Functional Characterization and Evolutionary Analysis of Glycine-Betaine Biosynthesis Pathway in Red Seaweed Pyropia yezoensis

The red seaweed Pyropia yezoensis is an ideal research model for dissecting the molecular mechanisms underlying its robust acclimation to abiotic stresses in intertidal zones. Glycine betaine (GB) was an important osmolyte in maintaining osmotic balance and stabilizing the quaternary structure of complex proteins under abiotic stresses (drought, salinity, etc.) in plants, animals, and bacteria. However, the existence and possible functions of GB in Pyropia remain elusive. In this study, we observed the rapid accumulation of GB in desiccated Pyropia blades, identifying its essential roles in protecting Pyropia cells against severe osmotic stress. Based on the available genomic and transcriptomic information of Pyropia, we computationally identified genes encoding the three key enzymes in the GB biosynthesis pathway: phosphoethanolamine N-methyltransferase (PEAMT), choline dehydrogenase (CDH), and betaine aldehyde dehydrogenase (BADH). Pyropia had an extraordinarily expanded gene copy number of CDH (up to seven) compared to other red algae. Phylogeny analysis revealed that in addition to the one conservative CDH in red algae, the other six might have originated from early gene duplication events. In dehydration stress, multiple CDH paralogs and PEAMT genes were coordinating up-regulated and shunted metabolic flux into GB biosynthesis. An elaborate molecular mechanism might be involved in the transcriptional regulation of these genes.


Introduction
Plant performance and yield responses to water deficit stress conditions have been extensively studied [1][2][3]. Water deficiency causes cell metabolic disorder and cell membrane mechanical damage. A decline in leaf relative water content (RWC) reflects a loss of turgor that results in limited cell expansion and thereby reduced growth in crop plants [4,5]. Glycine betaine (GB) (N,N,N-trimethylglycine) is a quaternary ammonium compound found in bacteria, halophilic archaebacteria, marine invertebrates, plants, and mammals [6][7][8][9]. As an organic osmolyte, GB can maintain the osmotic balance and stabilize the quaternary structure of complex proteins under abiotic stresses. The protective role of GB has been widely reported in plants [10][11][12]. Treating plants with exogenous GB or increasing GB content in transgenic plants would greatly enhance their tolerance to osmotic stresses. Under chilling-stressed conditions, the heterologous expression of the betA gene significantly increased the tolerance of transgenic cotton seedlings to chilling injury through the accumulation of high levels of GB [13]. GB is also known to stabilize both photosystem II complexes and Rubisco under abiotic stress.
Pyropia yezoensis, previously known as Porphyra yezoensis [24], Rhodophyta, is a typical intertidal macroalgae. During daily low tides, P. yezoensis is routinely exposed to high levels of light, dehydration, and extreme fluctuations in temperature and osmotic pressure due to the seawater-to-air transition [25]. The blades can tolerate dehydration with a water loss of up to 85%, and recover physiological activities immediately upon rehydration [26]. These characteristics make P. yezoensis an ideal model for studying the molecular mechanisms of stress tolerance in intertidal-zone seaweeds. However, to date, there have been few reports investigating the cellular osmolytes in P. yezoensis. Is GB present in Pyropia cells? What are the roles of the GB in response to osmotic stress in Pyropia? What influences the diversity and conservation of the GB biosynthesis pathway in Pyropia? To address these questions, in this study, we computationally identified genes involved in the betaine biosynthesis pathway. We identified the variation of enzymatic activities and transcriptions for key genes during dehydration as well as re-hydration processes.

GB Contents in Dehydrated P. yezoensis Blades
To determine whether GB exists in red seaweed P. yezoensis and whether it is involved in desiccation tolerance, we detected the amounts of GB in the dehydration process. In 40-day old P. yezoensis blades, 6.720 µmol/g (fresh weight, FW) of GB was detected. The amount increased sharply to 19.924 µmol/g FW (a 2.96-fold increase compared to the control) when 50% of water was lost in the blades ( Figure 1). However, after an extension of dehydration with water loss at 80%, the GB amount dropped slightly, although there was still a 1.990-fold increase compared to the basal level. To determine whether GB exists in red seaweed P. yezoensis and whether it is involved in desiccation tolerance, we detected the amounts of GB in the dehydration process. In 40-day old P. yezoensis blades, 6.720 μmol/g (fresh weight, FW) of GB was detected. The amount increased sharply to 19.924 μmol/g FW (a 2.96-fold increase compared to the control) when 50% of water was lost in the blades ( Figure 1). However, after an extension of dehydration with water loss at 80%, the GB amount dropped slightly, although there was still a 1.990-fold increase compared to the basal level. The seaweeds were desiccated until the total water content reached 50% and 80%. Three replicates were produced for each treatment. Error bars indicate standard errors. The statistical significance compared to the control sample is indicated with one or two asterisks (p value is lower than 0.05 (*) or 0.01 (**), respectively).

