Naoki Sato

We previously showed that the chloroplast of the glaucophyte Cyanophora paradoxa does not have fatty acid synthesis activity, based on both experimental and bioinformatic analyses. The present study was intended to extend the bioinformatic analysis to a wider range of Glaucophyta, using published genomic data. The results confirmed that the four species of Glaucophyta analyzed do not have prokaryotic acetyl-CoA carboxylase, which is found in the chloroplasts of Rhodophyta and Viridiplantae. Only the eukaryotic type of acetyl-CoA carboxylase was identified. Some components of prokaryotic fatty acid synthase (FAS II), such as acyl carrier protein (ACP), 2-oxoacyl ACP synthase (KAS I/II), and FabD, were detected in the glaucophytes, but other components especially KAS III were not detected. In contrast, the eukaryotic fatty acid synthase (FAS I) was found in Gloeochaete as well as Cyanophora. Although the genome data is still to be amended, the available evidence suggests that the lack of fatty acid synthesis in the chloroplast is a trait of not just Cyanophora but also shared by the phylum Glaucophyta. The relationship between photosynthesis and fatty acid synthesis is discussed with special attention to the origin of chloroplasts.

Triacylglycerol (TAG) has been frequently reported in cyanobacteria; however, unlike in plants and algae, the isolation of TAG as a pure substance and its subsequent chemical characterization have proven challenging. The slr2103 gene in Synechocystis sp. PCC 6803, which encodes a putative acyltransferase involved in TAG biosynthesis, has been considered evidence for the existence of TAG. However, the identification of acylplastoquinol (APQ) as the major component of the “TAG fraction” obtained through thin-layer chromatography (TLC) has raised questions about the actual presence of TAG in cyanobacteria. To address concerns regarding potential chemical and biological contamination in the detection of TAG in cyanobacteria, we developed 1D- and 2D-TLC methods to separate submicromole quantities of TAG and APQ from Synechocystis cells. Both compounds were convincingly identified using NMR and LC/MS. TAG levels depended upon culture conditions. Well-aerated cyanobacterial cultures exhibited minimal TAG levels, while TAG accumulation reached approximately 1% of total lipids in static or slowly swirled senescing cultures, where photosynthetic activity had declined substantially. Under these conditions, we observed numerous lipid globules, approximately 72 nm in diameter, located at the periphery of the cells. These findings provide critical insights into TAG and APQ accumulation in cyanobacteria, elucidating the role of lipid globules and offering perspectives on TAG biosynthesis in cyanobacteria, as well as the potential function of APQ in photosynthesis.

