Analysis of Ficus hirta Fig Endosymbionts Diversity and Species Composition

: Endosymbionts living in plants and insects are pervasive. Ficus (Moraceae) has very special inﬂorescences (which we also call ﬁgs) enclosed like an urn, and such inﬂorescence is usually parasitized by ﬁg wasps. Ficus breeds ﬁg wasp larvae in its ﬁgs and adult ﬁg wasps pollinate for Ficus , Ficus and its obligated pollinator formed ﬁg-ﬁg wasp mutualism. Previous studies have found that this conﬁned environment in ﬁgs may have provided protection for ﬁg wasps and that this has left some imprints on the genome of ﬁg wasps during the coevolution history of ﬁgs and ﬁg wasps. Research on the diversity of both bacteria and fungi in ﬁgs are fewer. Our study explored the diversity of endosymbionts in Ficus hirta ﬁgs. We utilized high-throughput sequencing and biological database to identify the speciﬁc microorganism in ﬁgs, then conducted microorganism communities’ diversity analysis and function annotation analysis. As a result, we identiﬁed the dominant endosymbionts in ﬁgs, mainly some insect internal parasitic bacteria and fungi, plant pathogen, endophytes, and saprotroph. Then we also found bacteria in Ficus hirta ﬁgs were more diversiﬁed than fungi, and bacteria communities in female ﬁgs and functional male ﬁgs were different. These ﬁndings may give us more insight into the coevolution and interaction among endosymbiont, ﬁg, and ﬁg wasp.


Introduction
Endosymbiotic microorganisms living in plants and insects are very common in nature, such as plant rhizospheric bacteria, insect gut microbes, and intracellular symbiotic bacteria. According to the previous research, almost all higher plants that have been studied can be found with endosymbionts, so plant endosymbionts prevail in nature [1,2]. For plants, endosymbionts can be found in leaves, petioles, fruits, thorns, seeds, bark branches, and roots, and has rich biodiversity [3,4]. Then it is also very common for insects to live with endosymbionts in their bodies [5]. Intracellular symbiotic bacteria can reside in many different organs and tissues of insects and even live inside the cell [6]. Endosymbionts usually have interactions with their hosts, they can not only cause diseases to the host, but also have many other effects on the host. For instance, plant endophyte can promote vegetative growth of host, increase biomass (yield), and increase the stress resistance of plants [7][8][9], while insect endosymbionts can improve nutrient metabolism of host insects, enhance host stress resistance and immunity [10][11][12], manipulate the reproduction of hosts, and influence insect growth, development, longevity, and evolution [13,14].
Ficus(Moraceae) is one of the largest genera of higher plants which has about 800 species worldwide [15].  and fig wasp have formed a strong obligate mutualism system which is a coevolutionary mode system with a long history and close relationship among animals and plants [16,17]. This mutualism system is a defining model for plant-insect coevolution and interaction researches and contributes greatly to ecosystem functioning, biodiversity, and agriculture [18,19]. The development of next-generation sequencing has led to a surge in effort to characterize the microbiomes of various vertebrate hosts, a necessary first step to determine the functional role these communities play in host evolution or ecology [20]. However, utilizing the next-generation sequencing to explore the endosymbionts in figs is rare in the past. Thus, to find out the  [21]. However, some researches have shown that although the fig was closed, there were still many microbes inside [4,22]

