Genome Sequencing and Evolutionary Analysis of Marine Gut Fungus Aspergillus sp. Z5 from Ligia oceanica

Marine-derived metabolites continue to be a prolific source of bioactive natural products with a high tendency to become drug candidates. ¹ Research interests in secondary metabolites (SMs) from marine-derived fungi have increased in recent years, as many of them are structurally unique and possess interesting biological and pharmacological properties. ² Various microbes, including fungi, are found in the gut of marine isopods^3–6 and may serve as food or defense for marine isopods by producing SMs. ⁷ While studies have shown that gut microbes exist in the worldwide distributed genus Ligia, ⁸ knowledge of their SMs is scanty. Gut-derived fungi from Ligia may be a prolific source of bioactive marine natural products. In this study, a fungus was obtained from the gut of the marine isopod Ligia oceanica and was identified as Aspergillus sp. based on the analysis of the ribosomal internal transcribed spacers and the 5.8S rRNA gene sequence. This strain of Aspergillus sp. Z5 has been deposited in the China Center for Type Culture Collection with accession number CCTCC M 2015238.

The genus Aspergillus is the major contributor to new bioactive SMs of marine fungal origin. ² Therefore, they are recognized as a prolific source of pharmacologically valuable new compounds. ⁹ We also obtained many compounds with new structure and bioactivity from Aspergillus sp. Z5 (unpublished data). The genome sequencing of Aspergillus sp. Z5 from the gut of marine isopod L. oceanica may provide fundamental molecular information on elucidating the metabolic pathway of SMs in this species.

For genome sequencing, Aspergillus sp. Z5 was cultured in glucose minimal media with 30 g/L sea salt and incubated at 30 °C, 200 rpm, for 24 hours. Isolation of genomic DNA isolation was performed as described by Green and Sambrook. ¹⁰ The genome of Aspergillus sp. Z5 was sequenced using the Illumina HiSeq 2000 technology at the Shanghai Majorbio Bio-pharm Technology Co., Ltd. A library with a fragment length of 300 bp was constructed, and a total of 24,591,619 paired-end reads were generated. Approximately 23,727,930 high-quality reads, which provided a 137.5-fold depth of coverage, were assembled with SOAPdenovo version 1.05. ¹¹ Based on the assembly, the genome size was estimated to be 33.8 Mbp with a GC content of 50.5% (Table 1). The assembly was organized in 1,089 contigs, which are linked by paired-end reads into 246 scaffolds (>1,000 bp). The average base is found in a scaffold of N₅₀ size 1,256,305 bp and a contig of N₅₀ size 195,844 bp (Table 1). This whole-genome shotgun project has been deposited at DDBJ/EMBL/GenBank under the accession no. LDZW00000000. The version described in this paper is LDZW01000000.

OPEN IN VIEWER

Table 1.

Comparison of genome characteristics.

ATTRIBUTE	ASPERGILLUS SP. Z5	A. NIDULANS FGSC A4*	A. ORYZAE RIB40*	A. FUMIGATUS AF293*
Assembly size (bp)	33.8 M	30.5 M	37.9 M	29.4 M
Predicted ORFs	11791	10687	12090	9840
GC content (%)	50.5%	50.0%	46.3%	50.0%
Gene/genome (%)	55%	50%	45%	49%
Contigs	1089
Contig N50 (bp)	195,844
Contig N90 (bp)	37,806
Scaffolds (>1000 bp)	246
Scaffold N50 (bp)	1,256,305
Scaffold N90 (bp)	381,266

Note: *Data are from the calculation based on the database of AspGD (www.aspgd.org/).

Protein-coding sequences were predicted by Augustus 2.5.5 and annotated using BLAST searches of nonredundant protein from the NCBI database. The genome of Aspergillus sp. Z5 encodes 11,791 predicted proteins, and the total length of genes was 18.7 Mbp, which makes up 55.3% of the genome (Table 1). The ratio of gene/genome in Aspergillus sp. Z5 is higher than that in three known Aspergillus species (Table 1). The rest of the noncoding genomic sequences are made up of intergenic sequences (including introns, promoters, and terminators), noncoding RNAs (rRNAs and tRNAs), origins of DNA replication, centromeres, and telomeres. Of the total predicted genes, 11,178 genes (94.8%) encode known function proteins, 210 genes (1.8%) are considered to encode hypothetical proteins, and 403 (3.4%) genes have no match in the database.