1IntroductionThis is a general commentary to the publication by Ito et al. (2025). Plastoquinone serves as a crucial electron carrier in the photosynthesis of cyanobacteria and chloroplasts. Plastoquinone-B (PQ-B) is specifically known for containing an acyloxy group within its prenyl group. Acylplastoquinol (APQ) is an ester of the reduced form of plastoquinone (Mori-Moriyama et al. 2023). The structure of APQ was established by 1H- and 13C-NMR; however, the determination of its isomeric structure is still pending. APQ has been confirmed by two independent research groups through LC/MS analysis (Ishikawa et al. 2023, Kondo et al. 2023). In reviewing the paper by Ito et al. (2025), I found some data to re-examine seriously as detailed below.2Critical examination of the paper2.1Quality of mass spectral dataFigure 1B is an MS/MS spectrum of palmitoyl plastoquinol (16:0-APQ) in Cyanidioschyzon merolae presented as Figure 1 in Ito et al. (2025). The signal intensity was extremely low, as evidenced by the row of low peaks, each representing 1 count, the minimum unit of digital data. The base peak at m/z=153 has only 10 counts. Furthermore, many other spectra (both APQ and PQ-B) exhibit a similar lack of quality. This suggests a very low quantity of the target substance, which could easily be attributed to cross contamination. Ito et al. (2025) detected primarily saturated APQ, which is also questionable. Mori-Moriyama et al. (2023) and Tanikawa et al. (2025) showed that Synechocystis APQ contains both saturated and unsaturated fatty acids. However, Figure 3 in Ito et al. (2025) only presents 16:0- and 18:0-APQ. This discrepancy may arise from the analytical method used. The authors analyzed the total lipid fraction by LC/MS, where each molecular species of APQ appears as separate peaks. The smaller peaks corresponding to unsaturated APQ may be obscured by overlapping peaks from glycerolipids and pigments. Additionally, the fragmentation pattern of unsaturated APQ differs from that of saturated APQ, complicating their detection. It would be more effective to isolate APQ first and then analyze the molecular species using LC/MS.I found it strange that no [M+NH4]+ signal was detected in the MS/MS spectra presented in Ito et al. (2025) , while a clear [M+NH4]+ signal was consistently observed in the MS data of Synechocystis APQ reported by the same group (Kondo et al. 2023). The signal for the de-prenylated fragment (m/z=389 for 16:0-APQ) was either not observed or very weak in Ito et al. (2025). The de-prenylated fragment is crucial in identifying the APQ molecular species, as the acyl fragment is not a reliable marker (see the next section). If the [M+NH4]+ signal was indeed ionized correctly, we would expect to see the same signal as a prominent parent ion in the MS/MS spectra. Given the very low intensity of the signal and the discrepancies observed, I suspect that the equipment may not have been properly operated. The same argument applies to the MS/MS spectra of PQ-B.2.2Acyl fragmentThe assignment of the acyl-derived fragment in the APQ mass spectrum remains enigmatic. Namely, 16:0-APQ yields a fragment with an m/z=237 (C14H29C+=C=O), rather than expected 239, which is typically found for the palmitoyl fragment (C15H31C=O+) resulting from esters. Ito et al. (2025) noted that the acyl fragment generated from 16:0-APQ has an m/z=237, but they did not provide an explanation for this assignment (Fig. 1A, which is adapted from the original Fig. 1 with annotations). Ishikawa et al. (2023) identified this fragment as the RC+=C=O ion, yet they also failed to clarify the underlying mechanism. Currently, there is no established explanation for this unusual fragment. The fragmentation scheme in Figure 1 in Itoh et al. (2025) will have to be corrected.2.3Slr2103 OrthologsIto et al. (2025) stated that the presence of APQ has been demonstrated in only four species of cyanobacteria. However, Tanikawa et al. (2025), published online prior to the submission of Ito et al. (2025), identified additional cyanobacterial species that also produce APQ, including Gloeobacter, which lacks slr2103 orthologs. Furthermore, while Ito et al. (2025) claimed that slr2103 orthologs are only conserved within cyanobacteria, Mori-Moriyama et al. (2023) highlighted that the plant PES1, which has an extra domain, is closely related to the cyanobacterial Slr2103 family. This suggests that both plants and algae have the potential to synthesize APQ.2.4Other pointsAPQ is an unexpectedly unstable substance that should be handled with caution. Its instability is likely due to its susceptibility to oxidation, which leads to de-acylation. The amount of APQ can decrease during concentration process such as evaporation, drying of thin-layer plates, or other manipulations commonly used in lipid analysis. The low APQ content suspected in the data of Ito et al. (2025), along with the high variability in quantitative data observed in Tanikawa et al. (2025), may result from degradation during these manipulations. It is essential to establish a reliable method for the quantitative analysis of APQ in the future.3DiscussionAPQ is a recently discovered substance that remains challenging to analyze. The analysis of APQ should be conducted with care, utilizing various methods rather than relying solely on a single technique such as LC/MS. In this context, I would like to draw the readers' attention to a recent paper by Das et al. (2025), which unfortunately did not provide any methodological details or mass spectrometry data. I encourage the authors of Ito et al. (2025) to improve their analysis by carefully revising or correcting their initial findings. This would be a constructive step forward in advancing research on the newly identified compounds known as plastoquinone-related lipids.

Archaeplastida, a group of photosynthetic organisms with primary plastids, consists of green algae (plus plants), red algae, and glaucophytes. In contrast to green and red algae, information on lipids and lipid biosynthesis still needs to be included in the glaucophytes. The chloroplast is the site of photosynthesis and fatty acid synthesis in all photosynthetic organisms known to date. However, the genomic data of the glaucophyte Cyanophora paradoxa suggested the lack of acetyl CoA carboxylase and most components of fatty acid synthase in the chloroplast. Instead, multifunctional fatty acid synthase and acetyl CoA carboxylase are likely to reside in the cytosol. To examine this hypothesis, we measured fatty acid synthesis in isolated chloroplasts and whole cells using stable isotope labeling. The chloroplasts had very low activity of fatty acid synthesis, if any. Most processes of fatty acid synthesis, including elongation and desaturation, must be performed within the cytosol, and the fatty acids imported into the chloroplasts are assembled into the chloroplast lipids by the enzymes common to other algae and plants. Cyanophora paradoxa is a rare organism in which fatty acid synthesis and photosynthesis are not tightly linked. This could question the common origin of these two biosynthetic processes in Archaeplastida

