half-strength MS salts, 3% sucrose (v/v), and 10% PEG2000 (v/v) [60].One-year-old daylily plants, which were cultured in pot with soil, were left unwatered for two weeks, whereas the control plants were regularly watered every two days.Relative soil water content was determined at the same time.

Statistical Analysis
At least three independent biological replicates were performed for each experiment.Statistical analysis was performed with Excel 2010 and SPSS26.Figures were made using Origin8 software.

Isolation of a Natural Albino Mutant of Daylily
Albino seedlings were observed during the selection and development of new varieties from self-pollinated progeny of daylily 'Black-eyed stella' and one of its parents, 'Stella de oro' in May 2020.The leaf colour of the albino seedlings was light yellow (Fig. 1A).The albino seedlings displayed a dwarf phenotype (Fig. 1B), and survived for less than 15 days in soil.However, these seedlings survived longer when grown on the medium containing half MS salts and 3% sucrose under sterile culture conditions (Fig. 1C-F).The chlorophyll and carotenoid content of albino seedlings was determined and compared with those of wild-type plants.The results showed that the content of chlorophyll a (Chla), chlorophyll b (Chlb) and carotenoids decreased significantly compared with that of the wild-type plants (Supplementary Fig. 1).Albino phenotypes are generally attributed to changes in the metabolism of photosynthetic pigments and chloroplast development.Therefore, a microscopic analysis was performed to verify whether the chloroplast development of the mutant plants was altered.Extremely few chloroplasts were found in the albino mutant leaf cells (Fig. 1G,H).To better understand chloroplast development, transmission electron microscopy was used for investigating the morphology of mesophyll plastids.Both the number of thylakoids and the extent of stacking in the grana were significantly reduced in albino mutant plants (Fig. 2).
To understand the mechanism of the decrease in chlorophyll content, the precursors of chlorophyll and carotenoids biosynthetic pathways were measured.Except PBG, the content of the precursors of chlorophyll biosynthesis significantly decreased in the leaves of albino mutants compared with that of wild-type plants (Supplementary Fig. 3).

Genetic Analysis of the Albino Phenotype
Daylily 'Black-eyed stella' is a cross between two diploid parents 'Stella de oro' and 'Little celena' (https:// www.daylilies.org/DaylilyDB/).Albino mutants were isolated from the self-pollinated offspring population of 'Black eyed stella' and one of its parents, 'Stella de oro'.The F2 population of 'Black-eyed stella' was used for further analysis, and the segregation ratio was in accordance with the expected Mendelian ratio of 3:1 (Table 1).Furthermore, all cross-pollinated offspring between H. 'black-eyed stella' The surviving F2 plants should have consisted of twothirds of heterozygous plants and one-third of homozygous plants, because the albino mutant plants died before flowering.The plants with self-pollination progeny that did not appear as albino seedlings were designated as homozygous plants (Alb +/+ ), and the other F2 plants were then marked as heterozygous (Alb +/− ).

