Abstract[Objectives]This study was conducted to explore the physiological mechanism of flower color variation in the white mutants of Anthurium andeaeanum.[Methods]The seven white mutants of ‘Alabama’ and ‘Turenza’ were used as materials to analyze the pigment types， flavonoid types and content and anthocyanin content in the wild type and mutants.[Results]The white spathe mainly contained flavonoids， flavonols， dihydroflavonols and dihydroflavonols; the white mutants of ‘Alabama’ had a higher total flavonoid content than the wild type， while the white mutants of ‘Turenza’ showed an opposite trend; and the spathe of the wild type had the highest anthocyanin content， and the pink part of the twocolor mutant or the spathe of the binary color mutant contained trace anthocyanins， while no anthocyanins were detected in the white part of the mutants.[Conclusions]The main cause of the white mutants of A. andeaeanum is related to anthocyanin metabolism.
Key wordsAnthurium andeaeanum; White mutant; Flavonoids; Anthocyanin
Plant color is the resultant of the synergistic effect of various internal and external factors， and is affected by factors such as anthocyanidin type， content and distribution[1]. Plant anthocyanins include carotenoids， flavonoids and alkaloids[2]. Spathes of Anthurium andeaeanum are mainly red， and there are also special varieties which have orange， coral， white， green， brown and twocolor spathes[3]. A. andeaeanum spathes are rich in color and have high ornamental value and economic value. The existing research shows that A. andeaeanum spathes mainly contain flavonoid anthocyanins， green spathes mainly contain chlorophyll， and brown spathes also contain a small amount of carotenoids[4]. Li et al.[3]detected three kinds of anthocyanins in the red spathe of ‘Arizona’ chimera， namely cyanidin 3rutinoside， geranium 3rutinoside and peonidin 3rutinoside， but in the chimeric orangered mutant， only cyanidin3rutinoside and geranium3rutinoside were detected， without any change in the composition of flavonoid glycosides and flavonols. Cong et al.[5]found that there were three kinds of anthocyanins in the red spathe， but no anthocyanins were detected in the white mutant， indicating that the anthocyanin type and content in spathes of these mutants had changed.
In recent years， scholars have screened various types of flower color mutants from the invitro culture materials of A. andeaeanum. In this study， with seven white mutants of ‘Alabama’ and ‘Turenza’as test materials， through the color phenotype and pigment qualitative and quantitative analysis of spathes， the differences in anthocyanidins between the mutants and wild type and between different white mutants were explored， providing an experimental basis for elucidating the physiological mechanism of white mutation in A. andeaeanum. Materials and Methods
Plant materials
The test materials were the A. andeaeanum control varieties （the wild types） and their somatic mutants preserved in the laboratory， including five white mutants of ‘Alabama’ and two white mutants of ‘Turenza’ （Fig. 1）.
Determination of color phenotype
The Royal Horticultural Society Color Card （RHSCC） was used to analyze the flower color of the A. andeaeanum mutants， in which the twocolor mutant was determined after distinguishing the color patches.
Qualitative analysis of pigments in spathes
According to the method of reference[6]， 0.1 g of the tested varieties and mutants were cut into pieces and added into 10 ml centrifuge tubes， respectively. Into each of the centrifuge tube， about 5 ml of petroleum ether， 10% hydrochloric acid aqueous solution and 25% ammonia water were added， followed by mixing and standing. The change in the color of each solution was observed after standing for a certain period of time.
Color reaction of flavonoids
According to the method of the references[6-8]， 0.1 g of the spathe was taken and extracted with 0.1% hydrochloric acidmethanol solution. After dilution， 2 ml of the extract was taken for color reaction.
Determination of flavonoids
According to the method of reference[9]， 1.00 ml of flavonoid extract was accurately pipetted into a 25 ml volumetric flask， and the absorbance was measured at 510 nm. An absorbanceconcentration standard curve was drawn with the concentration of rutin solution as the abscissa and the absorbance as the ordinate. The linear regression equation was calculated as y=8.705 7x-6E-05 （R2=0.999 5）， and the total flavonoid content in the sample was calculated according to the regression equation.