GB Biosynthesis Pathway in P. yezoensis
The biosynthesis pathway of GB exhibited great diversity in plants, animals, bacteria as well as unicellular diatoms. Mainly two metabolic pathways lead to the GB biosynthesis, namely the choline oxidation pathway and the glycine methylation pathway. To identify candidate genes encoding the key enzymes related to GB biosynthesis in P. yezoensis and to further refer to its evolutionary history, we searched genomic and transcriptomic information published for P. yezoensis and other algal genomes including primary endosymbionts green algae and red algae, as well as the secondary endosymbionts diatom and Ectocarpus, via a blast program using the corresponding homologous genes from the above organisms as queries.
In the choline oxidation pathway, the first enzyme that catalyzes the oxidation of choline into choline aldehyde is choline dehydrogenase (CDH). In P. yezoensis, we found seven genes encoding proteins with similarity to CDH homologs, while only three and one were found in Chondrus and the unicellular red algae Cyanodioschyzon, respectively. All of them harbored a GMC-oxred N conserved domain at the N-terminus (Table 1). We carried out the phylogenetic analysis to probe the evolutionary forces driving the expansion of CDH genes. In the evolutionary tree, the algal CDHs are grouped together, while the bacteria and animal homologs are in two other clusters, suggesting an ancient separation in the evolution of algal CDHs. The PyCDH1 showed a phylogeny with strong affiliation with its red-algal homologs. The other six PyCDHs formed a sister group with two homologs from Chondrus ( Figure 2). The most plausible explanation for the extraordinary expansion of CDH genes was the explosive gene duplication events in the P. yezoensis genome.  BADH catalyzes the last step converting betaine aldehyde into betaine. Thus, we also computationally identified the putative BADH genes in P. yezoensis. PyBADH contains an ORF of 2428 bp, encoding 691 amino acids. PyBADH protein harbors a conservative motif x-x-x-E-L-G-G-Kx-x (Figure 3), which is characterized as dehydrogenation, and several residues that are characterized as being involved in NAD + binding and catalytic sites [27]. The phylogenetic analysis of BADHs showed that the PyBADH formed a close clade with unicellular red algae homologs ( Figure 4). BADH catalyzes the last step converting betaine aldehyde into betaine. Thus, we also computationally identified the putative BADH genes in P. yezoensis. PyBADH contains an ORF of 2428 bp, encoding 691 amino acids. PyBADH protein harbors a conservative motif Figure 3), which is characterized as dehydrogenation, and several residues that are characterized as being involved in NAD + binding and catalytic sites [27]. The phylogenetic analysis of BADHs showed that the PyBADH formed a close clade with unicellular red algae homologs ( Figure 4).
PEAMT catalyzes the methylation of phosphoethanolamine to form choline, and is characterized as being essential for betaine biosynthesis in plants. In P. yezoensis, we found one putative PEAMT gene encoding 468 amino acids. The functional domain analysis of PyPEAMT showed that PEAMT belongs to the N-methyltransferase superfamily. Mar. Drugs 2019, 17, x FOR PEER REVIEW 5 of 12     In the glycine methylation pathway for GB biosynthesis, the sequential methylation is catalyzed by two AdoMet dependent methyltransferases in bacteria: glycine sarcosine methyltransferase (GSMT) and sarcosine dimethyl-glycine methyltransferase (SDMT). When blasting against P. yezoensis unigenes using bacteria homologs as the query, the best hit goes to PyPEAMT, as all of them have the conserved methyltransferase domain. However, in the phylogeny tree with both SDMT/DMT and PEAMT homologs included, the PyPEAMT was grouped in the PEAMT clade ( Figure 5). Therefore, we deduced that the glycine methylation pathway might not exist in the P. yezoensis genome.
PEAMT catalyzes the methylation of phosphoethanolamine to form choline, and is characterized as being essential for betaine biosynthesis in plants. In P. yezoensis, we found one putative PEAMT gene encoding 468 amino acids. The functional domain analysis of PyPEAMT showed that PEAMT belongs to the N-methyltransferase superfamily.
In the glycine methylation pathway for GB biosynthesis, the sequential methylation is catalyzed by two AdoMet dependent methyltransferases in bacteria: glycine sarcosine methyltransferase (GSMT) and sarcosine dimethyl-glycine methyltransferase (SDMT). When blasting against P. yezoensis unigenes using bacteria homologs as the query, the best hit goes to PyPEAMT, as all of them have the conserved methyltransferase domain. However, in the phylogeny tree with both SDMT/DMT and PEAMT homologs included, the PyPEAMT was grouped in the PEAMT clade ( Figure 5). Therefore, we deduced that the glycine methylation pathway might not exist in the P. yezoensis genome.