RNA Sequencing (RNA-seq) Analysis and Real-Time PCR Confirmation
RNA-seq was performed to compare gene expression patterns between the mutant and wild-type plants.Then, 41.96 G clean data were acquired.The effective data volume of each sample ranged from 6.74 G-7.17 G.The Q30 bases were distributed in 95.47-95.55%,and the average GC content was 46.59%.A total of 53479 unigenes were spliced and had a total length of 55899490 base pairs (bp) and an average length of 1045.26 bp (Supplementary Table 3).These results showed that the quality of the RNAseq data was good enough for subsequent analysis.A total of 7952 DEGs, including 4069 up-regulated genes and 3883 down-regulated genes were identified between daylily albino mutant and wild-type plants.GO function enrichment analysis showed that many DEGs were involved in biological processes related to cellular components and molecular function, such as cellular process, metabolic process, cell part, organelle, membrane, membrane part, membrane-enclosed lumen, binding, catalytic activity, and transporter activity (Supplementary Fig. 2).KEGG significant enrichment analysis revealed that the DEGs were involved in 125 metabolic pathways.The metabolic pathways with the most DEGs were carbohydrate, lipid, amino acid, and energy metabolism.The downregulated DEGs were pyruvate, amino sugar, nucleotide, starch and sucrose metabolism (Supplementary Fig. 3, Supplementary Table 4).
Six DEGs were randomly selected from genes encoding for terpene, carotenoid and chlorophyll metabolism and one transcription factor for quantitative real-time PCR (qPCR) validation, because all these genes were related to the albino trait (Fig. 3).The expression level of DEGs between qPCR and RNA-seq data was coincident (Supplementary Table 5).Therefore, the RNA-seq data described here can be considered reliable.
Given that the level of chlorophyll and carotenoid in albino mutants were much lower than those of wildtype plants (Supplementary Table 2), the DEGs related to chlorophyll biosynthesis were analysed with the RNA-seq data, showing that most of the DEGs related to chlorophyll biosynthesis were down regulated (Fig. 4, Ref. [61][62][63]).The fold change range was 0.20-0.49(Supplementary Table 6).CAO, a gene necessary for transforming Chla to Chlb in chlorophyll oxygenase biosynthesis [64], was upregulated (Fig. 4).This result might be responsible for the observed decrease in the Chlorophyll a/b ratio in the mutants.
The direct precursor of carotenoids biosynthesis is geranylgeranyl pyrophosphate (GGPP), which is also a precursor of chlorophyll biosynthesis [24].Therefore, the DEGs of terpene metabolism were analysed in detail.The results indicated that HfDXS, which encodes a key enzyme in the MEP pathway, was significantly negatively regulated (Figs.3,4).The expression level of HfDXS in albino mutant plants was down-regulated 3.5 times compared with that in the wild-type control (Supplementary Table 5).Regarding the mevalonate pathway (MVA) path- way for terpene metabolism, the genes encoding acetyl-CoA C-acetyltransferase and hydroxymethylglutaryl-CoA synthase were down-regulated (Fig. 4).These DEGs act upstream of the metabolic pathways of MEP and MVA.Therefore, changes in these DEGs likely substantially contributed to decreases in chlorophyll and carotenoid content in the leaves of the mutant plants.Regarding other genes relevant to carotenoid biosynthesis, two up-regulated and three down-regulated DEGs were found (Table 2).The up-regulated DEGs were genes involved in the biosynthesis of ϵ-carotene and lutein.The foldchange values were 2.40 and 8.62, respectively (Table 2).The down-regulated DEGs were genes involved in biosynthesis of aldehydes, capsaicin and 2-cis-4-trans-xanthin, and the fold change range was 0.38, 0.43 and 0.19, respectively (Table 2).The transcription factor WRKY genes regulate the expression of DXS genes [65,66].Therefore, the bioinformatic analysis of the differential expression of the HfWRKY genes with RNA-seq was performed and followed by quantitative PCR confirmation, showing that HfWRKY24 was remarkably upregulated (Fig. 3 and Supplementary Table 5).

Analysis of the Photosynthetic Characteristics in the Homozygous (Alb +/+ ) and Heterozygous (Alb +/− ) Plants
The photosynthetic pigment contents of Alb +/+ and Alb +/− plants were measured after being identified by the albino phenotype that appeared in self-pollination-derived progeny.The content of chlorophyll and carotenoids in the leaves of Alb +/− plants was significantly lower than that of the Alb +/+ plants (Supplementary Table 7).Measuring the photosynthetic characteristics of Alb −/− was difficult because of the small plant size and short surviving time.The results showed that not only the net photosynthetic rate (Pn), but also the transpiration rate (Tr), intercellular CO 2 (Ci), and water use efficiency (WUE) of Alb +/+ were significantly higher than those of the Alb +/− plants (Fig. 5A-D).The photosynthetic response curves of Alb +/− and Alb +/+ plants were further analysed because of the difference in Pn between the Alb +/− and Alb +/+ plants.The Pn of the Alb +/− plants was significantly lower than that of the Alb +/+ plants when the light intensity was higher than 100 µmol photons m −2 •s −1 (Fig. 5E), indicating that the light saturation point of Alb +/+ plants was higher than that of the Alb +/− plants.qPCR was performed to confirm whether the above differences in phenotypes between Alb +/+ and Alb +/− plants were accompanied by changes in the expression levels of HfDXS and HfWRKY24.The expression level of HfDXS in the Alb +/+ leaves was slightly higher than that in the Alb +/− leaves, but the expression level of HfWRKY24 in the Alb +/+ leaves was significantly higher than that in the Alb +/− leaves (Supplementary Table 8).In terms of photosynthetic characteristics, the Pn of Alb +/+ declined after drought treatment.However, the trend of decline slowed down after 4 d treatment.Meanwhile, the Pn of Alb +/− decreased continuously during the treatment period.The change trend of the stomatal conductance was consistent with that of Pn between Alb +/+ and Alb +/− after drought treatment (Fig. 6).

