Multifaceted Photoreceptor Compositions in Dual Phototrophic Systems : a Genomic Analysis

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Multifaceted Photoreceptor Compositions in Dual Phototrophic Systems : a Genomic Analysis © 2023 the Authors Published version Ihalainen, Janne A.; Dogan, Batuhan; Kurttila, Moona; Zeng, Yonghui; van Elsas, Jan Dirk; Nissinen, Riitta Ihalainen, J. A., Dogan, B., Kurttila, M., Zeng, Y., van Elsas, J. D., & Nissinen, R. (2024). Multifaceted Photoreceptor Compositions in Dual Phototrophic Systems : a Genomic Analysis. Journal of Molecular Biology, 436(5), Article 168412. https://doi.org/10.1016/j.jmb.2023.168412 2024 Multifaceted photoreceptor compositions in dual phototrophic systems – A genomic analysis Janne A. Ihalainen 1, ⇑ , Batuhan Dogan 1 , Moona Kurttila 1,† , Yonghui Zeng 2 , Jan Dirk van Elsas 3 and Riitta Nissinen 1,4 1-University of Jyva ¨skyla ¨,Nanoscience Center, Department of Biological and Environmental Science, 40014 Jyva ¨skyla ¨, Finland 2-University of Copenhagen, Department of Plant and Environmental Sciences, 2100 Copenhagen, Denmark 3-University of Groningen, Groningen Institute for Evolutionary Life Sciences, 9747 AG Groningen, the Netherlands 4-University of Turku, Department of Biology, 20500 Turku, Finland Correspondence to Janne A. Ihalainen: [email protected] (J.A. Ihalainen) https://doi.org/10.1016/j.jmb.2023.168412 Edited by: John Kennis Abstract For microbes and their hosts, sensing of external cues is essential for their survival. For example, in the case of plant associated microbes, the light absorbing pigment composition of the plant as well as the ambient light conditions determine the well-being of the microbe. In addition to light sensing, some microbes can utilize xanthorhodopsin based proton pumps and bacterial photosynthetic complexes that work in parallel for energy production. They are called dual phototrophic systems. Light sensing requirements in these type of systems are obviously demanding. In nature, the photosensing machinery follows mainly the same composition in all organisms. However, the specific role of each photosensor in specific light conditions is elusive. In this study, we provide an overall picture of photosensors present in dual phototrophic systems. We compare the genomes of the photosensor proteins from dual phototrophs to those from similar microbes with ”single” phototrophicity or microbes without phototrophicity. We find that the dual phototrophic bacteria obtain a larger variety of photosensors than their light inactive counterparts. Their rich domain composition and functional repertoire remains similar across all microbial photosensors. Our study calls further investigations of this particular group of bacteria. This includes protein specific biophysical characterization in vitro, microbiological studies, as well as clarification of the ecological meaning of their host microbial interactions. Ó2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Introduction Photosynthesis performed by plants, algae, and cyanobacteria provide the organic carbon, and therefore the energy needed for all other life on earth. While in this case light energy is converted into chemical energy through carbon fixation, many microrganisms also have machinery to store energy through light-driven proton gradients across the bacterial membrane either via anoxygenic photosynthesis (chlorophototrophs) typically utilizing bacteriochlorophyll a(BChl a) molecules or retinal-rhodopsin-based proton pumping (retinalphototrophs). The light-driven redox reactions produce a proton-motive force for chemical energy production in the form of ATP. 1 Organisms that utilize light as an energy source but are still dependent on organic carbon from their host are called photoheterotrophs. Intriguingly, several bacterial strains have been reported with both of these phototrophic machinery. They are referred to as dual photoheterotrophs. 2 Research Article 0022-2836/Ó2023 The Author(s). Published by Elsevier Ltd.This is an open access article underthe CC BY license (http://creativecommons.org/licenses/ by/4.0/). Journal of Molecular Biology 436 (2024) 168412 As the name says, anoxygenic photosynthesis does not produce oxygen as a side product. Still, some anoxygenic phototrophs can live in aerobic environments. Typically, the biosynthesis of the photosynthetic apparatus is suppressed by oxygen. However, organisms that strictly need oxic conditions for photosynthetic apparatus production and for photosynthetic electron transport are called aerobic anoxygenic phototrophs (AAP). 