Advances of the Flowering Genes of Gymnosperms

Flowering is an important stage in the life cycle of plants and also a turning point from vegetative growth to reproductive growth. This process is affected by many exogenous and endogenous factors. Some examples of the latter are endogenous hormones, plant growth status, nutrient composition, and flowering regulatory genes. Many gymnosperms have a long juvenile period. Previous studies attempted to shorten this period using traditional asexual propagation methods, but significant results have not been achieved. In recent years, molecular biology is used to study the flowering regulatory gene to obtain transgenic plants with early flowering trait. Thus, the production of gymnosperms is hastened, and economic efficiency is improved. Studies have shown that the flowering genes of plants act synergistically to form a complex network. In this paper, we reviewed the recent development in the study of the regulation of the flowering genes of gymnosperms, that is, from the floral meristemspecific gene, floral organ-specific gene, genes that inhibit plant flowering, and microRNA regulation of flowering. We provide a reference for the in-depth study on the genetic improvement of the flowering gene.


Introduction
Gymnospermae is an advanced plant between the fern and angiosperm, which is a more primitive seed plant.This species has been widely used in greening, medicinal, and dietary consumption.Some of these rare species are remains from the quaternary glaciers and are considered as rare endangered plants.Many gymnosperms have a long juvenile period.For example, the juvenile period of gingko is as long as 20-30 years (Singh et al., 2008).This plant takes 8-10 years to blossom and yield after being grafted.Many studies have long performed asexual reproduction technologies, for instance, cutting and grafting, for gymnosperms.Although these methods can shorten the juvenile period to a certain extent, the flowering and bearing of fruit still requires a long time.These characteristics are not conducive to the breeding of superior varieties and the application for high economic value.
The flower is the important reproductive organ of a plant.Flowering plays a central role in plant growth and species evolution.The flowering transformation is the process of transforming the plant from the vegetative growth to the reproductive growth.Flowering is affected by exogenous factors, such as light, temperature, exogenous hormones, moisture, and soil fertility, and endogenous factors (the genes involved in plant flowering regulation) (Levy and Dean, 1998;Khodorova and Michèle, 2013;Cho et al., 2016).The isolation of FLORICAULA (FLO) and DEFICIENS (DEF) genes from the model plant A. majus and the AGAMOUS (AG) gene cloned from A. thaliana indicate that studies on floral development have reached molecular level (Coen et al., 1990;Sommer et al., 1990;Yanofsky et al., 1990).The development of molecular biology has led to in-depth studies on the regulation of plant flowering, and considerable progress has been achieved.At present, the flowering regulatory genes have been isolated from A. majus, A. thaliana, O. sativa, P. radiata, P. mariana, and G. gnemon (Tröbner et al., 1992;Weigel et al., 1993;Yamaguchi et al., 2006;Mouradov et al., 1998;Rutledge et al., 1998;Shindo et al., 2001).Studies of these flowering regulatory genes and mutants have shown that these molecules can effectively promote the flowering period and shorten the juvenile period.Moreover, the widely existing microRNAs in plants are also involved in the regulation of floral organ development and the flowering time (Spanudakis and Jackson, 2014;Aukerman and Sakai, 2003).These flowering regulators constitute a complex network that collectively regulates the flowering (Jack, 2004).
In recent years, molecular biology technology has been employed to study the flowering regulatory gene.Some early flowering-improved plants with effectively shortened juvenile period have been obtained by genetically transforming the flowering gene.For example, the wild P.
