Plant Regeneration of Picea Asperata Mast. by Somatic Embryogenesis

Trees (2017) 31:299–312 DOI 10.1007/s00468-016-1484-4

ORIGINAL ARTICLE

Plant regeneration of Picea asperata Mast. by somatic embryogenesis

1 1 1 2 Yan Xia • Jianwei Zhang • Danlong Jing • Lisheng Kong • 1 1 Shougong Zhang • Junhui Wang

Received: 28 April 2016 / Accepted: 19 October 2016 / Published online: 31 October 2016 Ó Springer-Verlag Berlin Heidelberg 2016

Abstract MLV supplemented with 105 lM(±)-abscisic acid, 5% Key message This study provides a novel desiccation polyethylene glycol 4000, 3% sucrose, 0.1% activated indicator for somatic embryos and an efficient system charcoal (AC), 0.05% L-glutamine, 0.1% casein acid for somatic seedling production in Picea asperata Mast. hydrolysate, and 0.4% gellan gum. In this medium, Abstract Picea asperata Mast. is a widely distributed 415 ± 103 SEs with well-developed cotyledons were native spruce in China with outstanding wood properties obtained per gram (fresh weight) of EM. After partial and adaptability. In this study, we established an effective desiccation treatment under low light intensity, somatic system to propagate plantlets through somatic embryoge- embryos, depending on their color change, were divided nesis in P. asperata. The effects of cone collection date, into green cotyledonary embryos (GCEs) and non-green seed family, and plant growth regulators on the initiation of cotyledonary embryos (NGCEs). These embryos, along embryonal masses (EM) were analyzed. Besides the sig- with non-desiccated SEs as controls, were germinated on nificant effects of seed family and cone collection date on 1/2 MLV supplemented with 2% sucrose, 0.2% AC, 0.05% EM initiation, the highest induction frequency was L-glutamine, and 0.6% gellan gum. Only GCEs, with red obtained with modified Litvay medium (MLV) containing radicles, produced plantlets at the highest rates ([74.5%) 10 lM 2,4-dichlorophenoxyacetic acid and 5 lM 6-ben- among all of the treatments. This efficient system of zylaminopurine. The best tissue proliferation in suspension somatic seedling production in P. asperata includes cultures was obtained with an inoculum density of 1:7 (v/v, screened culture factors and a novel desiccation indicator sedimented cells:liquid medium) after 9 days in culture. to ensure SE quality before germination. The best medium for somatic embryo (SE) maturation was Keywords Spruce Á Embryonal masses Á Somatic embryo maturation Á Somatic embryo germination Á Partial Communicated by K. Klimaszewska. desiccation Á Indicator

Y. Xia and J. Zhang contributed equally to this work. Abbreviations Electronic supplementary material The online version of this 2,4-D 2,4-Dichlorophenoxyacetic acid article (doi:10.1007/s00468-016-1484-4) contains supplementary BAP 6-Benzylaminopurine material, which is available to authorized users. ABA Abscisic acid & Junhui Wang AC Activated charcoal [email protected] PEG 4000 Polyethylene glycol 4000

1 SE Somatic embryo State Key Laboratory of Tree Genetics and Breeding, Key EM Embryonal masses Laboratory of Tree Breeding and Cultivation of State Forestry Administration, Research Institute of Forestry, ECL Embryogenic cell line Chinese Academy of Forestry, Beijing, China GCE Green cotyledonary embryo 2 Department of Biology, Centre for Forest Biology, University NGCE Non-green cotyledonary embryo of Victoria, Victoria, BC, Canada FW Fresh weight 123 300 Trees (2017) 31:299–312

MLV Modified Litvay medium conditions for embryo germination (Stasolla and Yeung ANOVA Analysis of variance 2003). Some researchers have subjected SEs to partial desiccation to reach physiological maturity (Liao and Juan 2015; Stasolla and Yeung 2003). Desiccation con- tributes to a significant increase in germination and subsequent conversion into seedlings (reviewed by Kli- Introduction maszewska et al. 2016; Lelu-Walter et al. 2013; Stasolla et al. 2002). Long generation times and poor vegetative propagation The methods of SE desiccation in conifers can be make programs for the genetic improvement of coniferous classified as partial or full desiccation, depending on the species much more difficult than in broadleaf tree species. rate of water loss (Hay and Charest 1999). That partial However, modern biotechnology has provided new tools in desiccation significantly enhances the germination fre- addition to conventional plant breeding for conifer quency of spruce SEs was first reported by Roberts et al. improvement (Garcıá-Mendiguren et al. 2016; Park et al. (1990). To date, stimulating SE germination using partial 1998). Among these tools, somatic embryogenesis is of drying has been recorded in many species, including Picea great value, since an unlimited number of genetically rubens (Harry and Thorpe 1991), Picea glauca (Attree identical embryos can be produced (Grossnickle 2011; et al. 1991; Kong and Yeung 1992, 1995), Abies cepha- Klimaszewska et al. 2007; Park 2002). In general, the lonica (Salajova and Salaj 2001; Vookova´ and Kormut’a´k complete process of somatic embryogenesis starts with the 2006), pines (Pinus thunberghii, Pinus densiflora and Pi- initiation of embryonal masses (EM), followed by prolif- nus armandii var. amamiana) (Maruyama and Hosoi eration of EM, maturation of somatic embryos (SEs), and 2012), Picea morrisonicola (Liao and Juan 2015), P. abies plant regeneration. The first descriptions of somatic (Bozhkov and von Arnold 1998) and Picea omorika embryogenesis from immature zygotic embryos of conifers (Hazubska-Przybył et al. 2015). were reported in Picea abies (Chalupa 1985; Hakman and However, no desiccation method is generally applicable von Arnold 1985). Somatic embryogenesis is becoming the to every SE, because different species or even different most efficient method for cloning genetically improved genotypes of the same species require different treatments spruces (Klimaszewska et al. 2016). to maximize germination. Loss of water content by SEs Plant regeneration through somatic embryogenesis has during partial desiccation can be affected by moisture in been achieved in several Picea species, but to our knowl- the air and the duration of desiccation. In previous reports, edge, it has not been reported in P. asperata Mast., which the range of partial desiccation times varied widely, from 8 is one of the most important native to China species, also to 24 days in P. abies (von Arnold and Clapham 2008). known as dragon spruce (Kru¨ssmann 1983; Jing et al. Nevertheless, no study has proposed an effective indicator 2016). Because of its good wood and strong environmental of whether or not the desiccation is appropriate to stimulate adaptability, it is a primary species for afforestation in germination. The goals of this study were to find such an barren regions. However, its current propagation is mainly indicator and to determine major factors in each stage of dependent on seed sowing (Luo et al. 2001) and root cut- somatic embryogenesis. tings (Li et al. 2007), and these methods are not sufficient to meet the increasing demand. Seed production in P. asperata is lower than in most other conifers (Lin and Hu Materials and methods 2000). In addition, cuttings taken from mother trees are affected by age and position, which can markedly reduce Plant materials the quality of clones. Instead, the propagation of P. asperata by somatic embryogenesis has the potential to Immature female cones of P. asperata were collected from generate a stable supply of somatic seedlings for propa- open-pollinated elite mother trees at the National Spruce gating superior genotypes. Germplasm Bank in Gansu Province, China. Nine mother In some conifer species, the proper germination of SEs trees were chosen for weekly cone collection starting from is a crucial step for the transfer of plants from in vitro 29 May until 27 July, 2012. conditions to the natural environment. However, SEs are All cones were surface-sterilized by 10-min immersion sometimes inadequate for yielding viable seedlings able in 70% ethanol and then rinsed three times with sterile to acclimatize to the greenhouse (Hazubska-Przybył et al. distilled water under sterile conditions on a laminar flow 2015). The germination capacity of conifer SEs margin- bench. All seeds were removed aseptically from the cones. ally depends on genetic factors (Park 2002; Park et al. Immature zygotic embryos were then dissected out from 1998), so particular attention must be paid to the culture the megagametophyte under a Leica S6E stereomicroscope 123 Trees (2017) 31:299–312 301

