Membrane-Fluidization-Dependent and -Independent Pathways Are Involved in Heat-Stress-Inducible Gene Expression in the Marine Red Alga Neopyropia yezoensis

Heat stress responses are complex regulatory processes, including sensing, signal transduction, and gene expression. However, the exact mechanisms of these processes in seaweeds are not well known. We explored the relationship between membrane physical states and gene expression in the red alga Neopyropia yezoensis. To analyze heat-stress-induced gene expression, we identified two homologs of the heat-inducible high temperature response 2 (HTR2) gene in Neopyropia seriata, named NyHTR2 and NyHTR2L. We found conservation of HTR2 homologs only within the order Bangiales; their products contained a novel conserved cysteine repeat which we designated the Bangiales cysteine-rich motif. A quantitative mRNA analysis showed that expression of NyHTR2 and NyHTR2L was induced by heat stress. However, the membrane fluidizer benzyl alcohol (BA) did not induce expression of these genes, indicating that the effect of heat was not due to membrane fluidization. In contrast, expression of genes encoding multiprotein-bridging factor 1 (NyMBF1) and HSP70s (NyHSP70-1 and NyHSP70-2) was induced by heat stress and by BA, indicating that it involved a membrane-fluidization-dependent pathway. In addition, dark treatment under heat stress promoted expression of NyHTR2, NyHTR2L, NyMBF1, and NyHSP70-2, but not NyHSP70-1; expression of NyHTR2 and NyHTR2L was membrane-fluidization-independent, and that of other genes was membrane-fluidization-dependent. These findings indicate that the heat stress response in N. yezoensis involves membrane-fluidization-dependent and -independent pathways.

In addition, the presence of an intrinsic ability to memorize non-lethal heat stress, which enables survival under lethal temperature by establishment of heat stress tolerance, was confirmed in 'Bangia' sp. ESS1 [9]; however, other 'Bangia' species lack this heat stress memory [10]. Therefore, elucidation of mechanisms regulating responses to heat stress and acquisition of heat stress tolerance based on heat stress memory is indispensable for building strategies to protect the economically important Bangiales from seawater warming.
Responsibility to heat stress with changes in the gene expression pattern, morphology, and composition of metabolites is evolutionally conserved in green lineage including terrestrial plants and their closest algal relatives [11][12][13][14][15]. However, it seems that there are both conserved and diversified parts in the regulatory mechanisms of these physiological and molecular responses to heat stress among Archaeplastida. Indeed, in terrestrial plants, heat-inducible transcription of the gene encoding of the class A3 heat shock factor (HsfA3), which is a transcription factor regulating mRNA expression of heat shock protein (HSP) genes, is controlled by the APETALA2 (AP2) domain transcription factor dehydration-responsive element-binding protein 2A (DREB2A) [16][17][18]. In addition, expression of DREB2A is heat-inducible through regulation by the class A1 Hsfs, such as HsfA1a, HsfA1b, HsfA1d, and HsfA1e [18,19]. These findings indicate that the heat-stressinducible expression of HSP genes involves a stepwise activation cascade of transcription factors. However, genome and extensive transcriptome analyses have clearly demonstrated a lack of AP2-domain-containing proteins in red algae, suggesting that the heat signal transduction pathways in Bangiales differ from those in terrestrial plants. Moreover, although the presence of the HSF gene was confirmed in Bangiales, including N. yezoensis and Porphyra umbilicalis [20], as well as in unicellular red algae and Florideophyceae [20,21], little is known about the activation mode and function of HSF in red algae. Regulatory mechanisms of heat stress responses in Bangiales remain unknown.