OTU Clustering and Species Classification
The raw bacterial and fungal gene sequencing reads were demultiplexed, qualityfiltered by fastp version 0.20.0 [23] and merged by FLASH version 1.2.7 [24] with the following criteria: (i) The 300 bp reads were truncated at any site receiving an average quality score of <20 over a 50 bp sliding window, and the truncated reads shorter than 50 bp were discarded, reads containing ambiguous characters were also discarded; (ii) only overlapping sequences longer than 10 bp were assembled according to their overlapped sequence. The maximum mismatch ratio of overlap region is 0.2. Reads that could not be assembled were discarded; (iii) samples were distinguished according to the barcode and primers, and the sequence direction was adjusted, exact barcode matching, 2 nucleotide mismatch in primer matching.
Operational taxonomic units (OTUs) with 97% similarity cutoff [25,26] were clustered using UPARSE version 7.1 [25], and chimeric sequences were identified and removed. The taxonomy of each OTU representative sequence was analyzed, the corresponding sequence was respectively allocated to each taxonomic level: Domain, Kingdom, Phylum, Class, Order, Family, Genus, Species. The database used for comparison was as follows. Here the matching software we used was Qiime platform (http://qiime.org/scripts/assign_taxonomy.html, accessed on 10 March 2021), RDP Classifier [27] (version 2.11 http://sourceforge.net/projects/rdp-classifier/, accessed on 10 March 2021), the default confidence threshold was 0.7. Second, we also used the Nucleotide Sequence Database of NCBI (ftp://ftp.ncbi.nih.gov/blast/db/, accessed on 10 March 2021), the comparison method here used was blast, and e-value was default 1 × 10 −5 .

Analysis of Microbial Community Diversity
α-diversity refers to the diversity within a particular region or ecosystem and βdiversity is used to compare diversity between different ecosystems. So, to further explore the diversity and constitution of microbes in figs and the differences of microbial diversity between different gender figs, both α-diversity and β-diversity analyses were implemented. α-diversity of each sample was qualified with Ace and Shannon's diversity index.
Ace index is one of the commonly used indexes for estimating the total number of species in ecology, the index is bigger and the community abundance is higher. Here we use the following algorithm: CAceγ 2 Ace , f orγ Ace < 0.80 and , 0 .
(4) ni = The number of OTUs containing i sequences; Srare = The number of OTUs containing "abund" sequence or less than "abund"; Sabund = The number of OTUs more than "abund" sequence; abund = Advantage OTU threshold. The default value is 10.
The Shannon index is one of the indicators used to estimate microbial diversity in a sample, the bigger the index is the higher the diversity is. The calculation formula is as follows: Thereinto, sobs = The number of OTUs actually observed; ni = The number of sequences contained in the i-th OTU; N = All sequence numbers.
Then inter-group difference test of index was conducted by the Student's t-test and Wilcoxon rank-sum test.
To identify whether the microbe constituents varied between different gender figs, β-diversity was applied for comparison. In the β-diversity analysis (Including PCoA statistical analysis and PERMANOVA analysis), the distance between two pairs of samples needs to be calculated by the statistical algorithm to obtain the distance matrix, which can be used in the subsequent β-diversity analysis and visual statistical analysis. In PCoA statistical analysis we calculated and graphed based on the weighted unifrac-distance matrix [28,29]. In the subsequent PERMANOVA analysis, we used the Bray-Curtis distance matrix to decompose the total variance, analyzed the degree of explanation of sample differences by different grouping factors, and used the permutation test to analyze the statistical significance of the groupings [30,31]. The vegan bag in R (Version 3.3.1) was used for calculation. These operations were performed on the cloud platform (http:// cloud.majorbio.com/, accessed on 10 July 2021), which provided a variety of statistical analysis-related software programs and scripts, we just needed to follow the default steps provided by the cloud platform to operate.

Functional Prediction Analysis
To find out the major physiological types and functions of microbes in figs, we carried out Bugbase phenotype prediction and FaproTax function prediction analysis for bacteria OTUs and FUNGuild function prediction analysis for fungi OTUs [32,33]. After the functional prediction, the corresponding phenotype and functional prediction table were obtained. Then, the functional differences of the microbes from different genders of figs were tested by the Wilcoxon rank-sum test. These operations are also implemented on the cloud platform (http://cloud.majorbio.com/, accessed on 14 July 2021).

Result of OTUs Analysis
Total 685,210 clean reads were obtained in six samples of F. hirta, of which 315,536 and 369,674 were amplified by primers of 799F_1193R and ITS1F_ITS2R respectively. These reads were then aggregated at 97% sequence similarity to generate 867 OTUs, of which 718 and 149 OTUs belonged to bacterium and fungi respectively.