regulatory gene of the floral meristem.To date, LEAFY homologous genes, as an example LFY, FLO, NFL, and RFL, have been isolated from many angiosperms (Coen et al., 1990;Kelly et al., 1995;Weigel et al., 1995；Kyozuka et al., 1998).The LFY gene is the earliest expression of floral meristem-specific gene, and AP1/CAL and other genes are their downstream target genes (William et al., 2004).As a plant ages, the expression of the LFY gene gradually increases, reaching the highest level during reproductive growth.This process activates a series of downstream target genes to enable the plant to successfully complete flowering.The LFY gene can synergize with the AP1 gene to inhibit the activity of the EMF gene.This process promotes the conversion of the inflorescence meristem to the floral meristem and enables the plant to complete the flowering transition (Chen et al., 1997).Dornelas et al. (2005) transformed the PcLFY gene of P. caribaea to the A. thaliana lfy-26 mutant.The transgenic plants had the same phenotype as the wild plants.This result suggested that the LFY homologous genes of gymnosperms and angiosperms may have similar biological functions.
davidiana Dode needs 8-12 years to bloom, but the LFY mutant -P.davidiana can bloom in 5 months (Blázquez et al., 1997).Previous studies have shown that after transferring the LFY and APETALA1 (AP1) genes into citrus, both transgenic plants could bloom within a year, which is 3-5 years earlier than the wild plants, with normal fruit development and stable genetic trait (Peña et al., 2001).Therefore, studying the regulation of flowering genes in the gymnosperm can promote the early flowering of this species and is also conducive to breeding fine varieties and improving the economic benefit of gymnosperms.We reviewed the floral meristem-specific gene, floral organspecific gene, flowering suppression gene, and flowering microRNA.The expression patterns or functions of different flowering genes of gymnosperms are listed in Table 1.

Floral meristem-specific gene
In angiosperms, the LEAFY gene has an important role in determining the flowering time of plants.The  However, unlike angiosperms, gymnosperms have double-copy LEAFY homologous genes (Frohlich and Parker, 2000;Himi et al., 2001).Frohlich and Parke (2000) proposed the "mostly male theory" based on the differences in the spatiotemporal expression of double-copy LEAFY homologous genes in gymnosperms and the deletion of NLY genes in angiosperms.However, in the study of P. caribaea and G. parvifolium, the LEAFY homologous genes PcLFY and GpLFY were expressed in the female spherules (Dornelas et al., 2005；Shindo et al., 2001).This result weakened the basis of the "mostly male theory".
The LFY homologous genes of P. radiata, namely, PRFLL and NEEDLY, were found to be expressed in the vegetative and reproductive organs and constitutively expressed.However, the PRFLL genes were expressed in vegetative buds and male bulbs.The NEEDLY gene was mainly expressed in the female spherules (Mellerowicz et al., 1998;Mouradov et al., 1998).In G. biloba, the female plant LEAFY homologue gene Ginlfy and the male plant LEAFY homologue gene GinNdly have also been cloned (Zhang et al., 2002a;Zhang et al., 2002b).The result of the expression profile showed that Ginlfy was tissue-specifically expressed in the leaves of the flower buds, saplings, and mature female and male plants.The GinNdly gene was expressed in the roots, leaves, female and male flower buds, and young fruits of G. biloba as constitutive expression (Guo et al., 2005).In P. abies, the PaLFY gene was expressed in female scapes, and PaNLY was expressed in the surrounding tissues (Carlsbecker et al., 2004;Sundström et al., 2002).The PoLFY of P. macrophyllus was expressed in the female flower buds, and PodNLY was expressed in ovule primordium and epimatium, and PoNLY was expressed in the cones, quills, ovule primordia, and epimatium (Vázquez et al., 2007).In P. massoniana, the PmLFY and PmNLY genes main involved in the development process of the female cones and PmLFY also involved in strobile development (Chen et al., 2015).The difference in the spatial-temporal expression of the double-copy LFY homologous gene of gymnosperm may be the reason for the prolonged juvenile period of gymnosperms.The LFY gene may undergo functional differentiation in the long-term evolution and has different functions for inhibiting the development of vegetative organs and genital organs.