Heat stress increases plant membrane fluidity [22,23]. Therefore, homeostasis of the membrane physical state under heat stress is balanced by an increase in saturation level due to accumulation of saturated fatty acids, which increases membrane rigidity and lowers membrane fluidity [24][25][26][27][28]. Thus, sensing of membrane fluidization is thought to trigger plant heat stress responses [29][30][31][32][33]. Evidence for this includes an artificial increase in the saturation level of membrane fatty acids following pharmacological treatment with the membrane fluidizer benzyl alcohol (BA) [34][35][36] and the genetic inactivation of fatty acid desaturase and glycerol-3-phosphate acyltransferase genes in terrestrial plants [37][38][39]. In 'Bangia' sp. ESS1 (order: Bangiales), a moderate increase in the saturation rate of membrane fatty acids under heat stress conditions promotes acquisition of heat stress tolerance through the establishment of heat stress memory [9]. However, it is unknown whether heat-stress-induced gene expression is initiated by changes in membrane fluidity.
In the present study, we analyzed the relationship between membrane fluidization and the expression of genes encoding novel cysteine-rich domain proteins (homologs of HTR2 of N. seriata; [6]), HSP70s [5], and multiprotein-bridging factor 1 (MBF1; [40]) in the red alga N. yezoensis. Our findings demonstrate that both membrane-fluidization-dependent and -independent pathways regulate heat-inducible gene expression and that these pathways can be enhanced by dark conditions. These findings provide insights into the triggers of heat stress responses in poikilothermic organisms.
Three life cycle generations were subjected to temperature stress, which commenced at 12:00, 3 h after the start of light irradiation (9:00). Stress treatment was performed by incubating algae at 5, 15, and 25 • C for 0.5, 1, 2, 4, 6, 8, and 12 h. Since the duration of light irradiation was only 10 h, the 8 and 12 h incubations included 1 and 5 h of dark conditions, respectively. Experiments were also carried out with 24 h light.
Pharmacological treatment with benzyl alcohol (BA) and dimethyl sulfoxide (DMSO) was performed by incubating algae at 15 • C for 5, 15, and 30 min. Incubations at 25 • C for 8 and 12 h under light with BA or dark conditions with DMSO were also performed.
Temperature-stressed and pharmacologically treated gametophytes, sporophytes, and conchosporophytes were harvested, frozen in liquid nitrogen, and stored at −80 • C prior to use for gene expression analysis.

Identification and Phylogenetic Analysis of HTR2 Homologs
A homology search using our N. yezoensis transcriptome data [42] was performed against the NsHTR2 amino acid sequence to find candidate NsHTR2 homologs. The ORF finder (https://www.ncbi.nlm.nih.gov/orffinder/ (access on 30 March 2020)) was used to identify full-length open reading frames (ORFs) of these candidates. A BLAST search (https://blast. ncbi.nlm.nih.gov/Blast.cgi (access on 30 march 2020)) using predicted amino acid sequences of NsHTR2 and candidates from N. yezoensis as queries was performed to find HTR2 homologs from other algae and terrestrial plants. The amino acid sequences of all the obtained HTR2 homologs were used to construct a neighbor-joining phylogenetic tree with MEGA 7 (https: //www.megasoftware.net (access on 1 December 2021)) using ClustalW to align the sequences. The accession numbers of the sequences, except for NsHTR2, from DDBJ/EMBL/GenBank are shown next to the species names in the phylogenetic tree in Figure 1.
irradiation was only 10 h, the 8 and 12 h incubations included 1 and 5 h of dark conditions, respectively. Experiments were also carried out with 24 h light.
Pharmacological treatment with benzyl alcohol (BA) and dimethyl sulfoxide (DMSO) was performed by incubating algae at 15 °C for 5, 15, and 30 min. Incubations at 25 °C for 8 and 12 h under light with BA or dark conditions with DMSO were also performed. Temperature-stressed and pharmacologically treated gametophytes, sporophytes, and conchosporophytes were harvested, frozen in liquid nitrogen, and stored at −80 °C prior to use for gene expression analysis.