Determination of anthocyanin content in spathes
According to the pHdifferential spectrophotometry of Wrolstad et al.[10]， 25 mmol/L KCl and 400 mmol/L NaAc solutions were adjusted to pH 1.0 and pH 4.5 with HCl， respectively. The anthocyanin extract was diluted with KCl and NaAc solutions by the same times， obtaining two solutions which were determined at 520 and 700 nm for absorbance， respectively. The content of anthocyanins was calculated according to the following formula： Anthocyanin content （mg/g·Fw）=A*V*n*M/（┖*m）， wherein A is the corrected absorbance， i.e.， A=（A520-A700）pH1.0-（A520-A700）pH4.5; V is the total volume of the extract; n is the dilution times of the extract; M is the molecular weight of Cv3glu （449.2）; ┖ is the extinction coefficient of Cv3glu （26 900）; and m is the fresh weight of sample （g）. Results and Analysis
Phenotypic characteristics of white mutants of A. andeaeanum
According to the division standards of the developmental period of A. andeaeanum spathes[11]， the flower color differences were comparatively analyzed at fullexpansion stage （S5 stage） of spathes， and the phenotypic characteristics of the flower color are shown in Fig. 1.
According to the comparison with RHSCC， the wild types ‘Alabama’ （ACK） and ‘Turenza’ （TCK） belong to the "REDGROUP"; the mutants ‘A161010’， ‘A17012’， ‘A16081’ and ‘A16082’ belong to the "WHITEGROUP"， and among them， ‘A16082’ has white spathes， which show extremely light pink in the base color; for the two twocolor mutants， they show the spathes which are white in most part （‘A17014B’， ‘T15111B’） belonging to "WHITEGROUP" and pink in the middle chimeric parts （‘A17014Z’， ‘T15111Z’） belonging to the "REDPURPLE GROUP"; and ‘T15116’ is a binary color mutant with a white color on the front and a spotted pink color on the back. In addition to the difference in flower color， the spathes of the mutants are the same as those of the wild types， heartshaped， leathery， smooth and shiny. There are no significant differences in leaf morphology and color as well.
Qualitative analysis of pigments in the white mutants
According to the methods and standards of references[1，6]， qualitative analysis was carried out on the pigments of the test materials. The results are shown in Table 1. All of the test materials were colorless in the petroleum ether reagent， indicating that they did not contain carotenoids. Secondly， in the 10% hydrochloric acid solution， the control varieties ‘ACK’ and ‘TCK’ and mutants ‘A17014Z’， ‘T15116’ and ‘T15111Z’ all showed different degrees of red， indicating that they contained anthocyanins; and ‘A161010’， ‘A17012’， ‘A16081’， ‘A16082’， ‘A17014B’ and ‘T15111B’ exhibited different degrees of yellow， indicating that they contained flavonoid compounds. In addition， in the 25% ammonia reagent， some were yellow， indicating that they contained flavonoids; and some were orange yellow， indicating that they contained flavonols.