Floral organ-specific gene
The MADS-box gene is found widely in plants.This gene regulates all stages of the floral development and plays a decisive role in the development of a floral organ.In addition to the AP2 gene, most floral organ-specific genes contain the MADS-box DNA transcription factor region.Thus, this type of gene is called the MADS-box gene family.According to the ABC model of floral development (Bowman et al., 1991;Coen and Meyerowitz, 1991), the class A gene, AP1, is involved in the formation of sepal.The A and B genes determine the formation of the petal.APETALA3 (AP3), PISTILLATA (PI), DEFICIENS (DEF), and GLOBOSA (GLO) are Class B genes (Falkowski and Dubinsky, 1981;Sommer et al., 1990;Tröbner et al., 1992).The class B and C genes jointly inhibit the development of the pistil and stamen, while the class C gene determines the development of the carpel.The AG gene of A. thaliana belongs to class C gene (Yanofsky et al., 1990).However, the ABCDE model was deduced from the continuous study of flower development (Theissen and Saedler, 2001).The class D gene, like FLORAL BIDING PROTEIN 7 (FBP7) and FBP11 of P. hybrida, and AGL11 of A. thaliana, is the major gene that inhibits the development of ovules (Colombo et al., 1995;Rounsley et al., 1995;Angenent and Colombo, 1996).The class E gene is involved in the development of the petal, stamen, and carpel.SEPALLATA1 (SEP1), SEP2, SEP3, and SEP4 are typical examples of this class of gene (Pelaz et al., 2000;Ditta et al., 2004).By using the characteristics of MADSbox gene in different flower organs, some genes can be applied to plant genetic manipulation in order to reconstruct the horticultural traits of plants, and also shorten the juvenile period of long gymnosperms to reduce production time and increase yield.With the deepening of these studies, the application value of MADS-box gene will be gradually revealed.
According to reports, many MADS-box genes have been isolated from P. abies, P. mariana, G. gnemon (Tandre et al., 1995;Rutledge et al., 1998;Sundström et al., 1999;Winter et al., 1999).In P. radiata, PrDGL was expressed in emergent male cone primordia and persisted through the early stages of pollen cone bud differentiation (Mouradov et al., 1999).The GpMADS1, 3, 4 gene of G. Prvifolium were expressed during the early stage of ovule development in the differentiating nucellus and envelopes (Shindo et al., 1999).The studies of G. gnemon have shown that the expression of all four MADS-box genes (GGM 7,9,11,15) is limited to reproductive units (especially in the early stage of reproductive organ development), the GGM15 transcript is even restricted to male reproductive organs (Becker et al., 2003).These genes were specifically expressed in the reproductive organs but not in vegetative organs.The role of MADS box gene in the floral development of gymnosperm and angiosperm has been perceived to be very conservative (Theißen, 2001).However, new ideas have gradually emerged in subsequent studies.The MADS-box genes played a role in the growth of floral organs and were expressed during flowering initiation, fruit growth, differentiation, and formation of meristems, embryos, roots, and vascular tissues ( Van der Linden et al., 2002).This concept was confirmed by the expression of some gymnosperm.For instance, the homologue gene GBM5 of G. biloba in the stamens, ovules, and gametophytes of this species as well as in the young leaves of both male and female plants (Jager et al., 2003).In the study of C. japonica, the CjMADS15 gene was expressed in all organs except pollen, and was especially expressed in needles, CjMADS14 gene was expressed mainly in male and female strobili.These two genes play important roles during the development of male and female strobili in C. japonica (Katahata et al., 2014).The study of these MADS-box gene functions is of great theoretical significance for explaining various physiological phenomena in plants.