Results of OTUs Taxonomic and Community Species Composition Analyses
We tried to match all obtained OTUs of six samples with the specific taxonomic levels. In total, we identified 25 phyla, 65 classes, 155 orders, 240 families, 389 genera, and nearly 547 species for bacteria; 5 phyla, 16 classes, 37 orders, 76 families, 93 genera, and 109 species for fungi. However, about 78.97% OTUs of bacteria and 46.98% OTUs of fungal could not be annotated to species level, most of OTUs we could only classify into genus, family, order, or phylum level. Hence, for these unclassified microbes OTUs which we could not specify into species level, we designated these OTUs by their annotated genus, family, order, or phylum names.
From Figure 2a and Table S1, we can see that Wolbachia almost only exists in the functional male figs sequence library with 64.32% ± 26.90% abundance, while in the female figs library the abundance is 0.17% ± 0.16%. Ralstonia solanacearum and Burkholderia gladioli's sequences mainly exist in female figs with 37.37% ± 27.20% and 28.39% ± 48.35% abundance, but for functional male figs, abundances are 1.83% ± 1.24% and 6.29% ± 10.84%. Then, Rhodococcus erythropolis and Enterobacteriaceae exist in all six samples' sequence libraries, with 6.95% ± 7.26% and 6.15% ± 14.62% abundance in all libraries. But Enterobacteriaceae has a prominent quantity merely in Fhm_3, which is 35.99%, and in the rest of the groups, the abundance is 0.18% ± 0.22%. Moreover, Burkholderia-Caballeronia-Paraburkholderia and Rickettsia bellii also exhibit distribution bias between functional male and female figs, of which Burkholderia-Caballeronia-Paraburkholderia has 5.11% ± 0.76% abundance in female figs libraries and 0.62% ± 0.50% abundance in functional male figs. Rickettsia bellii only exists in Fhm_1 and Fhm_2 with 4.48% and 3.10% abundance.  (Figure 2b and Table S2).  In terms of fungus, Fusarium concentricum, species of Capnodiales, Simplicillium, Nectriaceae, Ascomycota, and Sctophoma are major fungi in all samples sequence libraries, with 28.72% ± 28.67%, 22.33% ± 32.36%, 16.61% ± 39.32%, 8.11% ± 11.07%, 11.29% ± 24.81%, and 2.77% ± 6.78% abundance in order. Of the above-listed fungi, only Ascomycota have detected sequences in all six samples, while others are absent at some samples. Fusarium concentricum has 47.47% ± 26.52% abundance in female figs sequence library, and 9.96% ± 17.21% abundance in functional male figs, and Capnodiales has 23.23% ± 40.14% and 21.43% ± 31.68% abundance in female and male figs respectively, while Simplicillium has 33.21% ± 55.12% in functional male figs and only one sequence is detected in female figs. (Figure 2b and Table S2).  Table 1). For the diversity indexes between female and functional male figs, there is no significant difference for both Shannon's diversity index and Ace index ( p-value 0.1273 0.0002 ** p-value < 0.01 means difference is extremely significant, and we marked the extremely significant p-value of the differences with **.

of 16
abundance in bacteria is higher ( Figure 3; Table 1). For the diversity indexes between female and functional male figs, there is no significant difference for both Shannon's diversity index and Ace index (  Table 2). The similarity of species between female and functional male figs was further calculated by PCoA analysis (Figure 4a,b). Bacteria are disparate in two types of figs; however, fungi are overlapped. The results of PERMANOVA analysis between different groups are similar to that of PCoA (Table S3). For bacteria, the R2 value between female and functional male figs is the highest (R2 = 0.5908), showing bacteria community divided by gender is more reliable than any other grouping modes, while for fungi it seems that communities do not vary with the fig's gender.    The similarity of species between female and functional male figs was further calculated by PCoA analysis (Figure 4a,b). Bacteria are disparate in two types of figs; however, fungi are overlapped. The results of PERMANOVA analysis between different groups are similar to that of PCoA (Table S3). For bacteria, the R2 value between female and functional male figs is the highest (R2 = 0.5908), showing bacteria community divided by gender is more reliable than any other grouping modes, while for fungi it seems that communities do not vary with the fig's gender.