(Leica Microsystems GmbH, Wetzlar, Germany) and containing 80 ml medium (same as the semi-solid main- placed horizontally on initiation medium. tenance medium without gellan gum). The flasks with EM were then placed on a gyratory shaker (110 rpm) and Initiation of embryonal masses cultured in darkness at 24 ± 1 °C. Five milliliters of well- growing suspension cultures (sedimented cells) were From previous studies (Hakman and von Arnold 1985; Ishii transferred to 80 ml fresh medium every 10 days. Stabi- 1995; Klimaszewska et al. 2011), the initiation medium lized cell lines were established after three subcultures. was based on modified Litvay medium (MLV) (Litvay After 30 days of culture in liquid medium, the volume of et al. 1985). In addition, this medium was supplemented sedimented cells was evaluated in each line. with 1% sucrose (Beijing Chemical Works, Beijing, The ECL 1931 suspension culture, which had a China), 0.1% casein enzymatic hydrolysate (Sigma, St. stable physiology (Table S1) and an undiminished ability Louis, MO, USA), 10 lM 2,4-dichlorophenoxyacetic acid to proliferate EM, was used for testing the effects of (2,4-D, Sigma), and 5 lM 6-benzylaminopurine (BAP, inoculum density (volumes of sedimented cells:liquid Sigma). Before autoclaving, the pH of the medium was medium = 1:5, 1:7, 1:9, 1:12, and 1:15) on the cell sus- adjusted to 5.8 ± 0.01 and 0.2% gellan gum (PhytagelTM, pension growth. Cellular growth was assayed by FW of Sigma) was added. After autoclaving at 121 °C for 20 min, 5-ml suspension culture and evaluated every 3 days from 0.05% L-glutamine (Sigma) was filter-sterilized and added day 0. Before weighting, the suspension culture was to the medium. Nine cone collection dates and nine open- pipetted onto filter paper disk to remove the liquid medium pollinated seed families (mother trees) were arranged in a in Bu¨chner funnel with a vacuum pump. full factorial design. Ten embryos were cultured in a Petri dish (90 9 15 mm) containing *30 ml semi-solid med- Maturation of somatic embryos ium and incubated in darkness at 24 ± 1 °C. A total of 2430 embryos were excised, with three Petri dishes per Eleven healthy-looking ECLs, at least one per seed family, treatment. All Petri dishes were randomly distributed on were used to assess the effect of genotype on maturation. the shelves of a growth chamber. After 5–7 weeks, the All the ECLs had been grown on semi-solid culture med- number of explants producing EM was recorded and the ium for several weeks. In order to stimulate embryo mat- initiation rate was calculated with the formula: (EM-pro- uration, 2,4-D and BAP were eliminated, while (±)- ducing explants/total explants) 9 100%. abscisic acid (ABA, Gibco-BRL, Gaithersburg, MD, Simultaneously, nine initiation media with various USA), polyethylene glycol 4000 (PEG 4000, Merck, combinations of 2,4-D (5, 10, or 15 lM) and BAP (2.5, 5, Darmstadt, Germany), and activated charcoal (AC, Sigma) or 10 lM) were tested in a full factorial design. Cones from were added. Furthermore, the concentrations of sucrose and three trees (1, 2, and 3) were collected on 9 collection dates gellan gum were changed. The maturation medium con- as described previously. Cones were stored up to 1 month sisted of MLV salts supplemented with 0.05% L-glutamine, in three layers of paper bags at 4 °C. Ten embryos per Petri 0.1% casein acid hydrolysate (Sigma), 60 lM ABA, 5% dish were cultured in darkness at 24 ± 1 °C. Each treat- PEG 4000, 6% sucrose, 0.3% AC, and 0.6% gellan gum. ment included three Petri dishes per seed family, per date, Five milliliters samples of suspension cultures of EM with a total of 7290 embryos in the test. (*0.2 g FW) were collected on a disk of sterile filter paper (WhatmanTM# 2), using a Bu¨chner funnel with a vacuum Proliferation and maintenance of embryonal masses pump, and each was washed with 10 ml of liquid MLV without PGRs. Finally, the filter paper with the attached Eight weeks from the beginning of EM induction experi- EM was placed on the 30 ml maturation medium in a Petri ments, proliferated EM were separated from explants and dish. After 7 weeks of culture at 24 ± 1 °C in the dark, placed on a fresh medium of the same composition. The numbers of cotyledonary embryos were counted for each culture conditions for EM proliferation were the same as treatment. The effects on embryo maturation were evalu- for the initiation cultures. Proliferating EM were subdi- ated by the number of cotyledonary somatic embryos per vided into small pieces [0.8–1.0 cm in diameter, *0.25 g gram (FW) EM. fresh weight (FW)] and sub-cultured on fresh maintenance Three ECLs (1113, 2711, and 1931), which had the medium biweekly. Necrotic tissue was removed at each highest maturation in the prior test, were selected for sub-culture. studies of the maturation medium. This medium consisted To bulk up the EM of different embryogenic cell lines of MLV supplemented with 0.05% L-glutamine, 0.1% (ECLs), EM of each cell line were transferred into liquid casein acid hydrolysate, and various combinations of ABA medium for suspension culture. Five to seven small clumps (30, 45, 60, or 90 lM), PEG 4000 (0, 2.5, 5, or 7.5%), of EM (*3 g FW) were put into 250-ml Erlenmeyer flasks sucrose (1, 3, 6, or 9%), AC (0, 0.1, 0.3, or 0.5%), and 123 302 Trees (2017) 31:299–312

