**1. Introduction**

Mitochondria are membrane-bound organelles known for supplying eukaryotic cells with energy through ATP to carry out cellular functions. This occurs due to aerobic respiration whereby pyruvate is oxidized to CO2 to generate reduced cofactors that drive the electron transport chain to chemiosmotically fuel ATP synthesis [1]. Despite the crucial role mitochondria have in supplying energy necessary for cellular functions and ATP for other biochemical pathways, it did not originate as a component of the eukaryotic cell. During the late 20th century, the theory of endosymbiosis became widely accepted and states that an aerobic bacterium was absorbed by, and formed an endosymbiotic relationship with, a pre-eukaryotic cell [2,3]. Though it became fully integrated into the Last Eukaryotic Common Ancestor (LECA), the proposed alpha-proteobacterium [4] maintained a portion of its circular genome carrying a conserved set of genes enabling the quick modulation of crucial energy acquisition proteins [5]. This remnant of the bacterial genome is referred to as mtDNA, the mitochondrial genome or chondriome.

**Citation:** Proulex, G.C.R.; Meade, M.J.; Manoylov, K.M.; Cahoon, A.B. Mitochondrial mRNA Processing in the Chlorophyte Alga *Pediastrum duplex* and Streptophyte Alga *Chara vulgaris* Reveals an Evolutionary Branch in Mitochondrial mRNA Processing. *Plants* **2021**, *10*, 576. https://doi.org/10.3390/ plants10030576

Academic Editor: Nunzia Scotti

Received: 18 February 2021 Accepted: 13 March 2021 Published: 18 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Mitochondria have retained their own gene expression machinery, combining bacteriallike traits with novel features that evolved in the host cell [6]. Quite a bit is known about mitochondrial transcription and RNA processing from the compact chondriomes of humans and mice, which can serve for an overview of the process. Briefly, a nuclearencoded RNA polymerase similar to those found in T3 and T7 bacteriophages [7] recognizes a promoter on both strands of mtDNA with the aid of two transcription factors [8,9]. These promoters occur in the only non-coding region (hyper-variable) and produce two long poly-cistronic primary RNAs known as heavy and light [10–14] with the aid of an elongation factor [14]. Individual mRNAs and tRNAs are removed from the primary transcripts by endonucleolytic cleavage by the enzymes RNaseP and RNaseZ, which precisely remove tRNAs, leaving most of the mRNAs as individual mRNAs with very short 3 and 5- UnTranslated Regions (UTRs), a process called the Punctuation Model [15–17]. Endonucleolytic processing between mRNAs with no intervening tRNA and between an mRNA with an adjacent antisense tRNA has been documented [13], but the enzymatic mechanisms responsible for these processing events are currently unknown. Having no 5- UTRs, these mRNAs lack canonical ribosome binding sequences and use an alternative ribosome binding mechanism that is unique to mitochondria [18]. Once cleaved from the primary transcript, mRNAs may be polyadenylated, which adds the final adenine in some transcripts' stop codons, stabilizes some, and acts as a degradation signal for truncated messages [19–21]. mRNA fragments, but not full-length mRNAs, may also be circularized [22].

The mitochondrial genomes of plants (embryophytes) are much larger than those in animal cells due to expansive intergenic regions, repetitive DNA, and introns [23]. Plant mitochondria share some transcriptional processes with vertebrates. Transcription in plant mitochondria is catalyzed by one or more nuclear encoded phage-like RNA polymerases [24,25], and transcription factors similar to those used by vertebrates are encoded in plant nuclear genomes [26], but their functions have yet to be demonstrated. Due to their sizes, plant chondriomes have multiple promoters dispersed throughout the chondriome [27], yielding multiple primary poly-cistronic transcripts. Post-transcriptional processing takes on an expanded role in plants requiring numerous RNA Processing Factors (RPFs) that target endo- and exo-nucleolytic enzymes, define mRNA ends, and modify transcripts [28]. The 5 termini of genes directly downstream of a transcriptional promoter are formed by the initial nucleotide added by the RNA polymerase [29]. For downstream genes in poly-cistronic transcripts, endonucleolytic cleavage between two genes will simultaneously produce the 5- UTR of one gene and the 3- UTR of an adjacent one. The lengths of these UTRs range from dozens to thousands of nucleotides consistent with the large intergenic regions of plant chondriomes [29–31]. To date, the best-defined cleavage mechanisms in plants are the precise removal of tRNAs by RNaseZ and PRORP. Similarly, tRNA-like secondary structures called t-elements also define intergenic cleavage sites recognized by endonucleases [31–35]. Most protein-coding genes are not separated by tRNAs, and their intergenic cleavage mechanism is hypothetical at this time but involves at least two nucleases [30,36]. Multiple 5 termini for each gene usually result from these processes [30,31]. The 3 ends are less variable and gene specific RPFs bind to them, presumably defining and stabilizing them [37–41]. The prevalence of group I and II introns in plant mitochondria creates an added layer of post-transcriptional processing. Neither class of intron is able to self-splice, so a group of nuclear-encoded RPFs are necessary for their removal [42]. In addition to the major construction of the mRNA coding regions, individual nucleotides are modified in a process known as RNA editing, which is common in higher plant mitochondria and chloroplasts [43]. Once the mRNA is no longer needed, it may be marked for degradation by way of polyadenylation by nuclear encoded factors in a manner similar to that of bacteria [44–47].