Our research group has reported some floral organspecific genes of G. biloba.The CONSTANT gene regulates the expression of the downstream FT gene in the leaves by responding to photoperiodic signals.The FT protein accumulates in the leaves and is transferred to the shoot tips to induce the flowering of the plant.The CONSTANT homologous gene GbCO has the highest expression level in the G. biloba bud tip and is regulated by a photoperiod.It promotes the early flowering by activating the downstream FT gene (Yan et al., 2017).The GbCOL16 gene of G. biloba is then cloned, with the highest expression in the leaves.The expression in the male spores is higher than that in the female sporophyll and young fruit.This phenomenon indicates that GbCO and GbCOL16 have similar flowering regulatory mechanisms with the CO gene (Wang et al., 2017).GbMADS9 of G. biloba belongs to the Bsister-class MADS-box gene and is expressed in the spherules and ovules.Overexpression of GbMADS9 leads to the early flowering of transgenic A. thaliana and enhances the ability of A. thaliana to tolerate permeation.Thus, GbMADS9 may be involved in the regulation of the flowering time of G. biloba and enhance the ability of this species to withstand abiotic stresses (Yang et al., 2016).The repression levels of GbSEP and GbMADS2 cloned from the female and male flowers of G. biloba are significantly higher than those in the roots, stems, and leaves.The expression of GbSEP in the female flowers is significantly higher than those in the male flowers, however GbMADS2 is opposite to GbSEP.The expression of the two genes increases as the flowers grow.This phenomenon indicates that these genes may be involved in the growth of G. biloba (Cheng et al., 2016;Wang et al., 2015).GbAGL66 is strongly expressed in the roots and flowers, with the highest level in the roots.This gene is also detected in the fruit.Thus, the GbAGL66 gene may be involved in the regulation of flower and fruit growth (Dou et al., 2017).GbAP2 gene is expressed in the roots, stems, leaves, male spores, female spores, and fruits and strongly expressed in the leaves and female spores.These characteristics indicate that GbAP2 may be involved in the growth and development of G. biloba (Zhang et al., 2017).These studies indicate that the MADS-box gene plays an important role in regulating the flowering time and floral organ development of G. biloba, and provides a reference for the flower-specific gene regulation of other gymnosperms.
Despite remarkable progress in the regulation of floral organ-specific gene, and numerous flower developmentrelated transcription factors have been cloned from various plants, how to use genetic engineering to precisely control plant development and how to use these transcription factors applied to agricultural production is still an urgent problem to be solved.

Genes that inhibit plant flowering
The EMBRYONIC FLOWER (EMF1 and EMF2) gene plays an important role in maintaining vegetative growth and inhibiting flower development (Calonje et al., 2008;Moon et al., 2003).The emf1 and emf2 mutants do not undergo vegetative growth but flowered directly (Sung 4 et al., 1992;Yang et al., 1995).These phenomena indicate that the loss of EMF function results in the early flowering of the plant.As the plant ages and the exogenous environment changes, the inhibition of EMF gene is gradually decreasing (Yang et al., 1995).When the expression drops to a certain level, the plant enters the flowering transformation.The expression of LFY and AP1 genes, begins to increase, overcoming the floral repression of EMF and enabling the plant to complete its flowering (Sung et al., 2003).Many angiosperm EMF genes, for instance in A. thaliana, O. sativa, and B. oleracea, have been reported (Sung et al., 1992;Li et al., 2006;Liu et al., 2012), but studies of EMF genes in gymnosperms have not been reported.
TFL is a member of the FT/TFL gene subfamily and plays a key role in flowering.To date, the effect of the TFL gene on flowering inhibition has been reported in many plants (Alvarez et al., 1992;Bradley et al., 1996;Nakagawa et al., 2002).The LFY and AP1/CAL are generally believed to suppress flowering by inhibiting the expression of the TFL1 gene in flowers.However, some studies have indicated that the TFL1 transcription was inhibited by AP1 but promoted by LFY (Shannon and Meeks-Wagner, 1991;Liljegren et al., 1999;Pillitteri et al., 2004;Serrano-Mislata et al., 2017).In the mutant strains without TFL1 function, the inflorescent meristem was rapidly transformed to the floral meristem, which significantly promoted flowering (Banfield and Brady, 2000).Studies have shown that after transferring the antisense MdTFL1 into apples, the transgenic plants blossomed within 8-10 months after being grafted (Kotoda et al., 2006).This characteristic indicated that removing the TFL gene to inhibit flowering could shorten the juvenile period.In gymnosperms, the PaFTL gene of Norway spruce was functionally similar to the TFL1-like gene (Karlgren et al., 2011).PaFTL1 could inhibit the development of male spherules and meristems, while PaFTL2 could inhibit the growth of needles and vegetative buds.
