miRNA and mRNA studies reveal pollination activates hormonal signaling and increases fruit set in the apomictic tree Zanthoxylum bungeanum Maxim

Background Apomixis is a form of reproduction that does not involve fertilization of female by male gametes but instead produces offspring from the female parent directly. The progeny of apomixis is genotypically identical to the female parent and so maintains any elite traits of the female parent. Apomixis has considerable potential in genetic plant breeding. However, the mechanism of apomictic reproduction remains unclear. Zanthoxylum bungeanum is an apomictic plant. Studies on miRNAs, mRNAs, hormone changes and fertilization process of the pollinated and non-pollinated materials of Zanthoxylum bungeanum allows screening of the important regulatory factors of the apomictic process at both physiological and molecular levels. This information should be of considerable help in understanding the mechanism of apomixis in this species. Results Our results show that Zanthoxylum bungeanum pollen can germinate on the stigma, and that the pollen tube can extend to the ovary wall after two days but that it then degenerates about the fifth day and before reaching the egg cell. So fertilization does not occur. Nevertheless, pollination did increase fruit set from 74.12% (unpollinated) to 89.31% (pollinated). The reproductive mode of the Zanthoxylum bungeanum cultivar ‘Hancheng Dahongpao’ was identified as obligate apomixis. Enrichment analysis of differential genes indicates that pollination activates genes involved in the synthesis of ABA, IAA, GA3 and JA, and that genes related to asexual development, genes involved in flower development, and transcription factors are also activated. RT-qPCR verification shows that the relative expression levels of mRNAs and miRNAs related to hormone signaling pathways and flower development were negatively correlated. Also, that miRNA is an important factor influencing hormone regulation after pollination of this apomictic species. The contents of ABA, IAA, GA3 and JA were all significantly increased following pollination.


3
Conclusions In general, although pollination does not result in fertilization in this species, it does activate mRNAs and miRNAs which serve as signals to activate the anabolic pathways for ABA, IAA, GA3 and JA. Increased levels of these hormones result in increased fruit set.
Background Zanthoxylum bungeanum Maxim. (ZB), common name Chinese prickly ash, is a shrub or dungarunga belonging to the rutaceae. There are about 250 species of the genus Zanthoxylum, which is widely distributed round the world, though most species of this genus originate in Asia [1]. Based on the color of the fruit at maturity, it is divided into two types: red Chinese prickly ash and green Chinese prickly ash. The young leaves of ZB can be eaten as a vegetable and the fruits, stems and roots can be used as medicines [2].
The fruit skin is one of China's eight most important condiments. Having a unique numbing taste. This product is in wide use in Asian cuisine and is an important seasoning for the traditional Chinese hot pot [3,4]. The fruit skin of ZB is a traditional herbal medicine in China and has been in use for more than 2,000 years [5]. Over 140 chemical compounds have been identified in ZB, including alkaloids, anthraquinones, flavonoids and free fatty acids [6][7][8][9]. Used as a medicine, the skin has anti-inflammatory, antibacterial and insecticidal effects [10][11][12][13]. For these reasons, ZB is an important species, that also has significant potential for commercial development.
Interestingly, the normal reproductive mode of ZB is apomictic [14] -a mode that produces viable seed without pollination. Hence, the genotype of the offspring is identical to that of the female parent [15,16]. In a commercial species, apomixis can be both a disadvantage and an advantage. The apomictic mechanism maintains the superior traits of the female for breeding, and can also provide a useful model for the genetic breeding of plant traits. However, the performance of apomictic plants following pollination has rarely been reported. Therefore, a study of differences between plants resulting from pollination and from non-pollination should help reveal the mechanisms of apomixis and thus provide a useful reference for plant genetic breeding research.

Pollen germination
Pollen germination tests were carried out on pollen on the in vitro germination medium, and pollen germination rate was counted and expressed as a percentage. The results 4 showed the pollen germination rate of ZB pollen increased from 4.30% after 3 h to 10.70% after 12 h ( Figure 1 and Table 1). In general, the germination rate of ZB pollen is low, which presents a huge obstacle to sexual reproduction in this species.   After 5 d, the pollen tube is broken down into callosum so the fertilization process cannot be completed. We observed 300 samples and found no successful fertilization in any of them. It is thus preliminarily concluded that the reproductive mode of ZB cv. 'Hancheng Dahongpao' is obligate apomixis.

Path enrichment analysis
The pathway enrichment analysis of differential genes reveals changes before and after pollination. It can be seen from Figure 3 that pollination has a significant effect on the various pathways of ZB than unpollinated. After pollination, hormone signal transduction is most active, as are metabolic pathways including starch and sucrose metabolism, glycolysis/gluconeogenesis and phenylpropanoid synthesis. Figure 3b shows the pathway of increased expression after pollination, and Figure 3c

Hormone content analysis
The transcriptome and miRNA sequencing shows many hormone pathways became active after pollination. We recorded the levels of four hormones ABA, IAA, GA3 and JA were for 6 pollinated and unpollinated materials. The results show the four hormones were significantly higher in the pollination material than in the unpollinated ( Figure 6). This indicates pollination can activate the synthetic pathway of these four hormones, it also suggests these four hormones participate in the development of embryos and fruits of ZB and, ultimately, in fruit set. Comparing the pollinated with the unpollinated ZB material (Table 2), it is clear that pollination significantly increased fruit set rate but had little effect on fruit size.