Identification and Phylogenetic Analysis of HTR2 Homologs
A homology search using our N. yezoensis transcriptome data [42] was performed against the NsHTR2 amino acid sequence to find candidate NsHTR2 homologs. The ORF finder (https://www.ncbi.nlm.nih.gov/orffinder/ (access on 30 March 2020)) was used to identify full-length open reading frames (ORFs) of these candidates. A BLAST search (https://blast.ncbi.nlm.nih.gov/Blast.cgi (access on 30 march 2020)) using predicted amino acid sequences of NsHTR2 and candidates from N. yezoensis as queries was performed to find HTR2 homologs from other algae and terrestrial plants. The amino acid sequences of all the obtained HTR2 homologs were used to construct a neighbor-joining phylogenetic tree with MEGA 7 (https://www.megasoftware.net (access on 1 December 2021)) using ClustalW to align the sequences. The accession numbers of the sequences, except for NsHTR2, from DDBJ/EMBL/GenBank are shown next to the species names in the phylogenetic tree in Figure 1.

Total RNA Extraction, DNA Removal, and cDNA Synthesis
Extraction of total RNA from gametophytes, sporophytes, and conchosporophytes, removal of genomic DNA contamination, quality check of purified total RNA, and synthesis of first-strand complementary DNA (cDNA) were performed as described in [41]. The thermal cycling parameters for the first-strand cDNA synthesis consisted of an initial sizes according to [41]. We used two real-time PCR machines. The thermal cycling parameters for the Applied Biosystems 7300 real-time PCR system (Life Technologies, Carlsbad, CA, USA) consisted of 95 • C for 5 min and 40 cycles of 94 • C for 30 s, 60 • C for 30 s, and 72 • C for 20 s. The thermal cycling parameters for the AriaMX (3000P) real-time PCR system (Agilent Technologies, Santa Clara, CA, USA) consisted of 95 • C for 3 min and 40 cycles of 95 • C for 5 s and 60 • C for 20 s. A dissociation curve was generated to check for specificity of amplification by heating from 60 to 95 • C in the Applied Biosystems 7300 real-time PCR system or by 95 • C for 1 min, 55 • C for 30 s, and 95 • C for 30 s in the AriaMX (3000P) real-time PCR system.

Statistical Analysis
Values are presented as the mean ± SD from triplicate experiments. Two-way ANOVA followed by a Tukey-Kramer test was used for multiple comparisons, and significant differences were determined using a cutoff value of p < 0.05.

Discovery of a Novel Protein Family Specifically Conserved in Bangiales
By homology searching of the transcriptome data of N. yezoensis [42], we found two homologs (Unigene16775 and CL1219.Contig1) of NsHTR2, which has been identified as a heat-inducible gene in N. seriata [6]. In addition, we identified eight homologs (OSX78858, OSX74332, OSX76372, OSX71170, OSX80752, OSZ76191, OSX79754, and OSX68660) from the red alga Porphyra umbilicalis [43] through an NCBI BLAST search with NsHTR2 as a query. We found no other homologs among florideophyceans, green and brown algae, or terrestrial plants, indicating that HTR2 is a Bangiales-specific protein. Phylogenetic analysis using amino acid sequences of the proteins revealed two clades ( Figure 1): an HTR2 clade containing all Neopyropia proteins and five P. umbilicalis proteins (OSX78858, OSX74332, OSX76372, OSX71170, and OSX80752) and a clade of HTR2-related proteins that contains OSZ76191, OSX79754, and OSX68660 from P. umbilicalis. Since Unigene16775 is more closely related to NsHTR2 than to CL1219.Contig1, we designated Unigene16775 and CL1219.Contig1 as NyHTR2 and NyHTR2L (HTR2-like), respectively. The DDBJ/EMBL/GenBank accession numbers of NyHTR2 and NyHTR2L are LC672062 and LC672063, respectively. It is unknown whether N. seriata has other HTR2 homologs.
Amino acid alignment indicated that all the HTR2 homologs have a unique domain at their C-terminal half in which the positions of 10 cysteine (Cys) residues were highly conserved ( Figure S1). Cross-brace zinc finger motifs, such as RING, PHD, FYVE, and ZZ fingers, are structural features that bind zinc atoms through eight Cys and/or His ligands, each with a specific biological role [44,45]. However, the Cys-rich domains of the HTR2 proteins differed from previously identified cross-brace zinc fingers and did not show sequence similarity to any conserved domains based on three-dimensional structure predictions with the Phyre2 Server (http://www.sbg.bio.ic.ac.uk/phyre2/html/ (access on 1 December 2021)). Thus, HTR2 homologs are novel proteins containing an unknown Cys-repeat motif. Since it is uncertain whether this motif is a new type of cross-brace zinc finger, we designated it the Bangiales cysteine-rich (BCR) motif.

