Fig 1.
The 3’-terminus of the 16S rRNAs capable of interacting with the 5’ UTR region of mRNAs is unstructured.
The secondary structure of the last 40 nucleotides of the E. coli 16S rRNA is presented. The antiSD core, the SD sequence, and the start codon are colored red, purple, and green, respectively. Note that the last 15 nucleotides of the 16S rRNA are unstructured and correspond to the most likely region to interact with the SD sequence in the mRNAs.
Fig 2.
Example of the output of the RNAhybrid program.
The interacting bases in the SD and antiSD sequences are colored in red. Note that -13.2 kcal/mol corresponds to the most significant value for a perfect hybridization of the canonical core of the SD and antiSD sequences.
Fig 3.
Distribution of the 16S rRNA copy number in the study organisms.
16S rRNA genes were identified in our set of 6,457 selected organisms using the MAST program with MEME position-probability matrices built considering a set of representative sequences of 16S rRNA genes from bacteria and archaea, as described in the Materials and Methods section.
Fig 4.
Histograms of the relative frequencies at which the antiSD/SD sequence bases interact in model organisms.
Based on our thermodynamic analyses, we predicted the bases of the leader sequences of the mRNAs in an organism most likely to interact with the 3’-terminus of their 16S rRNAs. We then represented the frequencies of the interacting bases as histograms. Positions with a relative frequency greater than 40% (green area up) were considered part of the SD sequence, and their counterpart in the 16S rRNA was considered an antiSD sequence. At each position, the frequency of the interacting base in the mRNA is indicated. The histograms illustrate SD:antiSD interactions for the model organisms E. coli (A) and B. subtilis (B). The E. coli antiSD sequence (5’-CCUCCUU-3’) and the B. subtilis extended antiSD sequence (5’-UCACCUCCU-3’) are shown in orange. The canonical CCUCC is shown in green.
Fig 5.
Histograms of the relative frequencies at which the nucleotides of the antiSD and SD interact in model-representative phylogenetic classes.
(A) Thermatogae, (B) Acidobacteria, and (C) Epsilonproteobacteria, as examples of phylogenetic classes in which histograms of the relative frequencies of SD:antiSD interacting bases are skewed to the left, centered, or skewed to the right. Their antiSD sequences and canonical CCUCC are shown in orange and green, respectively.
Fig 6.
Classification of antiSD sequences in accordance with their sequence and position in the 16S rRNA 3’-terminus.
Based on the frequency of interactions between the nucleotides of the 16S rRNA 3’-terminus and the bases of the 5’ upstream regions of mRNAs, we defined and classified the antiSD sequences in our set of reference organisms according to their position relative to the CCUCC, which is present in almost all 16S rRNA molecules. The letters in the figure represent the consensus of interacting bases in the rRNA of each group, while dots indicate non-interacting bases. The consensus sequences can be found in Fig 7. The CCUCC is enclosed in a red rectangle. Pie charts show the relative frequencies of bacteria and archaea in each group.
Fig 7.
Histograms of representative classes of the fifteen groups of antiSD identified in our analysis.
Of the 78 phylogenetic classes, one is selected from each of the ten antiSD classes we defined based on the similarity of their sequences, lengths, and positions within their 3’-terminus.
Table 1.
SD and antiSD sequences of organisms grouped by phylogenetic classes.
Fig 8.
Breakdown of the phylogenetic classes in group 8 (from Fig 7).
The histograms include three phylogenetic classes of archaea and five of bacteria. Group 8 is the only one that includes members from both domains. This group is characterized by having a relatively long antiSD sequence that starts further left from the CCUCC, compared to most bacterial antiSD sequences.
Fig 9.
Phylogenetic tree of representative organisms with different classes of antiSD/SD sequences.
To represent the phylogenetic distribution of organisms belonging to the different types of categorizations mentioned above, 323 representative members from our list of studied organisms were selected, and their 16S rRNA gene sequences were used to build the phylogenetic tree. The green and orange bars show the GC content and the percentage of mRNAs capable of establishing stable 16S rRNA/mRNA interactions, respectively. The names of the organisms are colored by one of the four groups to which they belong: colorless, yellow, green, and purple, to represent the groups i, ii, iii, and iv, respectively.
Table 2.
"SD-Scarce" organisms with conserved CCUCC.
Fig 10.
Histograms of the frequencies at which the bases of the 3’-terminus of the 16S rRNA interact with the 5’ upstream region of the mRNAs with relatively high frequencies for organisms lacking the canonical CCUCC.
A) and B) correspond to histograms of archaeal phylogenetic organisms, while C) corresponds to the Alphaproteobacteria class of the Bacteria domain. According to the frequencies of interaction with the leader regions of the mRNAs, the bases at the 3’ end of the 16S rRNA were defined as part of the antiSD sequence and are indicated in brown boxes. The canonical sequences CCUCC of the 16S rRNA are framed in green boxes, and the changes that move them away from the consensus are indicated in red boxes.
Table 3.
“SD-Abundant” organisms lacking a conserved CCUCC.
Table 4.
"SD-Scarce" organisms lacking a conserved CCUCC.