The FLC gene inhibits flowering, and the encoded MADS-box transcriptional regulatory protein is an inhibitor of flowering.The FLC gene regulates the flowering time by responding to vernalization that can downregulate the activity of the FLC gene and promote the flowering of the A. thaliana late flower type and late flower mutant (Sheldon et al., 2000).Studies have shown that FLC suppressed flowering by inhibiting the expression of two downstream flowering genes, namely, SOC1 (AGL20) and FT (Lee et al., 2000;Michaels and Amasino, 2001;Michaels et al., 2005), which promote flowering.In a recent study, the ft-1/flc-21 double mutant was identical to the flc-21 phenotype, but unlike ft-1 (Chen and Penfield, 2018).This phenomenon indicated that FLC was downstream of FT, which was inconsistent with previous findings.However, studies of FLC genes in gymnosperms have not been reported.Few studies have been conducted on the genes suppressing the flowering of gymnosperms.Thus, additional studies are needed to further understand the molecular mechanisms that inhibit these flowering genes.

microRNA regulation of flowering
The microRNAs are endogenous non-coding small RNAs of 21-24 nucleotides in length and found in a wide variety of plants.These molecules can regulate genes at the post-transcriptional level by completely or partially matching with the target gene mRNAs.These molecules regulate the expression of plants gene by transcriptional cleavage or inhibition of target gene mRNA translation (Reinhart et al., 2002;Bartel, 2004).The microRNA functions of some early identified plants are conserved, and studies on flowering microRNAs have focused on model plants (Reinhart et al., 2002;Chi et al., 2011).
A microRNA has multiple target genes or multiple microRNAs to regulate a target gene to form a complex flowering regulatory network.Acting as a posttranscriptional regulator, the microRNAs function broadly to control many aspects of plant biology and plant development, and play a key role in the regulation of plant flowering and floral organ development (Nag and Jack, 2010;Wu, 2013).The miR156 controls the transformation from vegetative growth to reproductive growth through the target regulation of SPL (SQUAMOSA Promoter-binding protein-like) transcription factors (Yang et al., 2011).The miR172 regulates the flowering time and floral organ development through the translation inhibition or cleavage of AP2-like family genes (Jung et al., 2007;Glazińska et al., 2009).According to reports, miR156 and miR172 are involved in regulating the timing of sensitivity of the vernalization response in Cardamine flexuosa, and modulated the expression of CfSOC1 gene to regulate flowering (Zhou et al., 2013).The miR159 modulates the MYB transcription factors and maintains the normal growth of anthers (Millar et al., 2005).The miR164 family (miR164a, miR164b, and miR164c) regulates the transformation among petals, pistils and stamens (Aida et al., 1997).The target genes are the NAC family of transcription factors (e.g., CUC1 and CUC2).
In the model plant A. thaliana, miR156 can postpone the flowering period, while miR172 can predate this period.The expression of miR156 decreases from the young to the adult, but the expression of miR172 increase (Wu et al., 2006).The miR156 negatively regulates the SPL3 expression and inhibits A. thaliana flowering.The miR156 target genes SPL9 and SPL10 positively regulate the expression of miR172 and have an indirect effect on inducing flowering (Wu et al., 2009).The miR172 negatively regulates the AP2 gene.This process reduces flowering inhibition and facilitates the successful complete flowering (Aukerman and Sakai, 2003).The miR164 negatively regulates CUP SHAPED COTYLEDON 1 (CUC1) and CUC2, and this process affects the growth of the meristem and the generation of floral organ primordia (Mallory et al., 2004).To date, many novel and conserved microRNAs involved in flower development in angiosperms have been discovery and profiled (Wang et al., 2012;Wang et al., 2014;Sun et al., 2015).In these studies, most of the conserved miRNAs in these species are highly conserved among plants (e.g., miR156, miR167 and miR172 etc.), indicating that these miRNAs have important and conserved functions in plant development.These works provide a good reference for the study of flower development of gymnosperms.