Table 1 ANOVA for embryonal masses (EM) initiation (%) in ab a ab b ab ab ab b ab P. asperata from nine open-pollinated trees at nine collection dates corresponding to embryo developmental stage 22.5 33.1 32.6 20.7 24.7 21.9 24.6 21.0 27.1 ± ± ± ± ± ± ± ± ± Source of variation df EM initiation (%) 27.4 42.2 38.1 23.0 29.6 27.4 35.1 24.1 35.1 Mean Mean square F test p value mn mn mn mn mn d n n n n e 11.5 5.8 5.8 7.0 Collection date (C) 8 1.591 160.950 \0.001 10.0 10.0 1 0.0 0.0 0.0 0.0 ± ± ± ± ± ±

Seed family (S) 8 0.118 11.906 \0.001 ± ± ± ± 6.7 6.7 6.7 4.4 0.0 10.0 10.0 0.0 0.0 0.0 C 9 S 64 0.019 1.879 \0.05 24 Jul Figure Error 162 0.010 mn mn mn n mn n n n mn d e 5.8 5.8 5.8 5.8 5.8 11.5 11.5 5.8 7.0 10.0 1 ± ± ± ± ± ± ± ± ± 5 ± gellan gum (0.2, 0.4, 0.6, or 0.8%) in an L16 (4 ) orthog- 3.3 6.7 3.3 3.3 3.3 6.7 6.7 6.7 5.6 10.0 onal design. Based on the results of the above experiments, 17 Jul Figure mn

in order to ﬁnd the best concentration of ABA during l–n l–n l–n mn n mn mn mn d

maturation, ECL 1931, the most proliferative line with a e 5.8 5.8 5.8 5.8 11.5 5.8 7.6 5.8 5.8 5.8 1 ± ± ± ± ± ± ± stable ability to produce mature embryos, was used to test ± ± ±

the different concentrations of ABA in a completely ran- 6.7 3.3 6.7 6.7 6.7 6.7 9.6 16.7 16.7 16.7 domized design. The tested medium consisted of MLV 10 Jul Figure d, k–n 1 supplemented with 5% PEG 4000, 3% sucrose, 0.1% AC, k–n i–l f–i c l–n 0.4% gellan gum, 0.1% casein acid hydrolysate, and var- e–h h–k j–m h–k 10.0 5.8 10.0 5.8 5.8 5.8 5.8 10.0 15.3 15.2 0.05 (Duncan’s multiple range test)

ious concentrations of ABA (75, 90, 97.5, 105, and ± ± ± ± ± ± ± ± ± ± \

120 lM). The culture conditions were as described above. p 20.0 56.7 30.0 13.3 36.7 26.7 36.7 20.0 46.7 31.9 e 3 Jul Figure

Partial desiccation of somatic embryos g–j g–j b–e d–g e–h e–h b e–h f–i e–h d, e 11.5 17.3 5.8 5.8 10.0 10.0 20.8 5.8 15.3 13.5 Six genotypes of mature SEs (1113, 2614, 2711, 1911, 1 1921, and 1931) were partially desiccated. Groups of 30 ± ± ± ± ± ± ± ± ± ± 43.3 70.0 56.7 46.7 60.0 40.0 53.3 53.3 53.3 53.0 embryos were transferred onto two layers of dry sterile 26 Jun Figure ﬁlter paper (Whatman TM 2) in plastic Petri dishes e–h b–d f–h b-e a g–j (35 9 12 mm) without lids. Subsequently, the dishes with a–c e–h e–h c–f d 11.5 15.3 0.0 10.0 5.8 5.8 5.8 0.0 10.0 14.3 embryos were placed in larger Petri dishes (90 9 15 mm) 1 ± ± ± ± ± ± ± ± ± ± containing 10 ml sterile deionized water. The large Petri 53.3 73.3 80.0 50.0 53.3 56.7 66.7 40.0 70.0 60.4 dishes were covered with lids and incubated at 24 ± 1 °C 19 Jun Figure c,

with a 16-h photoperiod at low light intensity (*15 lmol/ from nine open-pollinated trees at nine collection dates corresponding to various stages of embryo development 1

2 g–j e–h ab g–j e–h f–h d–g d–g a

m /s, LED ﬂuorescent tubes) for 2 weeks. a 11.5 5.8 11.5 10.0 15.3 10.0 10.0 20.8 26.5 21.5 asperata ± ± ± ± ± ± ± ± ± ±

Embryo germination and conversion . P 53.3 93.3 86.7 40.0 53.3 50.0 60.0 43.3 60.0 60.0 d 12 Jun Figure Somatic embryos with well-developed cotyledons were b, 1 g–j selected and placed horizontally on the surface of 1/2 MLV f–i f–i d–g g–j g–j f–i e–h f–h b germination medium containing 0.6% gellan gum, 2% 15.3 15.3 10.0 10.0 15.3 15.3 15.3 10.0 10.0 12.6

L ± ± ± ± ± ± ± ± ± ± sucrose, 0.2% AC, and 0.05% -glutamine. The growth of SD. Values indicated by the same letter are not signiﬁcantly different at all SEs in this medium was allowed to proceed at ± 46.7 46.7 60.0 40.0 43.3 46.7 53.3 40.0 50.0 47.4 5 Jun Figure c 24 ± 1 °C under a 16-h photoperiod of low light intensity 2

(18–20 lmol/m /s LED ﬂuorescent tubes) for 7–14 days. mn lmn lmn mn n mn lmn mn lmn d

