Ribosomal Protein Cluster Organization in Asgard Archaea

1. Introduction

Ever since the initial discovery of Archaea by Woese and Fox [1], the number of recognized lineages comprising the Archaea domain has rapidly increased. Currently, five superphyla (and 35-38 phyla) of Archaea are described in the NCBI database [2–4]. Of these, the recently discovered Asgard include several phyla and are the most diverse [5, 6]. To emphasize this, they have been referred to as a “superphylum” [7].

Amongst the three domains, there are approximately 102 recognized ribosomal protein (r-protein) families [8–11]. Essentially universal amongst these are seventeen large subunit and nineteen small subunit r-proteins. Many of these 36 r-proteins likely had an ancient origin possibly before the last universal common ancestor (LUCA) [10–16]. However, this conclusion is dependent on how the gene content of LUCA is identified. At least eight of these genes are found in multiple versions of LUCA (Rivas personal communication). The Archaea and the Eukarya also share eleven large subunit (LSU) r-proteins and twenty small subunit (SSU) r-proteins. However, neither group shares additional r-proteins with the Bacteria. For this reason, it has been hypothesized [17, 18] that the eukaryotic translation system originated from an earlier archaeal version and was subsequently expanded.

In Bacteria, approximately 32 r-proteins from both subunits and two translation-related proteins are grouped into seven well-studied clusters. They are, viz., alpha, beta/L10, L11, S10, S20, str, and the spc clusters [19, 20]. In E. coli and other well-studied organisms, these clusters have been experimentally shown to be operons [21]. In general, we will refer to them as clusters with the implication that many of them could likely be operons. The S10 and the spc operons (clusters) are the largest, each typically encoding ten-twelve r-proteins [21, 22]. Gene expression when known is frequently regulated by one of the r-protein components within the cluster. Besides these clusters, there are several smaller clusters as well. The r-protein clusters are widely conserved amongst bacterial species and remnants, and Bacteria-like clusters are also found in some chloroplasts and mitochondria [23]. However, it is not clear if the clusters seen in mitochondria are functional. If they are not, then their presence or absence is likely a mere reflection of their nonessentiality.

The universality of r-protein gene clusters is often linked with a long history of their association. They also represent ancient/regulatory relationships, with implications for early gene clusters in mini chromosomes. Eukaryotic transcription usually involves a single transcript. Thus, eukaryotic operons/clusters of genes are restricted to few organisms [24, 25]. While eukaryotes do have meaningful gene clusters, these clusters do not include r-protein genes [24, 26–29].

Earlier studies of non-Asgard Archaea showed that the universal r-protein genes are in clusters similar or identical to those found in Bacteria [20]. While three archaeal r-protein genes S4e, L32e, and L19e were found associated with the archaeal version of the spc operon, the nonuniversal archaeal r-protein gene L18e was part of the conserved L13 cluster. 17 r-protein genes were part of one of the ten previously unrecognized gene clusters. These clusters were found to be associated with genes involved in the initiation of protein synthesis, transcription, or other cellular processes. Since such associations in the universal clusters could not be found, it was posited that the ribosome had its own independent line of evolution [20].

It has been clear since their initial discovery that archaeal ribosomes more closely resemble those of Eukarya than Bacteria as has been documented in detail [30–32]. However, the non-Asgard Archaea have not been shown to have any strong specific relationship to Eukarya, and the matter was not initially aggressively pursued. Efforts to understand the relationship between the domains of life were rekindled when it was discovered that the then newly discovered Asgard Archaea have many nonribosomal features that are shared with Eukarya [7, 31, 32]. This has fostered the hypothesis that Eukarya may have descended from the Archaea [33]. Asgard Archaea were proposed to represent a historical link or bridge between Archaea and Eukarya [34–41]. A recent report has suggested that the genomic material is condensed and spatially distinct from the riboplasm within certain Loki- and Heimdallarchaeota cells, as further proof of the role of Asgard Archaea in eukaryogenesis [42]. Amongst the Asgard group, the Heimdallarchaeota have been proposed to be the closest to the Eukarya [43, 44]. Other analyses based on multiple gene sequences placed the Asgard between the TACK group of Archaea and Eukarya [7, 36, 40]. In contrast, one branch of the Asgard, namely, the Lokiarchaeota, and their close relatives were proposed to be closer to the Euryarchaeota, than to Eukarya [45].

Genes that are translated together frequently share similar origins, physical interactions, functions, or regulatory mechanism(s) [46]. The arrangement of genes in genomes is thus a window to understand how organisms are related. Assuming there was a universal ancestor of the three domains, it likely had a gene order/arrangement that possibly was a precursor of operons. Its evolutionary origins may have dated back to four billion years ago [47–51]. While some portions of this gene order underwent shuffling over evolutionary timescales [52, 53], the arrangement of some gene clusters continues to be conserved [54]. In particular, the translation machinery and the genes that encode the same are highly conserved. Their association frequently has had a long evolutionary history [55–58]. The origin of the ribosome has been posited to be strongly coupled with the early history and even origins of life [56, 59–63], which is further supported by recent papers by Bose et al. [64] and Bose et al. [65].