Determination of a Reference Gene for Quantitative Analysis of Heat-Inducible Gene Expression in Different Life Cycle Generations of N. yezoensis
We selected three candidate reference genes for qRT-PCR analysis of heat-inducible gene expression in three generations of the N. yezoensis life cycle, i.e., gametophyte, sporophyte, and conchosporophyte: GAPDH (glyceraldehyde 3-phosphate dehydrogenase), EF1α (Elongation factor1-α), and eIF4A (Translation initiation factor 4A). These were previously proposed as reference genes for qRT-PCR in N. yezoensis [46,47]. We validated the specificity and efficiency of the newly designed primer pairs (Table S1) by 2% agarose gel electrophoresis of amplicons of each candidate gene derived from cDNA libraries of gametophytes, sporophytes, and conchosporophytes. A single PCR fragment with expected size was amplified, which gave rise to a single peak in the melting curve (data not shown). The amplification efficiency (E) of each set ranged from 95.13 to 102.55%, and the linear correlation coefficients (R 2 ) ranged from 0.9948 to 0.9973 (Table S1), both of which were within the acceptable range (E: 90-105%; R 2 : 0.9910-0.9998). Moreover, we were able to generate standard curves for all primer sets using a tenfold dilution series with six dilution points in triplication (data not shown). Thus, all primer sets are suitable for further characterization of candidates as desired reference genes.
We calculated the expression stability of these candidates by measuring the cycle threshold (Ct) values of the genes. We synthesized template cDNAs for qRT-PCR using total RNAs separately prepared from three generations, each of which were a mixture of one control sample and triplicate samples treated with 5, 15, or 25 • C for 0.5, 1, 2, 4, 6, 8, or 12 h for each generation (a total of 22 samples for each generation). The Ct values of reference genes ranged from 17.4 to 23.8, 16.7 to 23.2, and 18.3 to 24.3 in gametophytes, sporophytes, and conchosporophytes, respectively ( Figure S2A). Of the candidates, EF1α displayed the lowest expression level, while both GAPDH and eIF4A had the highest expression levels ( Figure S2A). Next, we used geNorm software to rank the expression stability of each gene by calculating the expression stability value. Figure S2B shows the average expression stability (M) of three candidates. eIF4A was the most stable, with M < 1.5 in all of generations. Moreover, eIF4A was ranked as the most stable gene in all three life cycle generations using NormFinder (Table S2). Therefore, we concluded that eIF4A is a suitable reference gene for normalizing gene expression under both heat and cold stress conditions in all life cycle generations.
To confirm the above findings, we subjected all the candidate reference genes to qRT-PCR to examine NyMBF1 expression across N. yezoensis generations under heat (25 • C) and cold (5 • C) stress (see Table S3 for primers for NyMBF1). Our rationale was that it seemed plausible that NyMBF1 expression might be induced by heat stress, as occurs with numerous genes in terrestrial plants [48]. We incubated algae from each of the three generations separately at 5, 15, or 25 • C for 0.5, 1, 2, 4, 6, 8, or 12 h. When we normalized qRT-PCR data using eIF4A as the reference gene, it was clear that heat (but not cold) induced the expression of NyMBF1 in all three generations, with peaks at 2 and 8 h after exposure to 25 • C. In contrast, when we normalized the data using GAPDH or EF1α, we did not observe heat-dependent expression of NyMBF1 (Figures S3-S5, Table S4). These findings indicate that NyMBF1 is a heat-stress-inducible gene, and eIF4A is a suitable reference gene to use in monitoring gene expression under heat and cold stress in the three generations of N. yezoensis. Accordingly, we employed eIF4A as a reference gene in the rest of our study.