When the root elongated to *1 cm, the germinants were a 5.3 5.8 5.8 5.8 5.8 4.7 5.8 4.7 6.5 10.0 1 ± ± ± ± ± ± ± ± ± subcultured onto fresh medium as described above without ± AC, and the light intensity was increased to 50 lmol/m2/s. After 4–6 weeks on germination medium, the plants were EM initiation (%) Collection date 29 May Figure transplanted into forest containers ﬁlled with substrate Embryonal masses (EM) initiation (%) of (volume of peat:perlite:roseite = 3:1:1) and cultured in a Table 2 2345 16.1 6 13.3 7 6.7 8 3.3 9 6.7 Mean 12.7 The 6.7 data are expressed 12.7 as 9.8 mean Seed family chamber with the same light intensity for more than 1 10.0 123 Trees (2017) 31:299–312 303

1 month. Then the somatic seedlings were acclimatized in times. All collected data were analyzed with ANOVA a greenhouse. The germination rate was calculated by using SPSS 16.0. The means were compared by Duncan’s (germinated somatic embryos/total cotyledonary somatic Multiple Range Test (95% confidence level), and the embryos) 9 100% before transplantation into the forest standard deviations of the means were calculated. containers. The acclimatization rate was calculated as (acclimatized germinants/total transplanted germinants) 9 100% after culture in the greenhouse for Results 2 months. Embryonal masses initiation Histocytological and morphological changes The initiation of EM was affected by cone collection dates Thirty megagametophytes per seed family at each collec- and seed families; a significant interaction was found tion date were carefully dissected and photographed under among collection dates and mother trees (Table 1). We a fully apochromatic corrected stereo microscope (Leica classified the development stages of the explants (Table 2; M205 FA). Tissues were sampled from EM initiation cul- Fig. 1) and found that, on 29 May, the suspensor of zygotic tures after 8 weeks. Each sample was double-stained with embryos had elongated and the embryo had established acetocarmine and Evans blue to distinguish EM from non- polarity (Fig. 1a), and then the embryo entered the pre- embryonal masses (Gupta and Holmstrom 2005). Stained cotyledonary stage and became white and opaque on 5 June tissues were subsequently examined and photographed (Fig. 1b); on 12 June, the embryo had elongating cotyle- under a microscope (Leica DM6000 B). The morphology dons (Fig. 1c, d); and on 10 July, completely developed of somatic embryogenesis at each stage was examined and embryos reached morphological maturity with curved photographed using the stereo microscope. Embryos and cotyledons (Fig. 1e). The initiation percentages from 12 germinants after 9 days of germination were fixed in FAA and 19 June were significantly higher (both at 60%) than (volumes of 38% formaldehyde:acetic acid:70% etha- from the other collection dates (Table 2). These dates nol = 1:1:18), and then dehydrated in a graded aqueous corresponded to the mid-developmental stage of the ethanol series (70, 85, 95, and 100%), and embedded in cotyledonary embryo (Fig. 1c, d). The EM initiation paraffin. Each sample was sectioned serially at 6 lmona showed an increasing trend along with embryo develop- microtome (Leica RM2016). Subsequently, sections were ment up to 12 June, but gradually decreased as the zygotic stained with safranin and fast green using the method embryo matured after 19 June (Table 2). described by Johansen (1940), and then examined and Taking seed families into account, the best results for photographed by the digital scanner (Pannoramic 250 EM initiation were obtained from family 2 (42.2%), while Flash, 3DHISTECH Ltd., Budapest, Hungary). the worst were from families 4 (23.0%) and 8 (24.1%) (Table 2). The other families had initiation percentages Statistical analysis from 24.1 to 38.1%, and there was no significant difference among them. Meanwhile, the highest initiation percentage The experiments on EM initiation were three replicates (93.3%) was from family 2 collected on 12 June. When the (one dish with 10 embryos for each replicate) and repeated interaction among collection dates and seed families was twice. The rest of the experiments in this study were considered, about half of the families achieved their peak designed with at least six replicates and repeated three initiation on 19 June (Table 2; Fig. 1d). However, families

Fig. 1 Five development stages of explants. a Dominant embryo developed embryo with elongating cotyledon. The ﬁgure is represen- with established polarity. b Pre-cotyledonary embryo with shoot tative of at least 20 whole-mount explants from each collection date apical meristem. c, d Mid-stage cotyledonary embryos. e Completely (Bars 250 lm) 123 304 Trees (2017) 31:299–312

Fig. 2 Initiation and proliferation of embryonal masses. a, b Non- whole-mounts of embryonal or non-embryonal masses. e, f Embryonal embryonal masses induced from the suspensor and radicle ends. c, masses proliferation on solid and liquid media. (Bars a, c, 2 mm; b, d, d Embryonal masses initiated from the surface of explants; arrows 500 lm; e, f, 1 cm) indicate the embryo proper. The ﬁgure is representative of at least 20

2, 3, 5, and 8 had the highest initiation of EM on 12 or 26 embryogenic competence after placement on initiation June, whereas family 1 had equal initiations on 12 and 19 medium for 2–4 days. Callus was ﬁrst induced from the June (Table 2). radicle ends (Fig. 2a); most of it was non-embryonal Besides the explants without cotyledon primordia (Fig. 2b). EM started to appear around the upper half of (Fig. 1a), cells in immature zygotic embryos acquired zygotic embryos (Fig. 2c) after initiation for 7–10 days,

123 Trees (2017) 31:299–312 305

Table 3 Effect of 2,4-D and BAP on embryonal masses (EM) initi- ab a ab b ab ation in P. asperata 0.28 0.33 0.29 0.26 0.41

Initiation media PGRs (lM) EM initiation (%) ± ± ± ± ±

2,4-D BAP 0.68 0.77 0.65 0.53 0.65

1 5 2.5 10.8 ± 10.5d bc 2 5 5 20.5 ± 11.9 e–m f–m e–l h–m d–l c cd

3 5 10 14.8 ± 10.0 0.16 0.09 0.10 0.09 0.19 0.13 bcd 4 10 2.5 18.6 ± 11.2 ± ± ± ± ± ± 5 10 5 30.6 ± 18.2a 0.52 0.50 0.55 0.40 0.60 0.51 6 10 10 25.1 ± 14.0ab 7 15 2.5 23.3 ± 13.0abc