In order to gain further insight into the r-protein gene clusters in Asgard Archaea, the genome order of r-protein coding genes was analyzed and compared with non-Asgard archaeal and bacterial genomes.

2. Materials and Methods

3. Results

A total of ~552 Asgard genomes grouped under the various phyla that were available on the NCBI server (https://www.ncbi.nlm.nih.gov/datasets/genome/?taxon=1935183) as of July 2023 were examined [36, 73]. Of all the Asgard genomes, only thirteen occur as a single complete (scaffold) genome(s) (Table 1). They will henceforth be referred to as MKD1, FW102, B-35, bin132, bin108 (Lokiarchaeota), LCB4 (Odinarchaeota), bin27 and bin8 (Thorarchaeota), PM71, PR6, bin6, bin76, and bin272 (Heimdallarchaeota). The next genomes that are of reasonably good quality (assembled into less than 20 contigs) with annotated genes/proteins are those of the Thorarchaeote strains FW25 (twelve contigs) and BC (nineteen contigs) and the Heimdallarchaeote strains AC18 (eleven contigs) and SZ_4 bin2.246 (sixteen contigs). They will henceforth be referred to as FW25, BC, AC18, and bin2.246, respectively. These genomes provide useful information for some of the clusters of interest but typically not all.

The remaining Asgard genomes each consist of at least twenty contigs and frequently many more. Given the incompleteness of these genomes, the r-protein coding genes in a given cluster were often found in multiple contigs. Some of them, as illustrated in Supplementary Figures 1a-b, either end or begin a contig. The genomes (contigs) of many other genomes have not been annotated into ORFs. Hence, their gene order is not available and therefore was not considered when mapping the clusters. In summary, only the complete genomes and those available with less than twenty contigs, as defined above, were used for the comparative analysis of r-protein gene arrangement. Thus, the choice of non-Asgard archaeal genomes for representation in the figures for comparison is based on the four major groups classified under Archaea [74], viz., the Asgard, the DPANN, the TACK, and the Euryarchaeota [2, 7, 34, 36, 40, 41, 75–78]. Furthermore, the phylogenetic relatedness as described in the literature was also taken into account [7, 34, 37, 39, 41, 79]. E. coli and B. subtilis were used as representatives of Bacteria.

4. Discussion

Classical operons are found to occur in both Bacteria and Archaea [84, 85]. The conservation of the arrangement of the r-protein gene cluster is in congruence with how they are conserved in sequence [21, 22, 55]. A large chunk of 14 genes from the most highly conserved S10-spc cluster of r-protein coding genes was thus found to be similarly conserved in arrangement, whether it was Bacteria, Asgard, or non-Asgard Archaea. The arrangement of the four r-protein coding gene clusters as observed in the 17 Asgard archaeal genomes (Lokiarchaeota, Odinarchaeota, Heimdallarchaeota, and Thorachaeota) suggests that apart from the big cluster (S10-spc), the arrangement/order of the rest of the r-protein clusters is not conserved and is dispersed. In the context of one section of the Alpha operon found in the vicinity of the S10-spc cluster, the conservation of the location resembles what is observed in bacterial genomes. Assuming these Asgard genomes have been correctly assembled, a portion of the Alpha operon (L18e) cluster associated with the highly conserved and universal S10-spc cluster is likely a reflection of their association despite the phylogenetic distance between Bacteria and Asgard Archaea.

It has been suggested that the Heimdallarchaeota group is the most likely sister group of Eukarya [7, 36, 40, 43, 44]. The dispersive nature of the str/L30e cluster order and the unique arrangement of the S24e-S27ae, the L7ae, the L31e, and the L12-L10-L11-L1 clusters are particularly evident in the case of the Heimdallarchaeote genomes. The unique arrangement of the r-protein clusters, as observed in the case of the Heimdallarchaeota members of the Asgard in this study, may corroborate this. This could possibly be a marker for evolutionary divergence within the Asgard genomes. Additional complete Asgard genomes, when available, may help understand such divergences better. Examination of the plastid genome of A. thaliana and the gene order of (S10-spc cluster)-L36-S11 shows similarities with the order in Bacteria (E. coli, B. subtilis), though a section of this cluster is missing from the same (data not shown).

The dispersion of gene clusters in eukaryotes has been suggested to be directly correlated with higher rates of chromosomal/genome evolution [26, 86]. It remains to be seen if Asgard Archaea show a higher rate of genome evolution as compared to the non-Asgard Archaea.

Genome organization seldom remains in stasis with the phenomenon of lateral gene transfer constantly changing the dynamics of the same. It has been hypothesized that the Asgard proteome, particularly the eukaryotic signature protein coding genes, evolved through extensive horizontal gene transfer [36]. It is not clear from the few available complete Asgard genomes if the arrangement of the r-protein clusters is a result of this phenomenon. Nevertheless, the roles of gene order or arrangement in fashioning cellular physiology and genome evolution remain to be explored further [87].