Heat-Inducible and Dark-Stimulated Expression of NyHTR2 and NyHTR2L
Since HTR2 was identified as a heat-inducible gene in N. seriata [6], we analyzed expression profiles of both NyHTR2 and NyHTR2L under the same cold and heat stress conditions as those used to analyze NyMBF1 (Figures S3-S5 and Table S4) using primers, as shown in Table S3. NyHTR2 was expressed only in gametophytes and was induced by heat stress (Figure 2, Table S5). NyHTR2 expression peaked at 2 and 8 h after exposure to 25 • C ( Figure 2, Table S5). Since N. yezoensis was cultured under a short-day photoperiod (10 h light: 14 h dark) and the temperature treatments began 3 h into the light period, the NyHTR2 expression peak at 8 h occurred 1 h after the start of the dark period. Therefore, the first and second NyHTR2 expression peaks occurred under heat and heat-dark conditions, respectively. Moreover, NyHTR2L was induced by heat stress in all three generations, with expression peaks at 2 and 8 h after exposure to 25 • C heat stress ( Figure 2). NyMBF1 expression followed the same pattern in all three generations ( Figures S3-S5, Table S4). Therefore, the heat-inducible expression patterns of NyHTR2, NyHTR2L, and NyMBF1 were identical; although, temporal expression patterns of NyHTR2 during the life cycle differed from those of other genes. the first and second NyHTR2 expression peaks occurred under heat and heat-dark conditions, respectively. Moreover, NyHTR2L was induced by heat stress in all three generations, with expression peaks at 2 and 8 h after exposure to 25 °C heat stress (Figure 2). NyMBF1 expression followed the same pattern in all three generations ( Figures S3-S5 ,  Table S4). Therefore, the heat-inducible expression patterns of NyHTR2, NyHTR2L, and NyMBF1 were identical; although, temporal expression patterns of NyHTR2 during the life cycle differed from those of other genes.  Significant differences in expression level among the three life cycle generations from triplicate independent replicates as defined by Tukey's test (p < 0.05) in two-way ANOVA are shown in Table S5. G-gametophyte; S-sporophyte; CS-conchosporophyte.
We also reexamined the heat-inducible expression of NyHSP70-1 and NyHSP70-2 (see Table S3 for primers), which had been demonstrated previously [5]. In gametophytes, both genes had expression peaks at 2 h after heat exposure. NyHSP70-2, but not NyHSP70-1, also showed a second peak at 8 h after exposure ( Figure S6, Table S6). We observed similar expression patterns in sporophytes and conchosporophytes. Expression peaks in conchosporophytes occurred rapidly, at 1 h after exposure ( Figure S6, Table S6). These findings indicate that the appearance of a second expression peak under heat-dark conditions varies depending on the gene.
To further explore the drivers of the second expression peak, we changed the culture conditions to 24 h light. Under these conditions, we did not observe a second expression peak for NyHTR2, NyHTR2L, NyMBF1, or NyHSP70-2 in any generation (8 L and 12 L in Figures 3 and 4). There was no expression peak for NyHSP70-1 under either regular or 24 h light conditions (Figures 3 and 4). These findings indicate that the second expression peak requires darkness as well as heat stress. Thus, dark conditions result in a second, heatstress-dependent induction of NyHTR2, NyHTR2L, NyMBF1, and NyHSP70-2 expression after the first transient expression, which is inhibited by light illumination.

Relationship between Heat-Inducible Expression and Membrane Fluidization
Membrane fluidization, caused by heat stress, triggers heat-induced gene expressi in terrestrial plants [29][30][31][32][33]; however, it is unknown whether changes in the membrane physical state are related to gene expression in seaweeds. Thus, we examined the effec of a membrane fluidizer, benzyl alcohol (BA), and a membrane rigidizer, dimethyl sulfo ide (DMSO), on the expression of heat-inducible genes.