8 15 5 24.8 ± 14.6ab c–i a–d b–e d–l a–c b bc 9 15 10 21.0 ± 11.0 0.21 0.26 0.25 0.14 0.32 0.26 ± ± ± ± ± ± The data are expressed as mean ± SD. Values indicated by the same letter are not signiﬁcantly different at p \ 0.05 (Duncan’s multiple 0.71 0.92 0.87 0.60 1.01 0.82 range test) e–m b–g b–g d–k ab b Table 4 Embryogenic cell lines (ECLs) initiated and proliferated from nine seed families of P. asperata cultured on solid and liquid 0.16 0.13 0.06 0.15 0.36 0.27 medium ± ± ± ± ± ± 0.05 (Duncan’s multiple range test) 0.52 0.80 0.78 0.63 1.11 0.77 Seed Number of Frequencies of proliferated ECLs \ p family initiated ECLs Solid medium (%) Liquid medium (%) a–c a a–c b–f a–d a 174209 2 114 10 4 0.31 0.20 0.16 0.19 0.30 0.25 ± ± ± ± ± ± 3 103 11 7

462105 1.01 1.23 1.02 0.81 0.95 1.01 580113 674195 a–d a–c c–i c–h f–m b 795117 0.29 0.16 0.04 0.26 0.20 0.26

865182 ± ± ± ± ± ± 995149 0.95 1.00 0.80 0.74 0.48 0.78 Total 762 13 6

and thereafter kept proliferating. After initiation for d–l c–j j–m i–m lm c

5–7 weeks, EM were transferred to fresh proliferation 0.22 0.07 0.02 0.15 0.07 0.20 medium. EM were white and glossy (Fig. 2c, d), while ± ± ± ± ± ±

non-embryonal masses were green or brown (Fig. 2a, b). 0.61 0.68 0.33 0.37 0.25 0.44 ANOVA analysis revealed that both 2,4-D and BAP had signiﬁcant effects on the EM initiation percentage, while no interaction was found (Table S2). The highest per- g–m k–m k–m n n d SD. Values indicated by the same letter are not signiﬁcantly different at 0.09 0.02 0.06 0.03 0.12 0.06 centage of EM initiation was 30.6%, which occurred on ± ± ± ± ± ± medium supplemented with a combination of 10 lM 2,4-D ± Proliferation days and 5 lM BAP (Table 3). 0 day 3 days 6 days 9 days 12 days 15 days 18 days Mean

Proliferation and maintenance of embryonal masses

A total of 101 stable ECLs, 13.3% of the lines initiated, were established after 3 months with bi-weekly sub-culture Embryonal masses (EM) proliferation of ECL 1931 from ﬁve densities of inoculum at 3-day intervals onto semi-solid medium (Table 4; Fig. 2e). In the liquid Table 5 Inoculation density EM (g FW) of 5-ml suspension cultures 1:71:91:121:15MeanThe data are expressed 0.29 as mean 0.28 0.17 0.16 0.27 medium, some ECLs failed to proliferate or showed a 1:5 0.44

123 306 Trees (2017) 31:299–312

Fig. 3 Somatic embryo maturation in P. asperata. a Embryonal embryos matured for 5 and 7 weeks, respectively. The figure is masses transferred to maturation medium on day 0. b, c, d Embryos representative of at least 20 replicates at each stage. (Bars a, h, 1 cm; matured for 4, 7 and 14 days, respectively. e, f Pre-cotyledonary b–f, 250 lm; g, 500 lm) embryos matured for 3 and 4 weeks, respectively. g, h Cotyledonary minimal increase in sedimented cell volume. Forty-four Embryo maturation ECLs, 5.8% of the lines initiated, were successfully proliferated after 30 days of culture in liquid medium Immature SEs appeared successively on maturation med- (Table 4; Fig. 2f). The proportion of ECLs effectively ium after 4–7 days (Fig. 3a, b). Thereafter, the embryo established on semi-solid and in liquid medium were both started to elongate at both the shoot and radicle ends seed family-dependent: from 10 to 20%, and from 2 to 9%, (Fig. 3c, d) and continued the growth with cotyledon pri- respectively (Table 4). mordia developed after 5–7 weeks (Fig. 3e–h). Growth in liquid medium was significantly influenced All 11 ECLs generated mature SEs, producing between by initial cell inoculum density (Table S3). The best pro- 3 and 295 embryos per gram (FW) of EM, however, the SE liferation was obtained at the density of 1:7 (volume of numbers varied significantly among cell lines. Notably, sedimented cells to liquid medium, Table 5). The growth 1113, 2711, and 1931 were more productive (275, 295, and was also affected by proliferation days (Table S3), and the 247 SEs per gram of EM, respectively) than the rest of the fresh weight of EM reached the highest level on day 9 lines (Fig. 4). In the second experiment, ANOVA analysis (Table 5). showed that PEG 4000, sucrose, AC, gellan gum, and ABA

123 Trees (2017) 31:299–312 307

Table 6 Effect of ABA on somatic embryo maturation in P. asperata ABA (lM) Cotyledonary embryos (no./g FW EM)

75 189 ± 92b 90 213 ± 102b 97.5 392 ± 52a 105 415 ± 103a 120 119 ± 100b The data are expressed as mean ± SD. Values indicated by the same letter are not signiﬁcantly different at p \ 0.05 (Duncan’s multiple range test)

orthogonal experiment showed that 90 lM was the best concentration of ABA (Fig. 5). In the subsequent experi- Fig. 4 Number of mature somatic embryos per gram fresh weight ments, with the ABA concentrations increased from 75 to (FW) of embryonal masses (EM) from eleven embryogenic cell lines (mean ± SD). Values indicated by the same letter are not signifi- 97.5 lM, the number of cotyledonary somatic embryos cantly different at p \ 0.05 (Duncan’s multiple range test) increased however, it decreased once the ABA was sup- plied at concentration over 105 lM (Table 6). had significant effects on SE maturation. However, the line genotypes did not significantly differ from each other Partial desiccation with light and the color indicator (Table S4). Only a few SEs were induced without PEG 4000, but there were no significant differences in the After 14 days of partial desiccation with six genotypes of number of fully developed embryos over the concentra- mature SEs, two types were identified according to their tions of 2.5–7.5% (Fig. 5). Sucrose at 1 and 3% favored SE appearance. One was green cotyledonary embryos (GCEs, development, while high concentrations (6 and 9%) Fig. 6a), in which the radicals became red, and the cotyle- inhibited differentiation (Fig. 5). Somatic embryos in 1% dons and hypocotyls turned green. The frequency of GCEs sucrose were small, but those in 3% were large and robust. ranged from 70.0 to 74.4%, and did not significantly differ Few SEs were obtained from maturation medium without among the tested genotypes (Table 7). The other type was AC. The effects of the concentrations of AC tested (0.1, non-green cotyledonary embryos (NGCEs, Fig. 6d), with a 0.3, or 0.5%) did not significantly differ (Fig. 5). The lar- swollen appearance and a wrinkled, rough surface. The gest number of mature SEs in response to gellan gum was color of their cotyledons and hypocotyls was yellowish, 133 at 0.4%, and the maturation rate decreased with slightly translucent and they showed no sign of growth. decreases or increases in the gellan gum concentration. The Furthermore, histo-cytological examination showed that the