Daylily Albino Mutant Plants are Sensitive to Drought Conditions
As photosynthetic efficiency affects the drought tolerance of plants [51], RWC of Alb +/+ and Alb +/− plants were measured after drought treatment.The RWC of the leaves in Alb +/− plants declined immediately after drought treatment.Meanwhile, the RWC of the leaves in Alb +/+ plants showed a significant reduction only after 1 week of treatment (Fig. 6), indicating that the Alb +/− plants were more sensitive to drought conditions than the Alb +/+ plants.This result prompted us to investigate whether albino mutant seedlings were also sensitive to drought conditions.Owing to the short life of mutant seedlings cultured in soil, sterile culture with sucrose and PEG were used for drought stress treatment.Initially, the albino leaves grew similarly to the wild type (Fig. 7A), and as the number of days of stress increased, the albino leaves began to wilt and collapse (Fig. 7B).The mature leaves of albino seedlings almost completely wilted and were lodging after treatment (Fig. 7C), and dark brown spots appeared on the mature leaves (Fig. 7D).Meanwhile, the wild-type seedlings were almost unaffected by drought treatment (Fig. 7C).These results indicated that albino mutant is sensitive to drought, because high-MW PEG is considered the most suitable modelling system for modelling drought treatment in plants grown in vitro [67].Furthermore, the 1-year-old plants cul- tured in pots with soil were exposed to drought conditions.The results showed that the leaves of Alb +/− plants curled downward and wilted after 1 week of drought treatment (Fig. 7F-K).However, the Alb +/+ plants were almost unaffected after the first week of drought treatment (Fig. 7M-R).The leaves of Alb +/− and Alb +/+ plants were smooth and flat before drought treatment (Fig. 7E,L).In addition, the degree of curling of Alb +/− leaves was higher than that of the Alb +/+ leaves (Fig. 7K,R).These results indicated that Alb +/+ plants were more tolerant to drought treatment compared with Alb +/− plants.