Relationship between Heat-Inducible Expression and Membrane Fluidization
Membrane fluidization, caused by heat stress, triggers heat-induced gene expression in terrestrial plants [29][30][31][32][33]; however, it is unknown whether changes in the membrane's physical state are related to gene expression in seaweeds. Thus, we examined the effects of a membrane fluidizer, benzyl alcohol (BA), and a membrane rigidizer, dimethyl sulfoxide (DMSO), on the expression of heat-inducible genes. Treatment with 2.5 mM BA at 15 • C did not induce expression of NyHTR2 or NyHTR2L but did (like heat stress) induce expression of NyMBF1, NyHSP70-1, and NyHSP70-2 ( Figures 5 and 6). In contrast, treatment with 4% DMSO did not induce the expression of NyHTR2, NyHTR2L, NyMBF1, or NyHSP70-2, but did induce expression of NyHSP70-1 (Figures 5 and 6). This indicates that the heat-induced expression of NyHTR2 or NyHTR2L is not dependent on membrane fluidization, whereas that of NyMBF1, NyHSP70-1, and NyHSP70-2 is fluidization-dependent.
Cells 2022, 11,1486 10 of 18 Treatment with 2.5 mM BA at 15 °C did not induce expression of NyHTR2 or Ny-HTR2L but did (like heat stress) induce expression of NyMBF1, NyHSP70-1, and Ny-HSP70-2 ( Figures 5 and 6). In contrast, treatment with 4% DMSO did not induce the expression of NyHTR2, NyHTR2L, NyMBF1, or NyHSP70-2, but did induce expression of NyHSP70-1 (Figures 5 and 6). This indicates that the heat-induced expression of NyHTR2 or NyHTR2L is not dependent on membrane fluidization, whereas that of NyMBF1, Ny-HSP70-1, and NyHSP70-2 is fluidization-dependent.    Both cold stress and DMSO rigidify membranes; however, a 5 • C exposure did not induce expression of NyHSP70-1 ( Figure S6). To further assess whether NyHSP70-1 responds to low temperatures, we examined the expression of NyHSP70-1 and other genes in algae exposed to a temperature of 0 • C. Only NyHSP70-1 was induced by 0 • C treatment ( Figure S7). Thus, NyHSP70-1 is inducible by both heat and cold, whereas the other genes we examined were only inducible by heat.
Finally, we performed two experiments to confirm whether membrane fluidization is required for the second peak seen under dark conditions. First, we subjected the three generations of algae to 25 • C and 2.5 mM BA under continuous light. This treatment induced the expression of NyHSP70-1 and NyHSP70-2, as detected 1 h after BA application, but had no effect on the expression of NyHTR2 or NyHTR2L (Figure 7). BA application also induced NyMBF1 in gametophytes but not in conchosporophytes (Figure 7). In the second experiment, we subjected the three generations of algae to heat stress, either with or without 4% DMSO, under dark conditions. In gametophytes, DMSO reduced expression levels of all genes except NyHSP70-1 (Figure 8). This is consistent with our earlier results regarding the effect of DMSO treatment on NyHSP70-1 under dark conditions (Figures 4 and  S6). Therefore, there are at least two signal transduction pathways regulating the second expression peak under dark conditions at 25 • C: a pathway involving dark-dependent membrane fluidization that governs NyMBF1 and NyHSP70-2 expression and another pathway that, though heat dependent, does not involve membrane fluidization and that governs NyHTR2 and NyHTR2L expression.
induce expression of NyHSP70-1 ( Figure S6). To further assess whether NyHSP70-1 responds to low temperatures, we examined the expression of NyHSP70-1 and other genes in algae exposed to a temperature of 0 °C. Only NyHSP70-1 was induced by 0 °C treatment ( Figure S7). Thus, NyHSP70-1 is inducible by both heat and cold, whereas the other genes we examined were only inducible by heat.