Fig. 5 Effects of PEG4000, sucrose, AC, gellan gum, and ABA on SE maturation in P. asperata from three ECLs (1113, 2711, and 1931; mean ± SD). Values indicated by the same letter are not signiﬁcantly different at p \ 0.05 (Duncan’s multiple range test). P0, P2.5, P5, and P7.5 indicate 0, 2.5, 5, and 7.5% PEG4000; S1, S3, S6, and S9 indicate 1, 3, 6, and 9% sucrose; AC0, AC0.1, AC0.3, and AC0.5 indicate 0, 0.1, 0.3, and 0.5% AC; G0.2, G0.4, G0.6, and G0.8 indicate 0.2, 0.4, 0.6, and 0.8% gellan gum; and A30, A45, A60, and A90 indicate 30, 45, 60, and 90 lM ABA

123 308 Trees (2017) 31:299–312

Fig. 6 Different germination performance of P. asperata in GCEs germinated for 9 and 35 days, respectively. The ﬁgure is represen- and NGCEs. a A GCE after desiccation. b, c GCEs germinated for 9 tative of at least 20 replicates. (Bars a, d, 500 lm; b, c, e, f, 1 cm) and 35 days, respectively. d An NGCE after desiccation. e, f NGCEs

Table 7 GCE frequency (%), germination rate (%) of three types of SE, and survival rate (%) of germinated GCE from different ECLs of P. asperata ECLs GCE frequency (%) Germination rate (%) Survival rate (%) of germinated GCE Without desiccation Desiccation for 2 weeks NGCE GCE

1113 70.0 ± 6.7a 6.8 ± 4.6bc 18.3 ± 2.9ab 90.0 ± 5.0a 39.7 ± 5.5c 2614 71.1 ± 5.1a 1.5 ± 0.9c 6.7 ± 2.9c 86.7 ± 2.9a 87.5 ± 2.6a 2711 71.1 ± 10.7a 5.7 ± 3.4bc 5.0 ± 5.0c 76.7 ± 7.6c 42.2 ± 4.9c 1911 73.3 ± 8.8a 15.3 ± 9.1b 15.0 ± 5.0b 74.5 ± 4.3c 62.1 ± 18.4b 1921 74.4 ± 12.6a 4.2 ± 2.9bc 20.6 ± 1.0ab 77.8 ± 1.9bc 60.9 ± 15.2b 1931 71.1 ± 11.7a 28.1 ± 10.6a 21.8 ± 1.7a 85.2 ± 3.2ab 71.4 ± 4.8ab Mean 71.9 ± 8.3 10.3 ± 10.7b 14.6 ± 7.3b 81.8 ± 7.0a 60.7 ± 19.0 The data are expressed as mean ± SD. Values indicated by the same letter are not signiﬁcantly different at p \ 0.05 (Duncan’s multiple range test)

123 Trees (2017) 31:299–312 309

Fig. 7 Histology of undesiccated and desiccated SEs of P. asperata. procambium. c Procambium containing irregularly shaped cells. a Mature SE (control). b GCE after desiccation with organized d Cells in the procambium lined up in the same direction with large procambium (pc). c, d NGCEs after desiccation with poorly organized intercellular spaces. (Bars 200 lm) procambium cells in GCEs (Fig. 7b) were normal in embryos started to elongate on the medium. After structure, while those in NGCEs were irregularly shaped 7–14 days, numerous root hairs formed on the exposed part (Fig. 7c), or had large intercellular spaces (Fig. 7d). of the primary root (Fig. 8a), and needle primordia were ANOVA analysis revealed that genotype and embryo initiated from the shoot apical meristem (Fig. 8b). Root type (including non-desiccated embryos) both had signifi- elongation and primary needle development were evident cant effects on germination rate. A significant interaction at 35 days on germination medium (Fig. 8c). After trans- was found among genotype and embryo type (Table S5). plantation into forest containers, somatic seedlings grew GCEs had a higher germination rate than non-desiccated vigorously in the cultivation chamber (Fig. 8d). In the end, embryos and NGCEs (Table 7), and had more converted plantlets were successfully acclimatized in the greenhouse plantlets with well-developed roots and shoots (Fig. 6b, c). (Fig. 8d). The overall survival rate of acclimatized plant- The appearance of GCEs was used as an indicator for lets after 2 months was 60.6%, varying from 39.7 to 87.5% successful partial desiccation treatment of P. asperata. among the cell lines (Table 7). According to examination during desiccation and the results of GCE frequency analysis (Table 7), the indicator was genotype-independent in the tested genotypes, and its Discussion utility was demonstrated by the greatly increased number of SEs able to germinate. As shown in Fig. 6e, the radicles In this study, we subjected cotyledonary SEs to partial of the NGCEs hardly germinated after being placed on the desiccation under low light conditions. Recently, we found germination medium. Finally, the hypocotyls became that PDT promotes the transition of these embryos from hyperhydric, and the radicles turned dark (Fig. 6f). morphological maturity to physiological maturity by increasing the stress-related proteins in P. asperata SEs Somatic seedling production that are deficient in water and nutrients, and further induces photosynthesis in low light (Jing et al. 2016). However, After partial desiccation treatment, a total of 720 somatic different from our study, previous desiccation protocols embryos were transferred onto germination medium. With often specify darkness for the drying treatment of conifer the addition of 360 non-desiccated embryos, 1080 embryos SEs (Kong and Yeung 1992, 1995; Bozhkov and von were germinated. The cotyledons and hypocotyls of the Arnold 1998; Maruyama and Hosoi 2012; Liao and Juan

123 310 Trees (2017) 31:299–312

Fig. 8 Somatic embryo germination in P. asperata. Somatic embryos c Somatic embryo germinated for 35 days with elongation of the germinated for 9 days, showing elongation of the cotyledons (co) and primary needles. d Regenerated plants grown in a greenhouse. (Bars the primary root with root hair (rh) (a), and the anatomical structure a, 2 mm; b, 200 lm; c, d, 1 cm) of the apical meristem (sam) with initiated primary needles (pn) (b).