Discussion
This study reports the isolation of a natural albino daylily mutant (Alb −/− ) and homozygous lines (Alb +/+ ) based on the screening of self-pollinated progenies from a daylily heterozygous cultivar.Compared with the heterozygous cultivar, Alb −/− plants showed lower chlorophyll and carotenoid content, and Chl a/b ratio, while Alb +/+ plants showed a higher chlorophyll and carotenoid content (Supplementary Table 7).Genetic analysis revealed that the albino trait can be attributed to a single recessive nuclear gene.Furthermore, the results indicated that the mutated gene is related to drought sensitivity.
The analysis of RNA-seq and quantitative PCR indicated that expression of many genes was altered in mutant  2).To further understand the RNA-seq data, the top ten up-regulated and down-regulated DEGs have been analysed.Regarding the top 10 downregulated genes, there were 4 genes encoding photosystem II Psb, 1 gene encoded for a heat shock protein, 2 genes encoded for cytochrome, 1 gene encoded for isoprene synthase, 1 gene encoded chalcone synthase, and 1 gene encoded for NAD (P) H-quinone oxidoreducase (Supplementary Table 6).Psb plays a role in protecting the PS II system [68].So far no reports of downregula-tion of psb genes that affect leaf colour have been published.Regarding heat shock proteins, abnormal expression of genes encoding heat shock proteins was shown to cause abnormal chloroplast development, which in turn affects the colour of leaves [69,70].The expression of a gene encoding cytochrome was showed altered in albino mutants, but its gene function is related to plant growth and development, as well as responding to biotic and abiotic stresses, and does not appear to be a key gene that causes leaf colour mutations in plants [71,72].Isoprene synthase regulates an important reaction within the MEP pathway [63][64][65].However, no reports of downregulation of isoprene synthase-encoded genes to affect leaf colour have been published so far.Overexpression of the chalcone synthase-encoded gene increases the tolerance of plant leaves to strong light due to its high level of anthocyanin synthesis [73].The top 10 genes upregulated were genes encoded beta-glucosidase, polygalacturonase, allene oxide synthase, autophagy-related protein, probable linoleate 9Slipoxygenase, heat shock protein, ubiquitin-ribosomal protein, NADPH-dependent aldo-keto reductase, acetyl-CoA acyltransferase and DNA-directed RNA polymerase subunit beta (Supplementary Table 6).These genes are related to the activity of β-glucosidase and polygalacturonase, the induction of jasmonic acid by stress treatment, and cellular autophagy, respectively [74].Regarding the DEGs related to the albino phenotype [14,23,29,[75][76][77][78][79][80][81][82][83][84][85][86][87], 5 genes were down-regulated and 5 genes were upregulated (Supplementary Table 9).The seedlings of the LCYE mutant were white or light yellow, and can survive and grow until flowering [75].SGR Over-expressing resulted in an albino phenotype [76], and albinism causes up-regulated expression of the PPR gene [77].The loss-of-function of ALB3, CAO, DXR, and NUS1 resulted in an albino phenotype [14,29,[82][83][84][85][86].However, the expression changes of these DEGs in the albino mutant of daylily described here were completely opposite to those of the above-mentioned mutants.Downregulation of PORA caused dwarfism with a light green leaf colour [78].Downregulated gene encoded for δ-aminolevulinic acid dehydratase causing yellow spots on leaves [79].Therefore, the phenotype of the daylily mutant described here was different from these two mutants in surviving time and leaf colour.In particular, HfDXS, which encodes the first key enzyme of the MEP pathway, was down-regulated approximately five times in the mutant plants and increased slightly in Alb +/+ plants compared with the heterozygous plants.GGPP, an intermediate of MEP pathway, is a common precursor of chlorophyll and carotenoids biosynthesis [24].A similar phenotype with decreased chlorophyll and carotenoid content has been reported in Arabidopsis and tomato DXS mutants [9,10,88].
The results showed changes in the precursor contents of chlorophyll biosynthesis and a decrease in the Chl a/b ratio in mutant plants compared with wild-type plants.The possible reason was that Chla may not be as stable as Chlb.The decrease in the Chl a/b ratio can be attributed to the increased expression of CAO (Fig. 4), which encodes the synthesis of chlorophyll oxygenase.Changes in the content of chlorophyll biosynthesis precursors between albino and wild-type seedlings showed that urogen III synthase is the key enzyme for the decreasing chlorophyll content (Supplementary Fig. 2).However, the expression level of this gene revealed from the RNA-seq data did not change significantly.Changes in urogen III synthase activity may be a side-effect of a reduction in the expression level of HfDXS in the albino seedlings.Moreover, the chloroplast development of the mutant was impeded, indicating that the mutated gene of daylily affected not only photosynthetic pigment biosynthesis, but also chloroplast development.The results were consistent with the DXS loss-offunction mutants of Arabidopsis and tomato [9,10,88], supporting the notion that the albino phenotype described here might be influenced by the down-regulation of HfDXS.The higher expression level of HfWRKY24 in both the Alb +/+ and Alb −/− plants compared with the Alb +/− indicated that HfWRKY24 is likely not a major factor that induced the albino phenotype of daylily.
The present results also showed that the light saturation point and the net photosynthesis rate of the Alb +/+ plants were significantly higher than those of the Alb +/− plants.The higher value of light saturation points of Alb +/+ plants compared with the Alb +/− plants should be attributed to the high chlorophyll and carotenoid content of the Alb +/+ plants, because carotenoids protect chlorophyll from oxidative decomposition by light.In addition, studying the shade tolerance of Alb +/− plants will be valuable for the understory cultivation of daylily, because shade tolerant daylily is relatively scarce in the market.
Our results showed that the RWC of the Alb +/+ plants was higher than that of the Alb +/− plants after drought treatment, indicating that the Alb +/+ plants were more tolerant to drought conditions compared with the Alb +/− plantsas RWC is considered an important indicator of plant drought tolerance [50].Indeed, drought treatment experiments indicated that the Alb +/+ plants were tolerant to drought conditions, and the albino mutant were more sensitive to drought conditions compared with the Alb +/− plants.This finding can be due to the higher photosynthetic efficiency of Alb +/+ plants compared to that of Alb +/− plants, consistent with the study on genetic variation in a vacuolar H+-PPase encoded genes [51].The high value of net photosynthesis rate of the Alb +/+ plants compared with the Alb +/− plants can be due to the high concentration of photosynthetic pigments of Alb +/+ plants compared with the Alb +/− plants, because a high concentration of photosynthetic pigments can capture more light energy to form a developed root system.Conversely, difference between the albino mutant and heterozygous plants in drought tolerance can be the consequence of direct regulation of the drought sensitivity trait by the mutated gene.In addition, the Alb +/+ plants will be a good resource for the future breeding of daylily because of its drought tolerance trait.