1 These bacteria live in aerobic conditions and rely on diverse organic compounds as electron donors for the phototrophy. 3 AAP bacteria (AAPB) have been originally discovered in oligotrophic marine environments, 4 and are ubiquitous in both marine and freshwater ecosystems. 5,6 In aquatic ecosystems, they can constitute up to a third of the total bacterial populations. They are considered to play a significant role in carbon transformations in marine ecosystems. 3 AAPB have also been found in terrestrial systems, including soil biocrusts, Antarctic soils and plants. 7,8 AAPB have been detected in phyllosphere metagenomes, and are commonly present in epiand endophytic microbial communities of diverse plant species in arctic and boreal regions. 8 Proteorhodopsin genes have been identified in all three domains of life, but they are mainly studied from marine bacteria. 9,10 Microbial rhodopsins are recognized to be present in about 50% of all “heterotrophic” bacteria living in the surface ocean. 10 Further, terrestrial and glacial proteorhodpsins have been reported, 11,12 but studies on terrestrial retinalphototrophs are rare. Environmental factors impacting AAPB abundance, diversity and their photosynthesis are still relatively poorly understood. The AAP is driven by bacteriochlorophylls and as such they can utilize near infrared (NIR) light as an energy source. Rhodopsins consist essentially of one protein, opsin, and a maximum of two chromophores, a retinal and a carotenoid molecule, in the case of xanthorhodopsins. 13 The retinal has no cofactors and is produced in one metabolic step. 10 Under green light, the retinal pigment isomerizes and provokes a proton gradient across a membrane. Hence, in terms of excitation wavelengths, Chlorophyll a(Chl a) light absorption, taking place in plants, algae or cyanobacteria, do not disturb the functionality of AAPB or retinalphototrophs, and in principle, all photosystems can work in parallel in microbe-host systems. It is noteworthy that making rhodopsin-based proton pumps is rather ”cheap”, while making photosynthesis machinery for bacteria is ”expensive”, as the protein cluster for the latter is much larger than for the former one. 10,14 However, the latter can collect light much more efficiently and the turnover rate per absorbed photon is considerably higher than for a rhodopsin. The setting for host associated dual phototrophs - BChl aand retinal pigment containing bacterial systems in a Chl abased host - provides multifaceted requirements for bacterial photosensing. For example, they need to signal the fluency, intensity variability and spectral properties of light to their host. This in addition to sensing temperature and other external factors is necessary for optimal growth and functioning of microorganisms, and further benefits the well-being of the host. The wavelength sensitivity is conducted by using different photosensor proteins for each wavelength range. Figure 1 lists the photosensors which detect light at each wavelength range. The general mechanism of all photosensors follows the same pattern. Photon absorption by the chromophore leads to physical changes of the chromophore and its nearby environment. The flavin-containing systems, Light-Oxygen-Voltage sensors (LOV), Blue-Light Using Flavin domain (BLUF), and Cryptocrome/Photolyase (CRY/PHR) complexes rely on the rupture of a covalent bond (LOV) or charge transfer processes (BLUF, CRY/PHR) after photon absorption, whereas the rhodopsins, xanthopsins and bacteriophytochromes (BphP), rely on the photoinduced isomerization process of the retinal, p-coumaric acid, and bilin chromophore, respectively. 15 The changes in the chromophore environment induce changes in the protein environment of the photosensory domain. The biochemical signaling is performed by the effector domain of the complex. The effector domain often binds (or separates) to (from) its cognate response regulator (RR) functioning as a gene expression regulator. Other mechanisms are for example binding to other induction or repression system (LOV), or cysteinyl-adduct formation (LOV), or direct DNA binding (Photolyases, LOV), to mention just a few mechanisms. In the case of histidine kinases, a phosphotransfer reaction between the HK unit and RR takes place. 