3, x 8 of 16
Mean is mean, Sd is standard deviation, p-value is the false positive probability value, and Qvalue is the FDR value.

Result of Functional Prediction Analysis
The comparison of phenotype for bacteria between female and functional male figs is shown in Figure 5. There are five phenotypes with a slightly significant larger proportion in female figs: Gram-positive, facultatively-anaerobic, stress-tolerant, potentially pathogenic, and contain-mobile-elements, with the first phenotype's prominent group being Rhodococcus and the last four phenotypes being Ralstonia and an unclassified genus of Enterobacteriaceae; while two phenotypes with slightly significant larger proportion in functional male figs are aerobic and Gram-negative, and Wolbachia is the dominant group in both of them. According to the phenotype, the functions of functional male and female figs microbial communities are also different with slight significance with more proportion of plant-pathogen in female figs and more proportion of intracellular-parasites in functional male figs (Figure 5a,b).
Fungi in figs are mainly animal or plant pathogens and soil, wood, and undefined Saprotroph, but about 34.81% of fungi's physiological types were unknown ( Figure 6). As preceding results showed that there was no difference of fungi communities in functional male and female figs, and the functional annotation results of fungi are not that specific, so we did not conduct the Wilcoxon rank-sum test for fungi function difference between different gender.

Result of Functional Prediction Analysis
The comparison of phenotype for bacteria between female and functional male figs is shown in Figure 5. There are five phenotypes with a slightly significant larger proportion in female figs: Gram-positive, facultatively-anaerobic, stress-tolerant, potentially pathogenic, and contain-mobile-elements, with the first phenotype's prominent group being Rhodococcus and the last four phenotypes being Ralstonia and an unclassified genus of Enterobacteriaceae; while two phenotypes with slightly significant larger proportion in functional male figs are aerobic and Gram-negative, and Wolbachia is the dominant group in both of them. According to the phenotype, the functions of functional male and female figs microbial communities are also different with slight significance with more proportion of plant-pathogen in female figs and more proportion of intracellular-parasites in functional male figs (Figure 5a,b).  Fungi in figs are mainly animal or plant pathogens and soil, wood, and undefined Saprotroph, but about 34.81% of fungi's physiological types were unknown ( Figure 6). As preceding results showed that there was no difference of fungi communities in functional male and female figs, and the functional annotation results of fungi are not that specific, so we did not conduct the Wilcoxon rank-sum test for fungi function difference between different gender.

Ficus Hirta Figs Microbial Composition
According to microbial OTUs clustering results, we obtained 718 OTUs for bacteria and 149 OTUs for fungi. Although more sequences of fungi were produced than bacteria, bacteria sequences aggregated much more OTUs than fungi. This may show that bacteria in figs are more diversified than fungi.
For Ficus hirta, in functional male figs, Fusarium concentricum, fungi of Capnodiales and Ascomycota, and bacteria of order Rickettsiales (Including Wolbachia and Rickettsia bellii) are dominant communities; while in female figs, family Burkholderiaceae bacteria (Including Ralstonia solanacearum, Burkholderia gladioli and Burkholderia-Caballeronia-Paraburkholderia) and fungi of Simplicillium, Capnodiales, and Nectriaceae are dominant communities (Figure 2). Wolbachia (Anaplasmataceae) is a maternally inherited endosymbiont that mainly exists in arthropods and filarial nematodes [34,35]. Past research has manifested that Wolbachia can infect about 40% of arthropod species, Wolbachia is likely to be the most abundant endosymbiont among arthropods [36,37]. In addition, previous surveys have also shown that the incidence of Wolbachia in fig wasps is up to 59-67%, which is remarkably higher than that in other insects [38].  (Table S1). Rickettsia is a facultative eukaryotic intracellular symbiotic bacterium belonging to the subgroup Rickettsiaceae of Proteobacteria, similar to insect endosymbiont Wolbachia, Rickettsia is associated with reproductive ma-