Recently, molecular analyses of the genus Aspergillus have offered a better insight into their taxonomic and phylogenetic relations.^12,13 Phylogenetic relationships between filamentous fungi have often been based on ribosomal DNA sequences or single-gene families,^14,15 such as the genes encoding β-tubulin (benA), small rRNA subunit (rns), cytochrome oxidase subunit I (cox1), and RNA polymerase II second largest subunit gene (rpb2). Besides the ribosomal sequence analysis, we further performed multilocus sequence analysis of the four housekeeping genes (benA, rns, cox1, and rpb2) ¹³ from Aspergillus sp. Z5 and 23 Aspergillus species. The DNA sequences of these genes in each Aspergillus species were obtained from GenBank database and aligned with ClustalW for further analysis. The maximum-likelihood and neighbor-joining trees were generated using the MEGA 6 software with default settings. The phylogenetic analysis supported that Aspergillus sp. Z5 falls within the A. versicolor clade (Fig. 1A). Compared with the genome sequences available in the AspGD (http://www.aspgd.org/), the longest scaffold in the genome of Aspergillus sp. Z5 was 3.7 Mb in length and shared a high degree of colinearity with the reference chromosome of A. sydowii (Fig. 1B), a closely related species to A. versicolor.^16,17

OPEN IN VIEWER

Figure 1.

The evolutionary analysis of the genome of Aspergillus sp. Z5. (A) The phylogenetic position of Aspergillus sp. Z5 based on multilocus sequence analysis of the four housekeeping genes (benA, rns, cox1, and rpb2). The maximum-likelihood tree was generated using the MEGA6 software, and bootstrap values are shown as percentages (values below 60% are omitted). (B) The alignment of the largest scaffold in the Aspergillus sp. Z5 assembly was aligned against the genome of A. sydowii using BLASTN and visualized using the Artemis Comparison Tool.

In order to predict the clusters involved in SM biosynthesis, the reported genome was analyzed using antiSMASH pipeline. ¹⁸ As shown in Figure 2, 89 SM biosynthetic gene clusters and partial clusters were predicted in Aspergillus sp. Z5, including 10 nonribosomal peptide synthetase (NRPS) clusters, 10 polyketide synthetase (PKS) clusters, 8 terpene clusters, 2 PKS/NRPS hybrid clusters, 1 terpene/PKS cluster, and 58 clusters designated putative or other (Fig. 2). In total, there are 2,018 genes related to SM production (Supplementary Table 1, Fig. 2). Although the types of SM clusters could be accurately predicted according to the core SM biosynthetic genes encoding backbone enzymes, it is still not possible to predict with accuracy the boundaries of SM gene clusters or the functions of some clusters without backbone enzymes. ¹⁹ This is due to the fact that many of the genes surrounding the core SM biosynthetic genes often have unknown functions, making predictions of their involvement in the biosynthetic process of the SM almost impossible. ¹⁹ Based on the number of similar cluster genes and the identity of cluster genes (cutoff criteria: identity of 40% and query coverage of 50%) with other SM biosynthetic gene clusters, the types of most putative clusters without backbone enzymes were predicted in this study (Supplementary Table 1). However, the elucidation of these SM biosynthetic gene clusters is mainly dependent on experimental verification, followed by identification and characterization of SMs produced by the deletion strains.

OPEN IN VIEWER

Figure 2.

Graphical map of the putative SM gene clusters and the related gene numbers in the genome of Aspergillus sp. Z5.

The increasing availability of Aspergillus genomes has led to a rapid identification of secondary metabolism biosynthetic pathways (SMBPs) in recent years. However, only a small part of SMBPs are conserved between even closely related species. ²⁰ In this work, comparative analysis of SM cluster genes encoding backbone enzymes was performed between Aspergillus sp. Z5 and the other three well-annotated Aspergillus species (Aspergillus nidulans, Aspergillus oryzae, and Aspergillus fumigatus), revealing that some backbone enzymes are highly conserved among all these four Aspergillus species, including cluster 11 and 28 (siderophore), cluster 30, 32, 42, 64, and 86 (unknown product), and cluster 35 (neosartoricin and fumicycline A). Besides these conserved backbone enzymes among the four species, Aspergillus sp. Z5 shared the greatest number of backbone enzymes with A. nidulans, including cluster 2 (emericellin), cluster 15 (orsellinic acid/F9975/violaceols), cluster 17 (asperthecin), cluster 63 (asperfuranone), cluster 79 (terrequinone), and other clusters with unknown products (Supplementary Table 1). Much less backbone enzymes are conserved in A. oryzae or A. fumigatus (Supplementary Table 1). Although the products of some clusters in Aspergillus sp. Z5 were predicted based on the conserved sequences of SMBP backbone enzymes in genus Aspergillus, the functions of the rest of these 89 clusters remain unknown. In addition, some backbone enzymes with unknown functions were found in distantly related fungi such as Penicillium, Metarhizium, and Tolypocladium, but not in other Aspergillus species (cutoff criteria: identity of 40% and query coverage of 50%), ie, clusters 8, 14, 31, and 48, which are highly possible to be new compounds in Aspergillus species (Supplementary Table 1).

To the best of our knowledge, this is the first genome sequence of Aspergillus species isolated from the gut of marine isopod L. oceanica. Due to the limited knowledge and data about the marine isopod, it is currently impossible to explore the relationship between SMs from the Aspergillus species and its host. However, availability of this genome sequence presents a basis for exploring new bioactive compounds and elucidating the evolution of the metabolic pathway of SMs in marine-derived Aspergillus species.