16 To conclude, the photosensory domain architeture is rather conserved across the whole kingdom of life, whereas the effector domain variability is large. A plethora of functions are controlled by the effector domains and the interplay with their cognate RR or other suppressor/effector domains. Excellent studies on the variability of microbial photosensory systems and their functional mechanisms are available. 15,17–21 In this study, we provide a genomic analysis of photosensor composition in dual phototrophic bacteria. We compare the presence and multiplicity of photosensors of bacteriophytochromes (BphP), Light-oxygen-Voltage (LOV), Blue light photosensor (BLUF), and Cryptochrome/ Photolyase (CRY/PHR) systems, and photoactive yellow proteins (PYP) in dualphototrophic systems with those of similar species without dual phototrophic character (Fig. 1 and Tables 1 and S1). Further, we aim to pinpoint the differences and similarities of genome sequence levels of each photosensor among the dual phototrophic systems, some of which are newly discovered from plant associated bacteria. J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 2 Results Selection of dual phototrophic species - their origin and environmental conditions We performed whole genome sequencing for selected Sphingomonas strains previously identified as AAPB, based on our NIR fluorescence analysis. 8 The AAP positive bacteria were confirmed by presence of the reaction center specific pufM gene. When a xanthorhodopsin gene was observed in addition to the pufM gene, the species was identified as dual phototrophic system, similar to AAP5 strain reported by Koblizek and colleagues. 2 Hence, dual phototrophs are strains which contained both AAP and xanthorhodopsin gene patterns. We elaborated our analysis using the National Center for Biotechnology (NCBI) data base. We identified more dual phototrophic systems from the data base, but also strains with AAP genes only, strains with xanthorhodopsin genes only, and heterotrophic strains which did not contain any photoactive proton pumping system, but belonged to same genuses as those of photoheterotrophs. In our analysis, we have overall 62 strains, which contained more than 300 photosensors. All studied species with their access number, code used in this study, and source are presented in Table S1. Table 1 lists strains under comparison in this study. All AAPs and dual phototrophs contain genes for light harvesting complexes (LH), either LH1 or LH1 and LH2 genes. We name as ”Rho” strains those which contain only xanthorhodopsin gene. In addition to those, 23 similar strains but without phototrophicity are also listed. Peculiarly, strains TA304 and TA352 contain LH2 gene even though they lack photosynthetic reaction center genes, 12 and even more strikingly, OCH149 is a non-photoheterotroph but still contains LH2 gene. Interestingly, the dual phototrophs found in this work mainly originate from cold and harsh conditions with perpetual growth in dynamically changing habitats where irradiance and temperatures vary significantly across seasons. The Tardiphaga strains are collected from soils of glaciers in North East Greenland, with extremely short growing seasons but with ample amount of light during the growth period. 12 The Sphingomonas strains defined as dual phototrophs are either plant associated bacteria originating from arctic plants, collected in the northern region of Finland (J1U1, M1U20), 22 or from Svalbard (L1CD2B). The Sphingomonas strains which are not plant associated bacteria, but are dual phototrophs, originate either from permafrost in arctic Canada region (CGMCC and SE61), or from high altitude alpine lake in Austria (AAP5). 2 Photoreceptor gene collection Dual phototrophy requires precise and elaborate photosensing capabilities, as the organisms need to control which of the photosystems is synthesised. In addition, several of the studied strains J1U1, M1U20 and L1CD2B, which all belong to the bacterial family Sphingomonas, live in changing light conditions as they have been Fig. 1. Light, phototrophic systems and photoreceptors. In light sensing bacteria, photoreceptors and light harvesing systems work in parallel. The spectral regions utilized by bacterial photosynthesis systems, xanthorhodopsin and bacterial photosensors are marked as black arrows. The white arrows mark the spectral regions of ”competing” absorbance of Chl aand bmolecules The basic domain composition of the photosensors are shown as well. Rarely, a direct link between the phototrophic components with photosensors has been elucidated. Still, photoreceptors