Ficus Hirta Figs Microbial Composition
According to microbial OTUs clustering results, we obtained 718 OTUs for bacteria and 149 OTUs for fungi. Although more sequences of fungi were produced than bacteria, bacteria sequences aggregated much more OTUs than fungi. This may show that bacteria in figs are more diversified than fungi.
Wolbachia (Anaplasmataceae) is a maternally inherited endosymbiont that mainly exists in arthropods and filarial nematodes [34,35]. Past research has manifested that Wolbachia can infect about 40% of arthropod species, Wolbachia is likely to be the most abundant endosymbiont among arthropods [36,37]. In addition, previous surveys have also shown that the incidence of Wolbachia in fig wasps is up to 59-67%, which is remarkably higher than that in other insects [38].  (Table S1). Rickettsia is a facultative eukaryotic intracellular symbiotic bacterium belonging to the subgroup Rickettsiaceae of Proteobacteria, similar to insect endosymbiont Wolbachia, Rickettsia is associated with reproductive manipulation in host insects and the 16S rDNA of Rickettsia was 86% similar to that of Wolbachia as well [39,40]. Both of them can affect the reproduction of the host by parthenogenesis induction and killing of male progeny from infected females. Recent research show that Rickettsia has an effect on the environmental fitness of host insects and it can also enhance the host insect's ability to resist environmental adversity [40][41][42]. Bacteria of family Enterobacteriaceae have a large amount in Fhm_3, second only to Wolbachia (Figure 2a). Enterobacteriaceae are widely distributed and its host range is large including people, animals, plants, where they can establish themselves as parasites, symbionts, epiphytes, or saprophytes. Moreover, they can also live in soil or water. The previous study has also found Enterobacteriaceae bacterium is the dominant bacteria flora in Ficus hispida fig wasp [43].
Both Burkholderia gladioli and Ralstonia solanacearum belong to the family Burkholderiaceae, including some other bacteria of the same family, and they are prominent and dominant bacterial groups in female figs, while their numbers are relatively low in functional male figs. Family Burkholderiaceae is characterized by the presence of ecologically extremely diverse organisms and contains truly environmental saprophytic organisms, phytopathogens, opportunistic pathogens, as well as primary pathogens for humans and animals [44]. B. gladioli was initially known as a plant pathogen, but currently, B. gladioli is isolated not only from plants [45], but also found in diverse habitats, including soil, environmental water [46,47], and even the respiratory tract of immunosuppressed humans [48,49]. Studies have shown that B. gladioli often symbioses with fungi and plants as well [50]. B. gladioli is widely regarded as a germ that exhibits significant ecological niches divergence even within species [51], so it exists in all fig specimens with remarkable divergence in species abundance. Ralstonia solanacearum is a Gram-negative soil-borne pathogen that can cause bacterial wilt disease and lead to destructive losses of some economic crops, such as potato, eggplant, tomato, peanut, and tobacco [52]; it is a common plant pathogen.  [53]. F. concentricum has a wide host range, associating with multiple diseases on different hosts such as stem rot of Paris polyphylla var. chinensis and fruit rot of pepper and banana [54]. Fusarium fungi is a kind of important plant pathogenic fungi, widely distributed in nature [55]. Maybe just because of this, F. concentricum prefers existing in female figs and relatively exists less in functional male figs which are full of galled flowers (Figure 2b). The Capnodiales incorporate plant and human pathogens, endophytes, saprobes, and epiphytes, with a wide range of nutritional modes; several species are lichenized or occur as parasites in fungi or animals [56]. These biological characteristics may contribute to the distribution pattern of Capnodiales in Ficus hirta figs, both female and functional male figs have Capnodiales fungi without regularity. Moreover, the Simplicillium species are commonly found in soil, seawater, rock surface, decayed wood, air, and as symbiotic, endophytic, entomopathogenic, and mycoparasitic fungi, while a few insect-associated species have also been reported [57]. We can find that Simplicillium almost only existed in functional male figs (Table S1), this may indicate that the Simplicillium fungi we found in Ficus hirta figs are associated with fig wasps.