Chloroplasts are usually considered spheroid organelles, but this is not the only shape of chloroplasts. The chloroplast of Chlamydomonas has been typically described as cup-shaped. However, in old studies, it was also modeled as a complex shape with “perforations” or windows. Here, we reconstructed the cellular architecture of Chlamydomonas reinhardtii and C. applanata using an array tomography system installed on a field emission scanning electron microscope. C. reinhardtii chloroplasts resembled a baseball glove or a cup without a side, featuring numerous large and small holes that may facilitate the transport of metabolites and proteins produced in the Golgi apparatus fitted in the holes. In a lipid-accumulating, high-light condition, the chloroplast volume increased by filling the side cleft with an entire wall. Many accumulated large lipid droplets were accommodated within the chloroplast holes, which could have been considered as “chloroplast lipid droplets.” Mitochondrial meshworks surrounded the chloroplast. C. applanata chloroplasts appeared like a folded starfish or a cup with many side clefts and a few holes. There was a single mitochondrion or two that branched in a complex form. Tight contacts of various organelles were also found in C. applanata. These reconstructions illustrate the complexity of chloroplast shape, which necessitates a revised understanding of the localization of lipid droplets and the evolution of chloroplasts: The prevailing image of the spheroid chloroplasts that reminds us of the similarity between chloroplasts and cyanobacteria is no longer tenable.

Plastids in vascular plants have various differentiated forms, among which amyloplasts are crucial for starch storage and plant productivity. Despite the vast knowledge of the binary-fission mode of chloroplast division, our understanding of the replication of non-photosynthetic plastids, including amyloplasts, remains limited. Recent studies have suggested the involvement of stromules (stroma-filled tubules) in plastid replication when the division apparatus is faulty. However, details of the underlying mechanism(s) and their relevance to normal processes have yet to be elucidated. Here, we developed a live analysis system for studying amyloplast replication using Arabidopsis (Arabidopsis thaliana) ovule integuments. We showed the full sequence of amyloplast development and demonstrated that wild-type amyloplasts adopt three modes of replication, binary fission, multiple fission, and stromule-mediated fission, via multi-way placement of the FtsZ ring. The minE mutant, with severely inhibited chloroplast division, showed marked heterogeneity in amyloplast size, caused by size-dependent but wild-type modes of plastid fission. The dynamic properties of stromules distinguish the wild-type and minE phenotypes. In minE cells, extended stromules from giant amyloplasts acquired stability, allowing FtsZ ring assembly and constriction, as well as the growth of starch grains therein. Despite hyper-stromule formation, amyloplasts did not proliferate in the ftsZ null mutant. These data clarify the differences between amyloplast and chloroplast replication and demonstrate that the structural plasticity of amyloplasts underlies the multiplicity of their replication processes. Furthermore, this study shows that stromules can generate daughter plastids via assembly of the FtsZ ring.

Various studies have suggested the presence of triacylglycerol in cyanobacteria, but no convincing evidence exists. We purified a substance co-migrating with triacylglycerol and determined its structure using mass spectrometry, gas chromatography, and ¹ H- and ¹³ C-NMR. The major components were palmitoyl and stearoyl plastoquinols (acyl plastoquinol). Acyl plastoquinol has never been described before. The level of acyl plastoquinol was 0.4% of the total lipids. We still do not have clear evidence for the presence of triacylglycerol. If present, the maximum triacylglycerol level, must be at most 10% of acyl plastoquinol or less. The Synechocystis Slr2103 protein was suggested to synthesize triacylglycerol, but the product could be acyl plastoquinol. The possible roles of this novel compound in photosynthesis should be a new focus of research.

The order Parmales (Bolidophyceae) is a minor group of pico-sized eukaryotic marine phytoplankton that contains species with cells surrounded by silica plates. Previous studies revealed that Parmales is a member of ochrophytes and sister to diatoms (Bacillariophyta), the most successful phytoplankton group in the modern ocean. Therefore, parmalean genomes can serve as a reference to elucidate both the evolutionary events that differentiated these two lineages and the genomic basis for the ecological success of diatoms vs. the more cryptic lifestyle of parmaleans. Here, we compared the genomes of eight parmaleans and five diatoms to explore their physiological and evolutionary differences. Parmaleans were predicted to be phago-mixotrophs. By contrast, diatoms have undergone loss of genes related to phagocytosis, indicating the ecological specialization from phago-mixotroph to photoautotroph in the early evolution of diatoms. Furthermore, diatoms showed significant enrichment in gene sets involved in silica metabolism, nutrient uptake capacity, carbon concentrating mechanisms, and iron uptake in comparison with parmaleans. Overall, our results suggest a strong evolutionary link between the loss of phago-mixotrophy and specialization to a silicified photoautotrophic life stage early in diatom evolution after diverging from the Parmales lineage.