must play a decisive role which phototrophic system becomes active. J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 3 Table 1 The photoreceptor collection of the studied species, their origin and codes used in the phylogenetic trees. The NCBI accession numbers are listed in Supplementary Table S1. The dark grey color indicates the dual phototrophs, the grey the AAP systems and the light grey the systems with only xanthorhodopsins. The strains are ordered according to their total number of photosensor proteins. Source Genus Strain BLUF LOV BphP CRY PYP Category LH High arctic permafrost Canada Sphingomonas SE61 10 2 2 2 1 Dual LH1 Leaf of Oryza sativa Methylobacterium CBMB27 4 2 2 3 – AAP LH1 Stem surface of Oryza sativa Methylobacterium CBMB20 5 1 2 3 – AAP LH1 Antarctic sea ice Octadecabacter O307 5 – – 5 – Rho – High elevation lake Austria Sphingomonas AAP5 4 1 1 2 1 Dual LH1 Antarctica soil Sphingomonas PAMC26645 3 2 3 2 – None – Nodules of Lotononis bainesii Methylobacterium MS446 5 – 2 2 – Dual LH1 Greenland stream water Gemmatimonas TET16 4 – 1 4 – AAP LH1 High arctic permafrost Canada Sphingomonas CGMCC 3 1 1 2 1 Dual LH1 Glacial ice, NE Greenland Tardiphaga TA278 1 – 5 2 – Dual LH1,LH2 Wastewater Methylobacterium C1 3 – 2 2 – AAP LH1 Arctic plant endosphere, Kilpisja ¨rvi Sphingomonas S2H28 1 – 4 1 1 Rho – Glacial ice, NE Greenland Tardiphaga TA154 2 – 3 2 – Dual LH1,LH2 Contaminated soil Methylorubrum CM4 5 1 1 1 – AAP LH1 Arctic plant phyllosphere, Svalbard Sphingomonas L1CD2B 2 – 2 2 1 Dual LH1 Malus prunifolia Sphingomonas AP4 2 1 2 2 – None – Arctic sea ice Octadecabacter O238 2 – – 4 – Rho – Arctic plant endosphere, Kilpisja ¨rvi Sphingomonas J1U1 1 – 3 2 – Dual LH1 Glacial ice, NE Greenland Tardiphaga TA304 1 – 3 2 – Rho LH2 Glacial ice, NE Greenland Tardiphaga TA352 1 – 3 2 – Rho LH2 Plant associated Sphingomonas AK 2 1 2 2 – None – Tidal flat sediment Roseobacter YSTF 2 1 1 3 – AAP LH1,LH2 Arctic plant endosphere, Kilpisja ¨rvi Sphingomonas M1U20 1 – 3 2 – Dual LH1 Cornfield topsoil Xinjiang Sphingomonas NX02 1 – 2 2 – Dual LH1 Freshwater water column Limnohabitans G32 3 – – 3 – AAP LH1 Desert lake in north China Gemmatimonas AP64 2 – – 3 1 AAP LH1 Tidal flat area Roseobacter B14 1 1 – 3 – AAP LH1,LH2 Arctic plant endosphere, Kilpisja ¨rvi Sphingomonas J5HS3a 1 – 1 2 1 Rho – Marine coastal samples Dokdonia MED134 1 – 1 2 – Rho – Freshwater water column Limnohabitans B93 2 – – 3 – AAP LH1 Arctic plant endosphere, Kilpisja ¨rvi Sphingomonas S2U11 1 – 3 1 – AAP LH1 Freshwater water column Limnohabitans JIRII-31 2 – – 2 – AAP LH1 Plant associated Methylobacterium AM1 2 1 1 1 – AAP LH1 Marine coastal samples Congregibacter KT71 2 – – 3 – AAP LH1 Water column Roseobacter OCH149 1 1 – 3 – None LH2 Water column Dinoroseobacter DFL12 1 1 – 2 – AAP LH1,LH2 Root nodule Tardiphaga LMG 2 – 1 1 – None – Unknown Tardiphaga OK246 2 – 1 1 – None – Sea surface water, Japan Roseobacter AI77 – – 1 3 – AAP LH1 Freshwater water column Limnohabitans 15 K 1 – – 3 – AAP LH1 Water column Roseobacter OCH114 1 1 – 2 – AAP LH1,LH2 Freshwater Reservoir Limnohabitans 2D5 2 – – 2 – AAP LH1 Root nodule Tardiphaga TA581 1 – 1 1 – None – Root nodule Tardiphaga TA37S4 1 – 1 1 – None – Unknown Tardiphaga OK245 1 – 1 1 – None – Water column Pelagibacter SAR11 – – – 2 – Rho – Soil of Hengshui Lake Sphingomonas WHSC8 1 – 1 1 – AAP LH1 Arctic plant endosphere, Kilpisja ¨rvi Sphingomonas S3H21 – – 1 1 1 Rho – Root Tardiphaga P911 1 – – 1 – None – Unknown Sphingomonas TY – – 1 1 – None – Plant associated Tardiphaga VAF07 – – 1 1 – None – Sewage treatment plant Gemmatimonas T27 – – – 1 1 None – Antarctica soil Sphingomonas SO64 1 – – 1 – None – Plant associated Sphingomonas DCY99 – – 1 1 – None – Plant associated Sphingomonas CRA20 1 – 1 – – None – Freshwater lake Sphingomonas LM7 1 – 1 – – None – Alpine soil Sphingomonas DSM22537 1 – – 1 – None – Waste water bioreactor Sphingomonas CL51 1 – – – – None – Activated sludge Sphingomonas KC8 – – – 1 – None – J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 4 found either from plant leaf internal tissues (endosphere) or from plant leaf surfaces (phyllosphere) and therefore contain changing amounts of Chl molecules in their environment. The photosensing is performed by BLUF, LOV, BphP, Cryptochrome/Photolyase (CRY/PHR), and PYP photosensors. Table 1 shows the numbers of different photoreceptor proteins for each studied bacterial strain. Further, the Fig. 2 presents the average number of all photosensors