It should be noted that the arrangement of the r-protein coding gene clusters in the Asgard could be in some cases an artifact of poor genome completion. The occurrence of several partial genes, a universal gene annotated as a pseudogene, and several universal gene segments missing in the Asgard genomes in this study (Figures 1–5) suggests that the genomes in which they are found may have been incorrectly assembled. The noncontiguous/disjointed occurrence of the otherwise conserved segment of 14 cooccurring genes of the S10-spc cluster in the three Heimdallarchaeotes (bin6, bin272, and bin76) (Figure 1(a)) and the absence of three of the most universal and highly conserved genes uL3-uL4-uL23 from the genomes of Lokiarchaeotes bin132 and bin108 (Figure 1) suggest that the assembly of these genomes is likely not accurate. The incompleteness or misassembly of genomic contigs has been observed previously [88–92]. In fact, binning errors were implicated in the absence of r-protein genes from metagenome-assembled genomes (MAGs), and hence, analysis of genomics and phylogeny of uncultivated microbes, using MAGs, may be fraught with erroneous interpretations [93, 94].

However, we did observe one interesting feature in a non-Asgard archaeal MAG. Of the three non-Asgard archaeal genomes assembled from metagenomes that are available in the NCBI database [88], only the genome of DPANN archaeon GW2011_AR10 (GenBank accession: CP010424) is complete as a single scaffold/chromosome. Examining the order/arrangement of the r-protein gene clusters in this genome showed that four of the five clusters reported in this paper cooccur contiguously. The arrangement of these clusters in this genome is as follows: str-L30e cluster (rpoH-rpoB-HP-rpoA1-rpoA2-L30e-nusA-S12-S7-fusA-EF(1a)-S10)-5HPs-S10-spc cluster (L3-L4-L23-L2-S19-L22-S3-L29-IF(Sui)-RNP1-S17-L14-L24-S4e-L5-S14-S8-L6-L32e-L19e-L18-S5-L30-L15-secY)-HP-L34e-L14e-L18e (Alpha operon) cluster (S13-S4-S11-rpoD-tRNA(leu)-L18e-L13-S9-rpoN-S2)-HP-gltX-HP-L7ae cluster (L7ae-S28e-L24e-ndk-dUTP diphosphatase-infB-S6e-eIF2g-Utp24-rpoE1-rpoE2-S24e-S27ae-10 ORFS-gcp-S15-HP-S3ae). The L31e and L11-L1-L10-L12 clusters are not part of this section and occur separately as two distinct sections elsewhere in the genome. Only in B. subtilis (Bacteria), a partially similar feature is observed with the L11 cluster found immediately upstream of the S10-spc cluster.

The second non-Asgard MAG, that of DPANN Candidatus Micrarchaeum acidiphilum ARMAN-2 (GenBank assembly: GCA_009387755.1) is in 8 contigs, with the r-protein gene clusters split and not conserved as reported for some of the Asgard genomes in the paper. The third non-Asgard MAG, that of DPANN Candidatus Forterrea multitransposorum archaeon (GenBank accession no.: CP045477), has not been annotated into ORFs (open reading frames)/genes etc. Thus, these sequences derived from metagenome-assembled genomes may not be sound.

It is not clear if the cooccurrence of four r-protein clusters in the MAG of DPANN archaeon GW2011_AR10 is a result of the accuracy of the assembly and annotation/curation of the genome. If it is indeed an accurate assemble, then the assembly of other genomes in the database may need to be revisited for improving better quality which may help in an improved understanding of arrangement of these clusters. Thus, in the absence of good quality genomes, ascertaining the order of highly conserved genes of r-proteins can be challenging, and thus, well-curated genomes are vital towards understanding genome organization. The importance of a more robust curation of MAGS as well publicly available Asgard genomes [6, 95] cannot be emphasized enough to get a better picture of the genome organization of the same.

5. Conclusions

The discovery of the Asgard Archaea has redefined our understanding of the phylogenetic positions of the various groups that comprise the Archaea. The genomic arrangement of the r-protein coding genes in the Asgard superphylum within the domain Archaea suggests that the order may follow the phylogeny. The closer an Asgard is to eukaryotes, the more dispersed is the arrangement of the r-protein gene clusters. This feature suggests a possible ancient gain or loss of the operon strategy of gene regulation in the early history of life. Given that very few Asgard genomes are currently complete, new data and better means of genome assembly will facilitate a better understanding of genome order in the Asgard Archaea in the future.

Data Availability

The datasets used and analysed within the current study are available from the NCBI website as referenced in the paper.

Disclosure

A preprint has previously been published [96].

Conflicts of Interest

All authors declare they have no competing interests. LAK and ELS were student volunteers on this work. LAK was a fresh graduate from Clements High School, and ELS had just completed his 10th grade at Obra D. Tompkins High School, who volunteered with the group of Dr. George E. Fox at the University of Houston.

Authors' Contributions

MRT and GEF conceived and designed the study. MRT obtained the sequences. MRT, RVS, LAK, and ELS compared the organization of the ribosomal protein coding genes in the genomes. RVS, LAK, and ELS prepared the comparative chart for the gene clusters. MRT prepared the figures. MRT and GEF prepared the manuscript paper that was finalized with the help from all the authors. All authors read and approved the final manuscript.

Supplementary Materials