Transcriptional Variations of CDH, BADH, and PEAMT Genes Under Dehydration
We carried out real-time fluorescence quantitative polymerase chain reaction (PCR) to investigate the transcriptional variation of PEAMT, BADH, and four CDH genes in P. yezoensis during the dehydration process. PEAMT exhibited a gradual up-regulation with the aggravation of water loss, and reached a peak at 80% water loss. A synchronized sharp increase in transcription at the onset of dehydration was observed in all the four CDH genes. Although they returned to the basal level thereafter, CDH1, CDH2, and CDH3 increased in the subsequent re-hydration, with CDH3 even rising to a 5.06-fold increase compared to the control. BADH only showed slight up-regulation at 30%

Transcriptional Variations of CDH, BADH, and PEAMT Genes Under Dehydration
We carried out real-time fluorescence quantitative polymerase chain reaction (PCR) to investigate the transcriptional variation of PEAMT, BADH, and four CDH genes in P. yezoensis during the dehydration process. PEAMT exhibited a gradual up-regulation with the aggravation of water loss, and reached a peak at 80% water loss. A synchronized sharp increase in transcription at the onset of dehydration was observed in all the four CDH genes. Although they returned to the basal level thereafter, CDH1, CDH2, and CDH3 increased in the subsequent re-hydration, with CDH3 even rising to a 5.06-fold increase compared to the control. BADH only showed slight up-regulation at 30% water loss. Its transcription dropped thereafter and even went to half of the basal level in re-hydration (Figure 6a). water loss. Its transcription dropped thereafter and even went to half of the basal level in re-hydration ( Figure 6a).

CDH Activity under Dehydration Stress in P. yezoensis Seaweeds
We further detected the enzymatic activity of P. yezoensis CDHs in desiccated blades and found that its variation was well correlated with the transcriptional levels. The CDH enzymatic activity increased immediately at the early stage of water loss (1.54-fold) and then slightly dropped to the basal level under severe dehydration. When rehydrated, the enzyme activity increased 1.4-fold again (Figure 7).

CDH Activity under Dehydration Stress in P. yezoensis Seaweeds
We further detected the enzymatic activity of P. yezoensis CDHs in desiccated blades and found that its variation was well correlated with the transcriptional levels. The CDH enzymatic activity increased immediately at the early stage of water loss (1.54-fold) and then slightly dropped to the basal level under severe dehydration. When rehydrated, the enzyme activity increased 1.4-fold again (Figure 7).
Mar. Drugs 2019, 17, x FOR PEER REVIEW 7 of 12 water loss. Its transcription dropped thereafter and even went to half of the basal level in re-hydration ( Figure 6a).

CDH Activity under Dehydration Stress in P. yezoensis Seaweeds
We further detected the enzymatic activity of P. yezoensis CDHs in desiccated blades and found that its variation was well correlated with the transcriptional levels. The CDH enzymatic activity increased immediately at the early stage of water loss (1.54-fold) and then slightly dropped to the basal level under severe dehydration. When rehydrated, the enzyme activity increased 1.4-fold again (Figure 7). Figure 7. Profiles of CDH activity under dehydration stress in P. yezoensis. The algae reached water loss levels of 30 ± 5%, 50 ± 5%, and 80 ± 5%, respectively, as well as rehydration after 80% water loss Figure 7. Profiles of CDH activity under dehydration stress in P. yezoensis. The algae reached water loss levels of 30 ± 5%, 50 ± 5%, and 80 ± 5%, respectively, as well as rehydration after 80% water loss for 30 min. Three replicates were produced for each treatment. The symbol * represents significant differences from control values at p < 0.05 (Student's t-test).