as well as the average number of each photosensor in each photoheterotrophic category. Table 1 and Fig. 2 demonstrate that the photoheterotrophs have a larger amount and variety of photosensors in comparison to heterotrophs, and dual photostrophs have the largest amount of photosensors. In our analysis, dual phototrophs have on average about eight photosensors whereas single phototrophs, i.e. AAP or xanthorhodopsin based phototrophs (”Rho” category), have about five to six photosensors. Heterotrophs (”none” category in Fig. 2A) have on average about three photosensor proteins. Fig. 2B shows that all phototrophs contain about two CRY/PHR photosensors on average, the dual phototrophs and AAPs have more BLUF sensors than xanthorhodopsin based photoheterotrophs. The dual phototrophs have also the largest average number of BphP systems. The average number of PYP and LOV domains is low on average; however, variation can be observed so that AAP and dual phototrophs have a slightly higher LOV domain value than heterotrophs. We could not identify any LOV domains for ”Rho” category. Except for one case, all phototrophs contain multiple CRY/PHR, BLUF, or BphP genes. Several phototrophs contained two LOV domains, but many did not contain any LOV domains. The number CRY/PHR copies is generally two in each strain. Larger variation is observed with BLUF and BphP sensors. Often, with a relatively large number of BLUF genes a low number BphPs can be found, and vice versa. Some of the studied strains show genes for PYP. Often, PYPs are linked with BphPs. 23 However, in our search for gene locations, no strains’ PYP genes located or linked to BphPs. Recently, a wider genomic analyTable 1 (continued) Source Genus Strain BLUF LOV BphP CRY PYP Category LH Oceanic crustal fluid Gemmatimonas AH-D16 – – – – – None - Contaminated soil Sphingomonas MM1 – – – – – None – Plant associated Sphingomonas DM2 – – – – – None – Abbreviations: BLUF - blue light using flavin; LOV - light oxygen voltage; BphP - bacterial phytochrome; CRY - (CRY/PHR, cryptochrome/photolyase; PYP - photoactive yellow protein; Rho - xanthorhodopsin; LH - light harvesting complex Fig. 2. Numerical distribution of photosensors according to different phototrophic groups. (A) The number of photosensor-encoding genes vary depending on the type of phototrophy. The median line show the skewed (asymmetric) distribution of the number of photosensors. In ”None” group, the median is 2. The first quartile marks one end of the box and the third quartile marks the other. The segment lines describe the highest and lowest number of the photosensors and the outliers which were observed in the case of ”Dual” and ”None” groups. On average (the cross), there are 8.2 photosensors in dual phototrophs while AAPB and xanthorhodopsin-based phototrophs (”Rho” category) have 5.9 and 5.4, respectively. In the heterotrophs (the ”None” group), which does not contain either xanthorhodopsin nor light harvesting system, the approximate average number is 2.9 photosensors. (B) The bar plot shows an average number of each type of photosensor in Dual, AAP, Rho and None groups. The studied Rho-based systems do not have any LOV domains. J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 5 sis on the spread of PYP was reported suggesting multiple functions for PYP protein, depending on the species. 24 It has also been suggested that PYP could act as a UV-sensor. 25 Rhodopsins as photosensors are not analyzed in this study. Yet, it is worth mentioning that in the case of J5HS3a (S. faeni), both xanthorhodopsin and sensory rhodopsin genes were detected, while coexistence of xanthorhodopsin and sensory rhodopsin genes could not be observed in other instances. Comparison of the photosensory composition of the species in Table 1 identified a wider set of photosensory genes in the phototrophs, making it appealing to study the genomic sequences in more detail. In the Figs. 3–5 we provide the phylogenetic trees for CRY/PHR, LOV, BLUF, and BphP genes, respectively, of the strains studied, with the different categories highlighted. At first glance, one can see that the photosensor genes from each strain only partially cluster according to their taxonomy or to their photoheterotrophicity in the phylogenetic tree. When multiple photosensors are observed from the same strain, they often locate in different main branches. Still, across all the photosensors, the overall grouping remains according their genus. For example Sphingomonas always