Our understanding of mitochondrial transcription and RNA processing in algal species is mostly limited to the single-celled photosynthetic green alga *Chlamydomonas reinhardtii* P.A. Dangeard, which is a well-established model system [48]. *C. reinhardtii* has a small linear chondriome [49] that is unusual among algae but is a conserved trait among the Reinhardtinia clade of the Order Chlamydomonadales [50,51]. In this species, transcription is initiated on each of the two strands from promoters in a small intergenic region to produce two primary transcripts [52,53]. Each mRNA is endonucleolytically cleaved directly adjacent to the AUG start codon, leaving no 5- UTR, similar to those seen in animal systems. The 3- UTRs are comprised of various lengths of template-derived intergenic regions and may have non-template polycytosine and/or polyuracil tails added, presumably as part of the maturation process [54–56]. The poly-cytidylation of mitochondrial mRNAs seen in green algae is unusual and appears to be limited to the algal class Chlorophyceae [55]. It has been hypothesized that these leaderless mRNAs use an alternative ribosome-binding mechanism, but there is evidence that the mature mRNAs are circularized, which brings putative ribosome binding sites (RBSs) located in the intergenic regions of the *Chlamydomonas* chondriome upstream of the start codon to initiate translation [56]. mRNAs in *C. reinhardtii* are also poly-adenylated, which serves as a degradation signal consistent with mitochondria in other eukaryotes and bacteria [54,57,58]. mRNA editing, which is common among embryophytes, is missing in both the chlorophyte and streptophyte lineages of green algae [59], suggesting that some post-transcriptional processes were acquired by embryophytes after they invaded land.

The purpose of this study was to define the 5 and 3- UTRs of mitochondrial mRNAs in two algae, *Pediastrum duplex* Meyen (P: Chlorophyta, C: Chlorophyceae, F: Hydrodictyaceae) from the chlorophyte algal clade and *Chara vulgaris* Linnaeus (P: Charophyta, C: Charophyceae, F: Characeae) of the charophyte algal clade. *P. duplex* is a member of the same class as *C. reinhardtii*, but it has a circular chondriome that is several times larger [60,61], making the architecture more similar to that found in other algae. By defining the termini, we hoped to determine if RNA processing events seen in *C. reinhardtii* are also used among other chlorophytic algae or are related to its compact genome. *C. vulgaris* has a mitochondrial genome similar in size to *P. duplex*, but the gene content and synteny are more similar to bryophytes [62,63]. We analyzed the 3 and 5 ends from *C. vulgaris* to see if mRNA end processing resembled higher plants or chlorophytic green algae. Circular RT-PCR (cRT-PCR) and PacBio long-read sequencing were used to define the transcript termini of 12 mitochondrial mRNAs from each species. cRT-PCR allows the mapping of both the 3 and 5 ends of an RNA by artificially ligating them together followed by the production of a cDNA across the ligation site, PCR amplification of the sequences flanking the ligation site and sequencing of those amplicons. PacBio Iso-Seq is a long-read RNA sequencing technology which can sequence full-length RNAs including their 3 and 5- termini. This platform sequences poly-adenylated RNAs so organellar mRNAs must be artificially poly-adenylated to increase the likelihood they will be sequenced.