We reviewed some microRNAs involved in the regulation of flower development in gymnosperms, which are listed in Table 2.In the gymnosperm Norway spruce, of the 22 conserved miRNA families, 8 miRNAs conserved in embryonic plants and 13 other miRNAs conserved in angiosperms were detected, indicating that these miRNAs are present in the common ancestor of spermatophytes (Xia et al., 2015).It has been reported that pab-miR159a may regulate PaGaMYB expression in Norway spruce and may participate in seed germination and flower development (Yakovlev et al., 2010).Numerous microRNAs involved in the regulation of floral organ development in angiosperms have been identified from leaves and ovules of G. biloba.Some of these microRNAs were expressed in ovules, including the miR156, miR164, miR167, miR169, miR172, and miR390 families (but not miR164c and miR169b), indicating these microRNA plays an important role in the development of floral organ (Wang et al., 2016).Some conserved microRNAs (miR166a, miR166b, miR172, miR399, and miR776) were also isolated from P. taeda, and these molecules were expressed in the female and male gametophytes and needle tissues with almost identical levels.This result suggested that these microRNAs were involved in the development of the male and female gametophytes.Pta-miR157a and pta-miR157b were highly expressed in the needles and mature pollen, but lowly expressed in germinated pollen.This expression pattern is opposite to pta-miR161.2 and pta-miR164b, indicating that these microRNAs Participated in the development of pollen (Quinn et al., 2015).However, the specific mechanisms of microRNAs in the regulation of flowering of gymnosperms remain ambiguous.Further investigation on the specific regulatory functions of these microRNAs will contribute to the understanding of the floral development in gymnosperms.

Outlook
Flowering is a very complex physiological and biochemical process regulated by a network of genes composed of various flowering regulators (Schultz and Haughn, 1991).The LEAFY is lowly expressed during vegetative growth.Affected by diverse internal and external factors, a plant gradually releases the flowering inhibition.When a certain threshold value is reached, LEAFY activates the floral organ specificity gene AP1, and the synergistic effect smoothly completes the flowering conversion (Parcy et al., 1998).An in-depth study of the synergistic effects of flowering regulators of gymnosperms and the effects of regulatory networks on the floral development will provide a theoretical basis for reducing the juvenile period and genetically improving gymnosperms.To date, studies on the mechanism and application of the flowering regulators that regulate flowering have mainly focused on model plants, such as A. thaliana and N. tabacum (Eckardt, 2005).Studies on gymnosperms are rarely reported.Additional investigations should be performed to reveal the mechanism of flowering regulation of gymnosperms and the mechanism of their interactions.
With the development of molecular biology, the clonal identification of numerous plant flowering genes provides a theoretical basis to understand the evolutionary processes and phylogenetic relationships of plants.A series of studies have shown that the flowering genes were highly conserved among closely related species.This finding revealed the homology of plant origin.Therefore, the transcription of these flowering homologous genes into gymnosperms within a prolonged juvenile period facilitate the rapid blossoming reproduction of transgenic plants.This process will greatly increase the economic benefits of gymnosperms.However, given the difficulties in the tissue culture, in vitro regeneration, and gene transformation techniques of some gymnosperms, these genes can only be transcribed into ectopic expressions in plants.Several limitations should be solved to transcribe the flowering genes into corresponding gymnosperms to obtain flowering transgenic plant.Given the continuous development of tissue culture technology and molecular biology technology, the genetic improvement of gymnosperms will result in greater progress and better production of gymnosperms.

Table 1 .
Expression patterns or functions of different flowering genes of gymnosperms

Table 2 .
MicroRNAs involved in the regulation of flower development in gymnosperms