appear in the nearby branches (Figs. 3–5). Cryptochromes and Photolyases Cryptochromes and Photolyases marked in Table 1 were collected with a threshold of 35% sequence identity with Dinoroseobacter shibae (strain DFL 12) CRY/PHR sequence. The conservation of the FAD-binding domain’s strong homology across all the strains we examined underscores the notion that cryptochromes and 6_4 PHRs, despite their diversity in bacterial hosts, may share a common functional heritage. This conservation hints at the importance of these photoreceptor proteins in the adaptation of bacteria to UV radiation-induced DNA damage ? . The consistent presence of this homologous domain not only highlights the role played by cryptochromes and 6_4 PHRs in light-induced DNA repair but also suggests a shared evolutionary lineage among these proteins. 26,27 This shared ancestry, as evident from the robust FAD-binding domain homology, invites further exploration into the precise mechanisms governing the operation of these photoreceptors and their contribution to the adaptability of dual phototrophic bacteria. Additionally, the coexistence of FAD/NADbased photosensors, such as those observed in AP64 6_4 PHR 1 and TET16 6_4 PHR, broadens the scope of photoreceptor diversity and their potential functional significance in bacterial UV radiation response. Furthermore, while all cryptochromes are categorized within the PhrB and DPRP (Deoxyribodipyrimidine photo-lyase-related protein) family, the inclusion of an N-terminal DNA photolyase domain within all 6_4 PHRs is another finding. This feature suggests a dual role for 6_4 PHRs, not only in light sensing but also in direct involvement in DNA repair processes. 28,29 The presence of such a domain provides a foundation for future research to unravel the specific mechanisms by which these photoreceptor proteins contribute to bacterial survival and adaptation in the face of UV radiation-induced DNA damage. In summary, the shared homology in the FAD-binding domain across diverse bacterial strains points to a unifying theme in the functionality of cryptochromes and 6_4 PHRs, encouraging a deeper exploration of their roles in the context of bacterial phototrophy and DNA repair mechanisms. LOV domains - blue light sensing The LOV domains, highlighted in Table 1,were meticulously curated using a stringent sequence identity threshold of 35% relative to the LOV domain of Dinoroseobacter shibae (strain DFL 12). These LOV domains exhibit a unique structural architecture comprising at least one of the modules from the PAS9-PAS3-GAFHATPase-HWE_HK family. These functional modules include putative active sites, heme pockets, ATP binding sites, Mg binding sites, ATP-lid regions, and the signature G-X-G motif. This multifaceted architecture underscores the importance of these LOV domains as versatile light-sensing elements, potentially orchestrating a diverse range of biological responses in the organisms housing them. 30 BLUF domains - blue light sensing The BLUF domains marked in Table 1,were collected in accordance with the same 35% sequence identity threshold in comparison to the BLUF domain of Dinoroseobacter shibae (strain DFL 12), were consistently present in all the dual phototrophic strains examined. The conservation of this domain across the aligned strains suggests that these BLUF proteins belong to the group II BLUF proteins, denoting a shared ancestry and, potentially, a common functionality. 31 These BLUF domains also exhibit a distinctive set of conserved amino acids, specifically tyrosine, glutamine, and methionine, as depicted in Figure S8-10. Notably, a majority of the analyzed BLUF proteins are relatively short, comprising fewer than 200 amino acids. The absence of effector domains in these proteins implies that their signaling capabilities rely on changes in their oligomeric state, reflecting an elegant and efficient mechanism for translating light stimuli into biological responses. 31 However, it is worth highlighting that a few exceptions, namely NX02 BLUF 1, TET16 BLUF 1, C1 BLUF 1, J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 6 CBMB20 BLUF 1, and CBMB27 BLUF 1 domains, feature an additional C-terminal domain. According to protein family classification analysis provided by NCBI, these extra domains in NX02 BLUF 1 and TET16 BLUF 1 show homology to B12-binding domains. This suggests the possibility of an additional photosensor, such as CarH, which often plays a role in controlling carotenoid biosynthesis in various organisms. 