Microbial Diversity of Ficus Hirta Figs
The above discussion presents that in different genders of figs the dominant microbes are different but the α-diversity analysis reveals that there is no significant difference between female and functional male fig microbial communities in diversity and species abundance ( Table 2). Student's t-test conducted between fungi and bacteria for Shannon's diversity index and Ace index shows that bacteria Ace index is remarkably higher than fungi while the difference of Shannon's diversity index between bacteria and fungi is not conspicuous (Table 1). Relatively speaking, F. hirta figs bacteria community has high species abundance and higher uniformity of species distribution than fungi, but this kind of discrepancy does not exist between different gender fig endosymbiont communities.
Results of the β-diversity analysis show that bacteria communities differ with figs' gender whereas fungi communities in different gender figs are similar (Figure 4). Meanwhile, PERMANOVA analysis also confirms that the differences between the two bacteria groups are best explained by gender as a grouping factor, however, this trend is not distinct for fungi communities (Table S3). So far, with the addition of previous microbial species composition analysis, we can firmly conclude that in Ficus hirta figs bacteria are more abundant and diversified than fungi, and bacteria communities are different between female and functional male figs.
Differences of bacteria communities between different gender figs may be due to the large amounts of endosymbionts like Wolbachia, Rickettsia bellii or Enterobacteriaceae coexisted in fig wasp larvae. First, a large quantity of fig wasp symbiotic bacteria almost merely exist in functional male figs have caused the difference in bacteria communities between female and functional male figs; then, insect symbiotic bacteria can induce the host to produce antimicrobial peptides and defensins or symbiotic bacteria can secrete antibiotics to resist the infection of pathogenic microorganisms [58,59], and this may result in relatively less plant bacterium pathogen in functional male figs compared to that in female figs (Figure 5a

Functional Prediction Analysis
We analyzed figs endosymbionts from facets including species constitution and abundance, community structure, and diversity; perhaps we can get new points of view by analyzing the microbial function and physiological phenotype. In functional male figs, there are mainly some Gram-negative, aerobic, and intracellular parasite bacteria, this reflects the prevalence of Wolbachia, Rickettsia, Enterobacteriaceae such as Gram-negative bacteria in functional male figs ( Figure 5). We can also find that aerobic such as phenotype is mainly contributed by Wolbachia, and  (Figure 5a,b), and these phenotypes are mainly contributed by Ralstonia solanacearum according to BugBase phenotype prediction results (Figure 5c), as Enterobacteriaceae only appears in functional male figs (Figure 2a).
Fungi in figs are mainly animal pathogen, endophyte, lichen parasite, plant pathogen, soil saprotroph, and wood saprotroph; but many are still functionally unknown ( Figure 6). Then it is also worth noting that animal pathogen is prominently rich in Fhm_1, and species annotation results show that the Simplicillium fungi make up 96.86% proportion in Fhm_1 (Table S2). Accordingly, we can speculate Simplicillium may be a type of endophyte in the  [21,64,65]. But we guess that interaction between fig wasp and endosymbionts such as Wolbachia and Rickettsia can increase host resistance to diseases or adverse factors [41,42,[66][67][68][69], may also have a role in such gene family contraction.

Conclusions
Ficus hirta is a dioecious plant in which female trees only grow female inflorescence that can bear seeds and male trees grow functional male inflorescence with galled flowers that are parasitized by fig wasps. Such special structure and function of different gender figs would result in different endosymbionts contained in figs. Our research demonstrated that bacteria involved in different gender figs were different but the bacteria community diversity was similar. In gross samples of F. hirta figs, bacterial diversity was higher than fungi. Although we did not identify the significant difference of fungi communities between female and functional male figs, we found the distribution of a special fungi Simplicillium showed bias in different gender figs. If we increase the number of samples, we can probably find that fungi communities vary with the figs' gender.
Our research found out the main endosymbionts existed in F. hirta figs, but more samples and deeper and broader sequencing are required to discover new specific species.