Finally, we performed two experiments to confirm whether membrane fluidization is required for the second peak seen under dark conditions. First, we subjected the three generations of algae to 25 °C and 2.5 mM BA under continuous light. This treatment induced the expression of NyHSP70-1 and NyHSP70-2, as detected 1 h after BA application, but had no effect on the expression of NyHTR2 or NyHTR2L (Figure 7). BA application also induced NyMBF1 in gametophytes but not in conchosporophytes (Figure 7). In the second experiment, we subjected the three generations of algae to heat stress, either with or without 4% DMSO, under dark conditions. In gametophytes, DMSO reduced expression levels of all genes except NyHSP70-1 (Figure 8). This is consistent with our earlier results regarding the effect of DMSO treatment on NyHSP70-1 under dark conditions (Figures 4 and S6). Therefore, there are at least two signal transduction pathways regulating the second expression peak under dark conditions at 25 °C: a pathway involving darkdependent membrane fluidization that governs NyMBF1 and NyHSP70-2 expression and another pathway that, though heat dependent, does not involve membrane fluidization and that governs NyHTR2 and NyHTR2L expression.  (25 °C ) exposure under continual light. Relative mRNA levels were normalized using eIF4A as the reference gene. In the box plots, values on the Y axis represent the fold change of relative quantification of each gene. Different alphabetical letters denote significant differences in expression level among the three life cycle generations from triplicate independent replicates as defined by Tukey's test (p < 0.05) in two-way ANOVA. G-gametophyte; S-sporophyte; CS-conchosporophyte.

Discussion
In the present study, we examined the involvement of membrane fluidity in heat-stressinducible gene expression in the red alga N. yezoensis. We identified two genes encoding NsHTR2 homologs in N. yezoensis, which we designated NyHTR2 and NyHTR2L (Figure 1). The NsHTR2 homologs contained a novel Cys-repeat motif that was conserved only in Bangiales and which we therefore designated the Bangiales Cys-repeat (BCR) motif. We identified NyHTR2 as a gametophyte-specific gene (Figure 2). Both NyHTR2 and NyHTR2L were heat-stress-inducible genes but did not depend on the membrane fluidization that generally results from heat stress, as evidenced by their lack of induction by the membrane fluidizer BA (Figures 5 and 7). In contrast, expression of NyMBF1, NyHSP70-1, and NyHSP70-2 was induced not only by heat but also by membrane fluidization (BA treatment); thus, we consider them to be membrane-fluidization-dependent expressions (Figures 7 and  S7). Therefore, we conclude that N. yezoensis has both membrane-fluidization-dependent and -independent pathways of heat-inducible expression.
We found that dark conditions stimulated both the membrane-fluidization-dependent and -independent pathways, but in different fashions. The heat-inducible genes had an initial, transient low-level expression peak at 4-6 h after heat exposure in the light (Figure 2 and Figures S3-S6). A second expression peak occurred 1 h after the dark period began; however, this did not occur when N. yezoensis was grown under continual light at 25 • C (Figures 3 and 4). This indicates that the three generations of N. yezoensis could tolerate continual heat stress after the first peak; however, dark exposure reset the heat stress response and resulted in a second expression peak. Moreover, it is likely that membrane fluidization is involved in the second heat-induced expression peak of NyMBF1 and NyHSP70-2 because BA treatment also induced a second expression peak of these genes under continual light, whereas DMSO reduced the second expression peak in the dark period (Figures 7 and 8).
In contrast, the second, dark-induced expression peaks of NyHTR2 and NyHTR2L were not sensitive to BA application, indicating that they are independent of membrane fluidization. However, DMSO treatment reduced the dark-induced expression of these genes ( Figure 8). DMSO is a membrane rigidizer and is often used to study physiological responses to membrane rigidification, such as cold-stress-inducible gene expression [34,49]. Therefore, the fact that DMSO reduced NyHTR2 and NyHTR2L expression is inconsistent with the BA treatment results. In the unicellular green alga Chlamydomonas reinhardtii, transcripts of HSF1, HSP70A, HSP90A, and HSP90C accumulated under treatment with heat plus DMSO, an effect similar to that of the cytosolic translation inhibitor cycloheximide. Therefore, DMSO-dependent translational inhibition is proposed to be responsible for the enhanced expression of these genes caused by DMSO [50,51]. It is possible that translational inhibition by DMSO is also related to NyHTR2 and NyHTR2L expression in N. yezoensis; however, this remains to be confirmed.