Discussion
The data presented in this study strongly indicates that glycine betaine, as a compatible solute, plays essential roles in protecting P. yezoensis against osmotic stress. This is the second algal species (the first red algae) reported to have a betaine biosynthesis pathway. However, in the diatom T. pseudonana, the enzymatic activity of CDH was not provided [19]. We also computationally identified homologous genes of PEAMT and BADH in other three red algae as well as green algae (Figure 6b). The wide occurrence of the choline oxidation pathway in organisms strongly suggests its evolutionary conservation. Conversely, the GB methylation pathway is present only in bacteria and the diatom T. pseudonana, as reported recently. The PEAMT gene in P. yezoensis has high protein similarity to bacteria SDMT and was doubted to be the putative SDMT homolog catalyzing the methylation of glycine. However, the phylogeny analysis did not support this argument. Moreover, at least two proteins (GSMT, SDMT/DMT) or one integrative peptide with two substrate-specific methyltransferase domains such as in T. pseudonana are required by the three-step methylation of glycine [19], while only one domain was found in PyPEAMT. Therefore, we deduced that the glycine methylation pathway is bacteria-specific and the diatom T. pseudonana might obtain this pathway occasionally via horizontal gene transfer, transposable elements, etc.
Choline, the precursor of betaine, could also go into either the lipid metabolic pathway or the acetylcholine biosynthesis. Choline dehydrogenase was the key enzyme to determine the metabolic direction of choline into betaine biosynthesis. In the P. yezoensis genome, up to seven CDH paralogs were identified, representing the highest gene dose among known animal, algal, and bacterial genomes. The increase in particular gene families was frequently reported in genomic analysis. Usually, several mechanisms have been proposed to play essential roles in obtaining new genetic materials for environmental adaptation, including genome-scale duplication, regional duplication due to homologous recombination, and illegitimate recombination, as well as horizontal gene transfer from bacteria or archaea. No evidence of genome duplication was observed in P. yezoensis. Besides, in the phylogeny analysis of PyCDHs, they were clustered separately from bacteria CDH homologs, excluding the existence of horizontal gene transfer (HGT) events in the CDH expansion. Therefore, according to the close phylogenetic relationship of the six PyCDH paralogs, especially the high sequence similarities in the two paralogous groups (PyCDH2 and PyCDH4, PyCDH5 and PyCDH6), we deduced that the most plausible explanation was that multiple gene duplication events leading to the acquirement of new CDH genes happened in the evolution of the P. yezoensis genome, accounting for the rapid accumulation of betaine for cell protection when encountering osmotic stresses.
Another prominent observation in P. yezoensis GB biosynthesis was that CDH paralogs exhibited simultaneous up-regulation at the onset of osmotic stress, and then dropped to the basal level when dehydration was ongoing. Three of them also showed transcriptional elevation in re-hydration. The coordinated transcriptional pattern suggested the existence of an elaborate regulatory mechanism that could be induced by osmotic stress. In addition to this, our analyses showed an increase in the enzyme activity for CDH under moderate and severe osmotic stress in P. yezoensis. Further study of the cis-regulatory elements or conserved DNA motifs located in their promoter regions would help to illustrate the key components and their regulatory network in response to osmotic stresses.

Algal Material and Stress Treatment
The pure line RZ58 of P. yezoensis, established by clonal cultivation of an isolated single somatic cell and self-fertilization in the laboratory, was used for the experiments. Fresh leafy gametophytes were cultured in bubbling natural seawater with Provasoli's enrichment solution medium (PES) under 50 µmol photons m-2s-1 at 10 ± 1 • C and a 12:12 light (dark photoperiod before use). The seawater was bubbled continuously with filter-sterilized air and renewed every three days. P. yezoensis blades were subjected to dehydration by exposing them to the air for a period of time. They were collected when water loss was weighed to be 30 ± 5%, 50 ± 5%, and 80 ± 5%, respectively. For rehydration, severely dehydrated samples (water loss at 80 ± 5%) were transferred back into seawater, and collected 0.5 h later. Water loss was determined according to the method by Kim et al. [28]. Three biological replicates were produced for treatment. After harvesting and weighing, the samples were immediately frozen in liquid nitrogen and stored individually at −80 • C until use.