Chengchen SHEN et al. Preliminary Study on the Mechanism of Flower Color Variation in White Mutants of Anthurium andeaeanum
Color reaction of flavonoids from the while mutants of A. andeaeanum
On the basis of determining the flavonoid pigments contained in spathes， the color reaction analysis for flavonoid types was carried out by referring to the methods in references[6-8，11-12]. The results are shown in Table 2. ↘ In the reactions of hydrochloric acidmagnesium powder and hydrochloric acidzinc powder， the test materials all showed different degrees of pinkamaranth， indicating that they might contain flavonoids， flavonols， dihydroflavonoids and dihydroflavonols， but ‘A17014B’ was colorless， indicating that it might not contain flavonoids. ? In the metalsalt complexation reaction， both the flavonoid compounds and the lead acetate formed a precipitate， indicating that they contained an odiphenol hydroxyl group， or 3OH， 4keto or 5OH as well; in the ferric chloride reaction， all of them had different degrees of yellow， indicating that they contained phenolic hydroxyl structure; in the aluminum trichloride reaction， all of them were yellow except ‘A17014B’， indicating that they contained an odiphenol hydroxyl structure; and in the strontium chloride reaction， a precipitate was formed except ‘A17014B’， indicating that flavonoid compounds with an odiphenol hydroxy structure were contained. ? In the alkaline reagent （sodium carbonate） reaction， all of them showed different degrees of yellow， indicating that they contained flavonoids， dihydroflavonoids or dihydroflavonols. ? In the boric acid reaction， no bright yellow product was detected in both the wild types and the mutants， indicating that 5hydroxyflavone and 2▽hydroxychalcone were not contained. ? In the concentrated sulfuric acid reaction， all materials were brown to different degrees， indicating that flavonoids and flavonols might be contained.
Based on the above results， it could be considered that the wild types and white mutants of A. andeaeanum might contain flavonoids， flavonols， dihydroflavonoids and dihydroflavonols， and have phenol hydroxyl group， 3OH， 5OH， 4keto group， odiphenol hydroxyl group， and both odiphenol hydroxyl group and 4keto group structure; and the white part of ‘A17014B’ spathe did not contained an odiphenol hydroxy structure.
Analysis of total flavonoid content in while mutants of A. andeaeanum
The total flavonoid contents of A. andeaeanum spathes are shown in Table 3. In the white mutants of ‘Alabama’， the total flavonoid contents of the spathes were all lower than that of the control wild type， and among them， the mutants ‘A16082 ’ and ‘A161010’ showed the lowest contents， followed by ‘A17012’ and ‘A16081’， and the differences between them and the wild type reached a significant level. However， there were no significant differences between the twocolor mutant and the wild type， and between different color patches （‘A17014Z’ and ‘A17014B’）. In contrast， the total flavonoid contents of the ‘Turenza’ white mutants were higher than that of the wild type， and the total flavonoid content of the white part of the twocolor mutant （‘T15111B’） was higher than that of the pink part （‘T15111Z’）. The comprehensive results showed that there was a significant difference in the total flavonoid content between the wild type and the mutant， but the changing trends of the two tested varieties were opposite， and different trends were also observed between the different color patches of the twocolor mutants， indicating that the difference in total flavonoid content was not the main factor affecting the color change of spathes.
Analysis of anthocyanin content in while mutants of A. andeaeanum
The anthocyanin contents of the test materials are shown in Table 3. The differences in the anthocyanin content between the wild types and the white mutants of A. andeaeanum and between the different mutants were very significant. Among them， the wild types were the highest in the anthocyanin content of spathes， while no anthocyanins were detected in the white mutants or the content was very low in the white mutants; and in the twocolor mutants， the pink parts of the spathes （A17014Z， T15111Z） had an anthocyanin content significantly higher than that of the white parts （A17014B， T15111B）. In addition， the binary color mutant of ‘Turenza’， ‘T15116’， was also detected to contain anthocyanins， but the content was much lower than the wild type as the control. From this， it was judged that the difference in anthocyanin content is the main factor causing the white variation of A. andeaeanum spathes.
Discussion
Flower color is an important economic trait of ornamental plants. It is related to various factors such as pigment types， pigment contents and physical properties caused by internal or surface texture of petals， while anthocyanins play a major role[13]. Anthocyanins in flavonoids can reflect a variety of colors such as red， blue， purple and reddish purple[14]. In this study， the qualitative analysis of anthocyanidins showed that the flower pigments in A. andeaeanum spathes were mainly flavonoid pigments. The color reactions of flavonoids showed that A. andeaeanum spathes mainly contained anthocyanins， flavonoids， flavonols， dihydroflavonoids and dihydroflavonols. Compared with the wild type， the white spathes also contained flavonoids， but the hydroxyl groups or phenol hydroxyl groups contained were different. For instance， the white part of the spathe of the twocolor mutant （A17014B） did not contain an odiphenol hydroxyl structure. It has been clarified that flavonoid type， content and distribution determine the performance of flower color[15]. However， in this study， the white mutants of ‘Alabama’ had a higher total flavonoid content compared with spathes of the wild type， while the white mutants of ‘Turenza’ showed an opposite trend， indicating that no definite correlation was detected between flavonoid type and content and the change in the color of spathes.