Conclusions
The first natural albino mutant of daylily was isolated in this study.According to Mendelian proportions of the phenotypes in the progeny of heterozygous plants, the albino phenotype is caused by a single recessive mutation.The homozygous mutant plants suffered the pronounced downregulation of HfDXS1, causing a dramatic decline in chlorophyll and carotenoid levels and were unable to live in the soil for longer than 15 days.Heterozygous Alb +/− plants can be different from their homozygous Alb +/+ counterparts in terms of the occurrence of albino mutants in self-pollination-derived progeny.These heterozygous Alb +/− plants showed a reduced light saturation point, net photosynthetic rate, and water use efficiency, and reduced tolerance to drought compared with homozygous Alb +/+ plants, indicating that the presence of a mutated allele affected the viability of the heterozygous plants even in the absence of a visible albino phenotype typical of homozygous mutants.Further research will be needed for the exact mapping of the mutation described in this study, and for establishing a clear mechanism for its effects on plant viability, photosynthetic parameters, and drought tolerance.In addition, this study provided important resources for daylily breeding.

Fig. 1 .
Fig. 1.Comparison of the morphological and characterization of the plastid between albino and wild-type seedlings of daylily.(A) Seedlings of 15 d old wild-type (right) and albino mutant (left) seedlings cultured in the soil.(B) Height of seedlings of 15 d wild-type (left) and albino mutant (right) seedlings cultured in the soil.(C-E) Seedlings of albino mutant culture on the media containing half MS salts and sucrose under sterile conditions.(G,H) Light microscopy of the chloroplast of the leaf cells of albino mutant (G) and wild-type seedlings (H).The bars in (A-F) represent 1 centimeter, and in (G,H) represent 20 micrometers.Mean values ± SE (n = 3); significant values **p < 0.01.

Fig. 2 .
Fig. 2. Microscopic analysis of the plastids tissue between albino mutant and wild type seedlings of daylily.(A,C,E) Transmission electron microscopic examination of plastids of albino seedlings.(B,D,F) Transmission electron microscopic examination of plastids of wild-type.The bars in (A,B) represent 10 micrometers, (C,D) represent 2 micrometers, and (E,F) represent 1 micrometer.

Fig. 6 .
Fig. 6.Photosynthetic capacity and relative water content of Alb +/+ and Alb +/− plants of daylily after drought treatment.(A-E) Photosynthetic capacity and relative water content of Alb +/+ and Alb +/− plants of daylily were compared under progressive drought stress under laboratory conditions.Data were obtained from three two-year-old plants, and similar trends were observed in repeated experiments.(F) Soil moisture content during drought treatment.Mean values ± SE (n = 3); Significant values *p < 0.05, **p < 0.01; ns, no significance.