RNA Isolation and cDNA Synthesis
Total RNA was extracted from the thallus using the Plant RNA Kit (Omega, U.S.) in accordance with the manufacturer's instructions. RNA degradation and contamination were monitored on 1% agarose gels. Purity was evaluated using a NanoPhotometer spectrophotometer (IMPLEN, CA, U.S.). RNA concentration was measured using a Qubit RNA Assay Kit and Qubit 2.0 Fluorometer (Life Technologies, CA, U.S.). RNA integrity was assessed using an RNA Nano 6000 Assay Kit and the Bioanalyzer 2100 system (Agilent Technologies, CA, U.S.). The cDNA templates were synthesized using a First Strand cDNA Synthesis Kit (Roche, Germany) according to the manufacturer's instructions. Subsequently, 1 µg total RNA was used to synthesize the first-strand cDNA. The resulting cDNA mixture was diluted 10 times by adding nuclease-free water and stored at −20 • C.

Quantitative Real-Time PCR (RT-PCR)
Sequence-specific primers of PyBADH, PyPEAMT, and PyCDHs, based on their multiple alignments, were designed using the Primer 5.0 software (Premier Biosoft International, USA) ( Table 2). Quantitative real-time PCR (RT-qPCR) was performed with Light Cycler480 (Roche, Germany). PyUBC was simultaneously used as internal control to normalize the amount of messenger RNA (mRNA) in each reaction. The cycling parameters were as follows: 95 • C for 5 min, followed by 45 cycles of 95 • C for 10 s, 60 • C for 10 s, and 72 • C for 20 s. The mean amplification efficiency of each primer pair was checked with the LightCycle ® 480 gene scanning software version 1.5 (Roche, Germany). The 2−∆∆Ct method was used to assess the expression of relative genes [29]. The qPCR results shown are the average (±SD) of three biological repeats.

GB Extraction and Liquid Chromatography Analysis
About 200 mg of the material was ground with liquid nitrogen, then pre-cooled with 1 mL methanol, and ultrasonically extracted for 30 min. This was centrifuged for 10 min at 10,000× g (HITACHI, Tokyo, Japan) and the supernatant was mixed after repeating twice, concentrated to dry under a nitrogen blower (NDK200-2, Miulab, Hangzhou, China), dissolved by vortex oscillation in 1 mL water was added, and then filtered with 0.22-µm-pore filter membranes (Millipore).

Detecting Content of GB
The content of betaine was determined by HPLC (L-3000, RIGOL, Beijing, China). By comparing the peak area of the samples with that of the standards (Shanghai yuanye, Bio-Technology, Shanghai, China), the content of the samples was calculated.

Determination of Choline Dehydrogenase Activity
The choline dehydrogenase activity was detected using a Plant CHDH ELISA kit (Mlbio, Shanghai, China). Phosphate buffer saline (PBS) (100 µg material, 900 µL PBS) was added into a tube after grinding samples with liquid nitrogen. The samples were mixed and then centrifuged for 20 min (Hitachi, Tokyo, Japan) at 2500× g and the supernatant was taken. The kit was removed from the 4 • C environment and balanced at room temperature for 25 min. The standard wells, sample wells, and blank wells were set, and the absorbance (OD) of each well was measured at 450 nm by zero-setting the blank well.

Statistics
Data were analyzed using the Origin Pro 9.0 software (Origin Lab, Northampton, MA, USA). Datasets were normalized to test the statistical significance of the treatment effect and a one-way ANOVA was performed using the SPSS 22.0 software (IBM, USA). Significant differences between group means were determined as p < 0.05 via Tukey's and Fisher's tests.

Conclusions
In this study, we reconstructed the biosynthesis pathway of glycine betaine in the economically important seaweed P. yezoensis and experimentally detected the existence and rapid accumulation of GB as well as the transcriptional regulation under desiccation stress. The extensive illustration of the genetic foundation in the GB biosynthesis in P. yezoensis will undoubtedly further our understanding of the adaptive strategies of intertidal seaweeds in their evolutionary history and provide precious genetic resources for genetic engineering of algae and plants for abiotic stress tolerance.