2015). Since SEs of P. asperata changed color after des- our study, the developmental stage of the immature zygotic iccation, we divided the treated SEs into two types, GCEs embryos had a strong effect on the EM initiation rate. and NGCEs, depending on the color changes. Only GCEs Cotyledonary embryos in the mid-developmental stage had a high germination rate, especially with superior were the best explants. And the results clearly indicated growth of root hairs. These results suggest that the easily that some genotypes had an inherent ability to induce EM, observed color (green cotyledon with red radicle) is an while others consistently produced a gray or brown calli. indicator of high germination potential. Although different Furthermore, different combinations of 2,4-D and BAP had genotypes needed different times to reach the standard strong effects on EM initiation in P. asperata. We found color change, the color indicator was genotype- that 10 lM 2,4-D combined with 5 lM BAP produced the independent. highest percentage of EM initiation. Similar results have A high initiation rate of EM is important for the appli- been reported in some other spruce species, such as white cation of somatic embryogenesis technology. Induction of spruce (Klimaszewska et al. 2011), Norway spruce (Hak- EM in coniferous species is strictly dependent on explants man and von Arnold 1985), and Picea Jezoensis (Ishii (Stasolla and Yeung 2003). In P. asperata, immature 1995). zygotic embryos result in higher EM initiation rate than if The transition from EM to SE is a key developmental mature zygotic embryos are cultured, in agreement with switch, which determines the yield and quality of mature several studies (reviewed by Lelu-Walter et al. 2013). In SEs (von Arnold and Clapham 2008). In the maturation

123 Trees (2017) 31:299–312 311 step, our study showed that 105 lM ABA, 5% PEG 4000, References 3% sucrose, 0.4% gellan gum, 0.1% AC, 0.05% L-glutamine, and 0.1% casein acid hydrolysate was optimal for Attree SM, Moore D, Sawhney VK, Fowke LC (1991) Enhanced the maturation of P. asperata. Previous research has also maturation and desiccation tolerance of white spruce [Picea glauca (Moench) Voss] somatic embryos: effects of a non- shown that a series of factors affected SE differentiation in plasmolysing water stress and abscisic acid. Ann Bot conifers (Filonova et al. 2000; Joshi and Kumar 2013; 68:519–525 Montalbań et al. 2013; Stasolla and Yeung 2003; von Bozhkov PV, von Arnold S (1998) Polyethylene glycol promotes Arnold and Clapham 2008). Klimaszewska et al. (2016) maturation but inhibits further development of Picea abies somatic embryos. Physiol Plant 104:211–224 pointed out that a high frequency of differentiation and Chalupa V (1985) Somatic embryogenesis and plantlet regeneration maturation of SEs in most Pinaceae species occurs in the from cultured immature and mature embryos of Picea abies (L.) presence of ABA, sugars, and agents that restrict the Karst. Commun Inst For Cech 14:57–63 availability of water. As Hay and Charest (1999) discussed, Filonova LH, Bozhkov PV, von Arnold S (2000) Developmental pathway of somatic embryogenesis in Picea abies as revealed by ABA and osmotic stressors such as PEG and sucrose have time-lapse tracking. J Exp Bot 51:249–264 different modes of action, but they usually operate syner- Garcıá-Mendiguren O, Montalbań I, Goicoa T, Ugarte M, Moncaleań gistically to evoke better maturation responses in SEs. In P (2016) Environmental conditions at the initial stages of Pinus our study, the levels of gellan gum and sucrose in the radiata somatic embryogenesis affect the production of somatic embryos. Trees Struct Funct 30:949–958 maturation medium were higher than in the initiation and Grossnickle SC (2011) Tissue culture of conifer seedlings-20 years proliferation media. The results showed that high levels of on: viewed through the lens of seedling quality. In: Riley LE, gellan gum and sucrose promote the maturation of Haase DL, Pinto JR, technical coordinators (eds) National P. asperata SEs. Ramarosandratana et al. (2001) also Proceedings: Forest and Conservation Nursery Associations- 2010. Proc. RMRS-P-65. USDA Forest Service, Rocky Moun- demonstrated that SE maturation in maritime pine tain Research Station, Fort Collins, pp 139–146 improves on maturation medium with a high concentration Gupta PK, Holmstrom D (2005) Double staining technology for of gellan gum (Ramarosandratana et al. 2001). Moreover, distinguishing embryogenic cultures. In: Jain SM, Gupta PK AC had a significant effect on the maturation of P. asper- (eds) Protocol for somatic embryogenesis in woody plants. Springer, The Netherlands, pp 573–575 ata SEs. Although AC does, to some extent, reduce the Hakman I, von Arnold S (1985) Plantlet regeneration through somatic effect of ABA, a previous study has shown that the com- embryogenesis in Picea abies (Norway Spruce). J Plant Physiol bination of ABA and AC not only increases the yield of 121:149–158 Norway spruce cotyledonary SEs but also increases the Harry IS, Thorpe TA (1991) Somatic embryogenesis and plant regeneration from mature zygotic embryos of red spruce. Bot number of genotypes forming cotyledonary embryos, while Gaz 152:446–452 reducing the cost of embryo production (Pullman et al. Hay EI, Charest PJ (1999) Somatic embryo germination and 2005). desiccation tolerance in conifers. In: Jain SM, Gupta PK, The protocol defined in our study provides an effective Newton RJ (eds) Somatic embryogenesis in woody plants. Springer, The Netherlands, pp 61–96 tool for large-scale plant propagation. This procedure has Hazubska-Przybył T, Wawrzyniak M, Obarska A, Bojarczuk K (2015) potential for the clonal propagation of superior genotypes, Effect of partial drying and desiccation on somatic seedling quality and can further advance the clonal forestry of P. asperata. in Norway and Serbian spruce. Acta Physiol Plant 37:1–9 Furthermore, we have developed an indicator of optimal Ishii K (1995) Somatic embryogenesis in Picea glehnii and P. jezoensis. In: Mohan Jain S, Gupta PK, Newton RJ (eds) Somatic partial desiccation that can be used to identify viable SEs, Embryogenesis in Woody Plants. Kluwer Academic Publishers, and thus obtain large number of germinants. Dordrecht, pp 55–65 Jing D, Zhang J, Xia Y, Kong L, Ouyang F, Zhang S, Zhang H, Wang Author contribution statement YX and JZ designed and con- J (2016) Proteomic analysis of stress-related proteins and ducted the experiments; YX and DJ wrote the draft manuscript; LK metabolic pathways in Picea asperata somatic embryos during advised on experimental design and provided language revision; SZ partial desiccation. Plant Biotechnol J. doi:10.1111/pbi.12588 and JW provided plant tissue, laboratory facilities, and project Johansen DA (1940) Plant microtechnique. McGraw-Hill Book supervision. Company, New York Joshi R, Kumar P (2013) Regulation of somatic embryogenesis in Acknowledgements This study was supported by a grant from the crops: a review. Agric Rev 34:1–20 China Twelfth Five-Year Plan for Science Technology Support Klimaszewska K, Trontin JF, Becwar MR, Devillard C, Park YS, (2012BAD01B01). The authors acknowledge Xiao Long Shan For- Lelu-Walter MA (2007) Recent progress in somatic embryoge- estry Research Institute for cone collection and thank Prof. Iain nesis of four Pinus spp. Tree For Sci Biotechnol 1:11–25 Charles Bruce (Peking University, China) for critical reading of the Klimaszewska K, Overton C, Stewart D, Rutledge RG (2011) manuscript. Initiation of somatic embryos and regeneration of plants from primordial shoots of 10-year-old somatic white spruce and Compliance with ethical standards expression profiles of 11 genes followed during the tissue culture process. Planta 233:635–647 Conflict of interest The authors declare that they have no conflict of Klimaszewska K, Hargreaves C, Lelu-Walter MA, Trontin JF (2016) interest. Advances in conifer somatic embryogenesis since year, 2000. In:

123 312 Trees (2017) 31:299–312

Maria AG, Lambardi M (eds) In Vitro Embryogenesis in Higher Park YS (2002) Implementation of conifer somatic embryogenesis in Plants. Springer, New York, pp 131–166 clonal forestry: technical requirements and deployment consid- Kong L, Yeung EC (1992) Development of white spruce somatic erations. Ann For Sci 59:651–656 embryos: II. Continual shoot meristem development during Park YS, Barrett JD, Bonga JM (1998) Application of somatic germination. In Vitro Cell Dev Biol Plant 28:125–131 embryogenesis in high-value clonal forestry: deployment, Kong L, Yeung EC (1995) Effects of silver nitrate and polyethylene genetic control, and stability of cryopreserved clones. In Vitro glycol on white spruce (Picea glauca) somatic embryo devel- Cell Dev Biol Plant 34:231–239 opment: enhancing cotyledonary embryo formation and endoge- Pullman GS, Gupta PK, Timmis R, Carpenter C, Kreitinger M, Welty nous ABA content. Physiol Plant 93:298–304 E (2005) Improved Norway spruce somatic embryo development Kru¨ssmann G (1983) Handbuch der Nadelgeho¨lze, 2nd edn. Paul through the use of abscisic acid combined with activated carbon. Pareyc, Hamburg Plant Cell Rep 24:271–279 Lelu-Walter MA, Thompson D, Harvengt L, Sanchez L, Toribio M, Ramarosandratana A, Harvengt L, Bouvet A, Calvayrac R, Paques M Paˆques LE (2013) Somatic embryogenesis in forestry with a (2001) Effects of carbohydrate source, polyethylene glycol and focus on Europe: state-of-the-art, benefits, challenges and future gellan gum concentration on embryonal-suspensor mass (ESM) direction. Tree Genet Genomes 9:883–899 proliferation and maturation of maritime pine somatic embryos. Li Q, Zhang K, Luo JX (2007) A preliminary research on the In Vitro Cell Dev Biol Plant 37:29–34 variation of growth traits and clonal selection in clonal seed Roberts DR, Sutton BCS, Flinn BS (1990) Synchronous and high orchard of Picea asperata. J Sichuan For Sci Technol 28:21–27 frequency germination of interior spruce somatic embryos Liao YK, Juan IP (2015) Improving the germination of somatic following partial drying at high relative humidity. Can J Bot embryos of Picea morrisonicola Hayata: effects of cold storage 68:1086–1090 and partial drying. J For Res 20:114–124 Salajova T, Salaj J (2001) Somatic embryogenesis and plantlet Lin J, Hu Y (2000) Atlas of structure in the gymnosperms from China. regeneration from cotyledon explants isolated from emblings and Science Press, Beijing seedlings of hybrid firs. J Plant Physiol 158:747–755 Litvay JD, Verma DC, Johnson MA (1985) Influence of a loblolly Stasolla C, Yeung EC (2003) Recent advances in conifer somatic pine (Pinus taeda L.). Culture medium and its components on embryogenesis: improving somatic embryo quality. Plant Cell growth and somatic embryogenesis of the wild carrot (Daucus Tissue Organ Cult 74:15–35 carota L.). Plant Cell Rep 4:325–328 Stasolla C, Kong L, Yeung EC, Thorpe TA (2002) Maturation of Luo JX, Sun P, Li XQ (2001) Advances in genetic improvement of somatic embryos in conifers: morphogenesis, physiology, bio- spruce in abroad and breeding strategies of Picea asperata Mast. chemistry, and molecular biology. In Vitro Cell Dev Biol Plant J Sichuan For Sci Technol 22:31–40 38:93–105 Maruyama TE, Hosoi Y (2012) Post-maturation treatment improves von Arnold S, Clapham D (2008) Spruce embryogenesis. In: Sua´rez and synchronizes somatic embryo germination of three species MF, Bozhkov PV (eds) Plant embryogenesis. Humana Press, of Japanese pines. Plant Cell Tissue Org Cult 110:45–52 Totowa, pp 31–47 Montalbań I, Setień-Olarra A, Hargreaves C, Moncaleań P (2013) Vookova´ B, Kormut’a´k A (2006) Comparison of induction frequency, Somatic embryogenesis in Pinus halepensis Mill.: an important maturation capacity and germination of Abies numidica during ecological species from the Mediterranean forest. Trees Struct secondary somatic embryogenesis. Biol Plant 50:785–788 Funct 27:1339–1351

123 本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具