32 Hence, it is conceivable that this dual phototrophic strain set may possess an extra layer of photoreceptor complexity, enabling them to fine-tune their responses to light in the context of carotenoid biosynthesis regulation. Furthermore, the NX02 strains exhibit notable variations in their N-terminal regions when compared to each other. This divergence in their N-terminal regions hints at potential adaptations or specialized functions within this particular group of organisms, and it warrants further investigation to unveil the intricacies of their light-sensing capabilities and how these variations may impact their overall responses to environmental stimuli. Phytochromes - red light sensing The phytochromes marked in Table 1 were collected with a threshold of 35% sequence identity with Sphingomonas faeni S3H21 phytochrome sequence. We restricted for an analysis where all bacteriophytochromes contained the conserved DIP-motive in the chromophore binding pocket. 33 Phytochromes are red – far-red light sensors. They have a bilin chromophore that undergoes cis–trans photoisomerization after absorbing light in the red or far-red region. Notably, the absorption spectra of the red absorbing state (Pr-state) and the far-red absorbing state (Pfrstate) vary exactly at the absorption wavelength of Chl amolecules with a very steep difference. 20 In canonical BphPs, the Pr-state is the resting state in darkness, and light activation drives the protein to the Pfr-state. Bathy type of BphPs, however, function conversely: in darkness the system is in Pfr state and far-red light activation drives the system to the Pr-state. A structure related genome Fig. 3. Phylogenetic tree of CRY/PHR domains The phylogenetic tree is based on an alignment process, and the maximum likelihood trees show sequences with 85% or more sequence similarity in the same branch. The color coding reflects the genus of the strains (See Table 1) and the dual, AAP and Rho classes are marked with shades of grey. Branch length.s are not shown. J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 7 comparison study by Velaquez-Escobar et al. suggested that RxxPRxSF motif in the so-called tongue region 34 and HExT motif indicates a histidine kinase activity of protein. 16 Following this guideline, the Figs. 5 and S11-15 and indicate that about twothirds of dual phototrophic systems’ BphPs may function as bathy BphPs. Observation of the phylogenetic tree indicates the lower half is completely bathy BphPs except for PAMC26645 BphP 2 and S2H28 BphP 2. Bathy BphPs can also be found on the second and sixth rows of the first branch. We note however, that the final decision of whether the particular strain is bathy or canonical needs to be characterized spectroscopically with isolated proteins, see for example in. 35 The HK proteins are typically divided into five subtypes (families), HisKA, HisKA_2, HWE_HK, HisKA_3, His_Kinase. 36 Typically BphPs are HisKA sensors. 36 The first branch (TA154 BphP 1 – CBMB27 BphP 2) are HisKA-family. The second branch (TA154 BphP 3 – PAMC26645 BphP 1) are HWE_HK family together with HATPase family. The third branch (TA278 BphP 2-SE61 BphP 1) are HWE_HK family. In the case of PAS-output domain containing strains the homologous superfamily comparison gives the PAS domains to be homologous with the PYP sensor domain, as observed with previous studies of BphP systems. 37,24 A tandem PYP-BphP construct has been linked to photosystem regulation, 38 and the PYP has been shown to accelerate the Pr-recovery once illuminated with blue light. 39 However, only MS446 BphP 1 includes the PYP-like domain inside of its PAS-domain. The function in these dual phototrops requires further investigation. Another detail about domain construction of the BphP sequences is that the AP4 BphP 2 and TA352 BphP 3 are actually PAS-less BphPs. All these observations brings attention to the rich variety of BphPs types in phototrophs. Fig. 4. Phylogenetic tree of LOV and BLUF domains The phylogenetic trees of LOV (A) and BLUF (B) are based on an alignment process, and the maximum likelihood trees show sequences with 85% or more sequence similarity in the same branch. The color coding reflects the genus of the strains (See Table 1) and the dual, AAP and Rho classes are marked with shades of grey. Branch lengths are not shown. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) J.A. Ihalainen, B. Dogan, M. Kurttila, et al. Journal of Molecular Biology 436 (2024) 168412 8