Yang et al.[4]found that different A. andeaeanum varieties differed significantly in anthocyanin content. Specifically， the red and pink spathes and the red or pink parts of the variegated spathes contained more anthocyanins， while the white spathes and the white parts of the variegated spathes contained very little anthocyanins， which is similar to the quantitative analysis of total anthocyanins in this study. In the present study， the two tested wild types had the highest anthocyanin contents， and the pink parts of the twocolor mutants or the spathes of the binary color mutant also contained trace anthocyanins， while no anthocyanins were detected in the white parts. And the total anthocyanin content showed significant correlation with the color of spathes， indicating that the main cause of the white mutants of A. andeaeanum is related to anthocyanin metabolism.
References
[1]AN TQ. Physiological biochemistry of flower color[M]. Beijing： China Forestry Publishing House， 1989.
[2]FENG H， YI SL， ZUO JQ， et al. Progress of flower color modification by genetic engineering and anthocyanin biosynthesis pathway[J]. Genomics and Applied Biology， 2014， 33（2）： 445-451.
[3]LI CH， NIU JH， HUANG SR， et al. Preliminary studies on partial mutation in color of Anthurium andraeanum ‘Arizona’ spathe[J]. Molecular Plant Breeding， 2015， 13（3）： 235-235.
[4]YANG L. Relationships between the composition and content of anthocyanidins and the expression of key genes and color formation in spathes of Anthurium andraeanum Linden[D]. Haikou： University of Hainan， 2014： 24-32.
[5]CONG HQ， XU L， XIN CY， et al. Comparison and determination of anthocyanidins in Anthurium mutant[J]. Journal of Anhui Agricultural Sciences， 2009， 37（17）： 7832-7834.
[6]ZHAO CL， GUO WM， CHEN JY. Preliminary study on the categories and contents of the flower color pigments of Prunus mume Sieb. Et Zucc[J]. Journal of Beijing Forestry University， 2004， 26（2）： 68-73.
[7]BAI CX， HU K， DAI SL， et al. Components of flower pigments in the petals of different color Chrysanthemum morifolium Ramat. cultivars[J]. Journal of Beijing Forestry University， 2006， 28（5）： 84-89.
[8]SU M. Study on biological characteristics and formation mechanism of flower color chimera in chrysanthemum[D]. Suzhou： Suzhou University， 2011.
[9]YAN JF. Study on pigment metabolism and variation mechanism of flower color chimera in chrysanthemum[D]. Suzhou： Suzhou University， 2014： 26-28.
[10]WROLSTAD RE， DURST RW， LEE J. Tracking color and pigment changes in anthocyanin products[J]. Trends in Food Science & Technology， 2005， 16（9）： 423-428.
[11]PENG JJ. Sequencing of Anthurium andraeanum transcriptome and analysis of differentially expressed genes in anthocyanin biosynthesis pathway[D]. Yinchuan： Ningxia University， 2015.
[12]GAO JM. Phytochemistry[M]. Beijing： Science Press， 2012.
[13]ZHAO YP， CHEN FD， GUO WM. Advances in genetic engineering of flower color of ornamenal plants[J]. Bulletin of Botany， 2003， 20（1）： 51-58.
[14]TANAKA Y， TSUDA S， KUSUMI T. Metabolic engineering to modify flower color[J]. Plant & Cell Physiology， 1998， 39（11）： 1119-1126.
[15]GENG J. Study on anthocyanin metabolic pathways and products of ornamental Malus spectabilis[D]. Chongqing： Southwest University， 2011.