Discussion
Double fertilization is one of the most common reproductive modes in plants, egg cells combine with the polar nuclei of pollen [23]. The fertilized egg cells then develop into embryos. The latter develops into endosperm, which provides energy for seed germination. Nevertheless, some plants can form viable seeds without fertilization having taken place. This indicates fertilization is not essential for seed formation [24]. Analysis of hormones in unpollinated and pollinated ovaries of various species reveals that fertilized eggs and developing seeds release hormones that are important determinants of fruit set [25,26].
There are many obstacles to the success of sexual reproduction in plants. In this study, the in vitro germination rate of pollen was very low, with a germination rate after 12 h of only about 11%. In addition, pollen research in the Zanthoxylum piperitum showed that the pollen germination rate was only 10%, which was mutually verified with the results of this study. This is far lower than for other species [27][28][29][30].  In most plants, pollination is followed by fertilization, the formation of a zygote and this the development of a fruit containing viable seed(s). In an obligate apomictic plant, fertilization fails but the fruit nevertheless develops and is contains viable seeds. By analyzing the pollination process in ZB, we show that pollination fails to combine the male and female gametes but it does increase the production of ABA, IAA, GA3 and JA and so the fruit sets. We speculate that in ZB, pollination fails to achieve fertilization but nevertheless delivers a signal which activates a series of hormone synthesis pathways, which produce hormones that promote both fruit set and fruit development (Figure 7).
This view is supported in tomato where it has been found that low levels of ABA lead to poor fruit development [31]. In cherry, ABA and its signal transduction related gene PaSnRK2.1/2.2/2.4 are highly expressed in the ovary and young fruit. This indicates ABA plays an important role in fruit development [32]. In Arabidopsis, ABA has also been shown to act as a switch for flower development. It has been shown that ABA activates photoperiod response genes and flowering genes to promote flowering, but ABA can also inhibit flowering, independent of the flowering gene [33]. The interaction between hormones is an important element of floral organ development. In the citrus ovule, it has been shown that IAA activates the expression of the GA3 biosynthetic gene, while also reducing the breakdown of GA3. Applications of exogenous IAA or GA3 to unpollinated citrus flowers restores fruit set and fruit development to the same levels as pollination [34]. ABI5 is a transcription factor that binds specifically to the auxin response element 5'-TGTCTC-3', thereby functioning as a transcriptional activator. It can form a dimer with Aux/IAA protein to participate in the expression of auxin while promoting the synthesis of JA [35][36][37]. ARF7 is also an auxin response factor with high expression levels in tomato ovaries and these regulate fruit development [38]. In many plants, successful pollination and fertilization can induce increases in auxin and gibberellin levels in the ovary [39][40][41].
Moreover, after pollination, auxin and gibberellin response genes were up-regulated, indicating auxin and gibberellin signals can be induced by pollination [42].
We selected 300 samples to identify the reproductive mode in ZB Hancheng Dahongpao. We found no evidence of successful fertilization. We conclude its reproductive mode is obligate apomictic. However, due to the complex genetic background of ZB genotypes and the wide phenotypic variability among its cultivars [43,44], we cannot exclude the possibility that some ZB genotypes may well be able to reproduce through normal sexual processes. We recorded fruit set and fruit size of pollinated and Each analysis was repeated three times.

Pollen germination
The collected ZB anthers were surface sterilized with 1% sodium hypochlorite for 8 minutes, washed three times with sterile water, and then the pollen in the anthers was transferred to a solid medium. The medium components were agarose (5.8 g/L), sucrose (50 g/L) and boric acid (0.01 g/L) [17][18][19].

RT-qPCR
The RT-qPCR reaction was carried out using the cDNA of the collected material as a template. The primers for RT-qPCR were designed by Primer Premier 5.0 (Palo Alto, CA, USA). These primers are listed in Table 3. The RT-qPCR reaction was carried out in a for 45 s. ZBUBQ and ZBUBA were used as reference genes to correct the relative expression levels of mRNAs [20] and U6 was used as a reference gene for miRNAs [21]. Declarations 12 X.F. wrote the paper. All authors (X.F., Y.M., T.Y. and A.W.) discussed the results and commented on the manuscript.

Funding
The funders had no role in the experiment design, data analysis, decision to publish or preparation of the manuscript. This study was financially supported by the National Key  Figure 1 In vitro germination and pollen morphology of Zanthoxylum bungeanum pollen.  Hormone signal regulation model after pollination of Zanthoxylum bungeanum.