On the basis of our findings, we conclude that there are multiple pathways regulating heat-stress-inducible gene expression in N. yezoensis. One is a membrane-fluidizationdependent pathway, which can be subdivided into a dark-enhanced pathway for NyMBF1 and NyHSP70-2 and a dark-independent pathway for NyHSP70-1. Similarly, moss membrane proteins, including the cyclic nucleotide-gated Ca 2+ channels CNGCb and CNGCd, are membrane-fluidization-dependent heat-sensing molecules [29,36,52]. Therefore, it could be valuable to explore the involvement of Ca 2+ influx and cyclic nucleotide-gated Ca 2+ channels in early heat stress responses in N. yezoensis. The other pathway regulating heat-stress-inducible gene expression in N. yezoensis is a membrane-fluidizationindependent pathway directing NyHTR2 and NyHTR2L expression, which is enhanced by dark conditions. The discovery of this pathway was unexpected because heat stress responses are generally thought to be triggered by membrane fluidization [30][31][32][33]52]. Thus, further examination of heat-stress-inducible expression of NyHTR2 and NyHTR2L might uncover novel mechanisms of heat stress responses in photosynthetic organisms.
Alone among the genes examined in this study, NyHSP70-1 was induced by both DMSO treatment and cold stress (Figures 6 and S7), indicating that it is triggered by coldstress-induced membrane rigidification. Similar cold-stress-inducible expression of HSP genes has been reported in various plant species [41,[53][54][55][56][57][58]. In our study, NyHSP70-1 expression was also induced by heat stress and was membrane-fluidization-dependent, but it was not affected by dark (Figures 4, 6-8 and S6). NyHSP70-1 is therefore an ideal model in which to investigate the regulatory mechanisms of heat and cold-stress-inducible gene expression based on physical changes in membranes in Bangiales.
Extracellular Ca 2+ influxes and reactive oxygen species (ROS) are involved in the early phases of heat stress responses in terrestrial plants [59][60][61][62][63]. These can be triggered by membrane fluidization [34][35][36]52,64]. However, little is known about how these signaling molecules may be involved in either of the heat-stress-induced gene expression pathways in N. yezoensis. Therefore, further examination of the membrane-fluidization-dependent and -independent pathways could provide novel insights into stress perception and early events in heat-stress-inducible gene expression in Bangiales.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/cells11091486/s1, Figure S1: Conservation of the Bangiales cysteinerich (BCR) motif in HTR2 homologs from Neopyropia yezoensis, Neopyropia seriata, and Porphyra umbilicalis; Figure Figure S3: Temperature-dependent expression patterns of NyMBF1 in Neopyropia yezoensis gametophytes; Figure S4: Temperaturedependent expression patterns of NyMBF1 in Neopyropia yezoensis sporophytes; Figure S5: Temperaturedependent expression patterns of NyMBF1 in Neopyropia yezoensis conchosporophytes; Figure S6: Temperature-dependent expression patterns of NyHSP70-1 and NyHSP70-2 with temporal comparisons in three life cycle generations of Neopyropia yezoensis; Figure S7: Induction of NyHSP70-1 by exposure to 0 • C in three life cycle generations of Neopyropia yezoensis; Table S1: Sequences of primers for candidates of reference genes used for qRT-PCR; Table S2: Expression stability analysis of four reference genes by NormFinder in three life cycle generations of Neopyropia yezoensis; Table S3: Sequences of primers for heat stress-inducible genes used for qRT-PCR; Table S4: Temperature-dependent gene expression profiles of NyMBF1 in three life cycle generations of Neopyropia yezoensis normalized using reference genes; Table S5: Gene expression profiles of NyHTR2 and NyHTR2L in three generations of Neopyropia yezoensis under heat stress normalized using eIF4A as the reference gene; Table S6: Gene expression profiles of NyHSP70-1 and NyHSP70-2 in three generations of Neopyropia yezoensis under heat stress normalized using eIF4A as the reference gene.