Cytochrome b5 occurrence in giant and other viruses belonging to the phylum Nucleocytoviricota

Introduction

Giant viruses, discovered just over 20 years ago, are enigmatic members of the global microbiome¹. Traditionally, viruses were defined as biological entities on the edge of life, lacking cellular structure and metabolism, and only reproducing in a host cell. That view of viruses changed in 2003 with the discovery of a giant virus, Mimivirus bradfordmassiliense (formerly known as Acanthamoeba polyphaga mimivirus or APMV)^2,3. (We refer to Mimivirus bradfordmassiliense as mimivirus.) The mimivirus genome epitomized an emerging and diverse group of viruses, then referred to as the nucleocytoplasmic large DNA viruses (NCLDV). Giant viruses, many larger than 0.2 μm and with genomes greater than 300 kbp, continue to be discovered in—or captured with—amoebae⁴, and have been identified in a wide range of hosts, including algae⁵. They are increasingly found in metagenomic datasets as well^6,7. With the growing number of giant viruses, a new classification was developed including a new class, the Megaviricetes within a new phylum, the Nucleocytoviricota³. Especially prominent to-date are members of the Mimiviridae, with AT-rich DNA genomes, and the proposed Pandoraviridae family, which have GC-rich DNA genomes. Multiple clades and species occur in both families. They have been isolated from marine, freshwater, and/or terrestrial sources in different parts of the world⁸.

The first giant virus discovered, Mimivirus bradfordmassiliense, encodes more than 1000 proteins, but fewer than 300 have predicted functions^2,9. Many mimivirus sequences encode proteins not found in any other virus, including protein translation enzymes (thought to be a signature of cellular organisms), DNA repair pathway components and putative homologues of enzymes involved in eukaryotic metabolism². Since publication of the first mimivirus genome, numerous other metabolic genes thought only to be found in cellular lifeforms have been found in diverse GV genomes from different environments. Using bioinformatic approaches, we and others have identified GV genes and proteins and their orthologues that are involved in known cellular primary metabolic pathways e.g. glycolysis, gluconeogenesis, TCA cycle, photosynthesis and lipid β-oxidation^10,11,12. It is already accepted that viruses can hijack host metabolic networks, but the GVs auxiliary metabolic genes represent another form of host metabolism manipulation, by expanding, rather than merely enhancing, host cells’ catalytic capabilities especially in harsh variable environments. However, the functionality of these genes remains unclear, and how they integrate into established host metabolic systems is completely unknown¹³.

Heme proteins are among the most versatile of metalloproteins found in nature and play essential roles in diverse and distinct biological functions from gas exchange to redox reductions¹⁴. Cytochrome b5, originally detected in the larvae of the silkworm Platysumia cecropia by Sanborn and Williams in 1950¹⁵, is a ubiquitous electron transport protein found in animals, plants, fungi and purple photosynthetic bacteria¹⁶. Soluble forms of cytochrome b5 function in photosynthetic energy production in bacteria and the reduction of hemoglobin in animals^16,17. In contrast, membrane-bound forms of cytochrome b5 function in the mitochondria and the endoplasmic reticulum (ER) of eukaryotes. ER-associated cytochrome b5 roles are shown in Fig. 1 and include roles in fatty acid biosynthesis driving fatty acid desaturase enzymes (Δ9-, Δ6-, Δ5- and Δ12-desaturases), in cholesterol biosynthesis, and in the reduction of cytochromes P450 enzymes where cytochrome b5 has been shown to alter substrate metabolism by different P450s¹⁶. No function has been attributed to cytochrome b5 in the mitochondria¹⁸.

Cytochrome b5 occurrence in giant and other viruses belonging to the phylum Nucleocytoviricota — **Fig. 1: Prominent viruses and cytochrome b5 roles.**

Previously a gene encoding a putative cytochrome b5 protein was found in the large double-stranded DNA virus Ostreococcus tauri virus OtV-2¹⁹, and the protein was expressed and characterized. This soluble viral cytochrome b5 protein had near identical spectral properties to purified membrane-bound, recombinant human cytochrome b5¹⁹. The crystal structure of the O. tauri OtV-2 cytochrome b5 revealed a single domain, comprising four β sheets, four α helices and a heme moiety, which is similar to that found in larger eukaryotic cytochrome b5 proteins¹⁹. The O. tauri OtV-2 cytochrome b5 also was shown to reduce a eukaryotic cytochrome P450 (Candida albicans CYP51) through the transfer of reducing equivalents, suggesting this viral cytochrome b5 can act as a bona fide cytochrome b5 in P450 reactions, as well as potentially be involved in modulating lipid and sterol biosynthetic enzyme activity. This added credence to our current hypothesis that novel viral heme proteins, including viral cytochrome P450 enzymes²⁰, can modulate either viral or host sterol or lipid content to maximize energy production and facilitate viral replication. This possibility has particular relevance to viruses that have lipid envelopes and/or rely on controlled passage through the membrane for virion release.

The current work describes the discovery of multiple cytochrome b5 enzymes in numerous giant virus genomes and from metagenome datasets that are currently publicly available. Identification of enzyme systems that seemingly cannot function in the inert virion particle yet are identifiable across a wide range of viral species and ecotypes, points to key aspects of NCLDV biology that remain unexplored.

Results

Overall distribution of cytochrome b5 in the virosphere

Among the viruses, we identified numerous cytochrome b5 genes encoded in the genomes of specific groups within the Nucleocytoviricota including Ostreococcus viruses, which are atypical giant viruses, based on the absence of RNAPol2. Genes were assigned as encoding cytochrome b5 based on the conservation of sequence and secondary structure identity with known eukaryotic cytochromes b5. Key factors in sequence assignment include the retention of secondary structure elements occurring in the order β1-α1-β4-β3-α2-α3-β5-α4-α5-β2-α6, the absolute conservation of two histidine residues residing in the loops of between helices α2 and α3 and between helices α4 and α5, and the overall domain identification using Pfam models (Fig. 2). The conservation of the two α2-α3 and α4-α5 histidine residues is key, as these conserved histidine imidazoyl side chains bind the heme cofactor in authentic cytochrome b5 proteins.

**Fig. 2: Amino acid sequence alignment of viral cytochromes b5.**

A phylogeny of the Nucleocytoviricota with the branches that contain cytochrome b5 genes highlighted, appears in Fig. 3A, and details of the viral cytochromes b5 are presented in Table 1. Thus far, our searches revealed cytochrome b5 genes to be present in the Mimiviridae genera Mimivirus, Moumouvirus and Megavirus (formerly known as Groups A, B and C respectively) and in tupanviruses (Tupanvirus altamarinense and Tupanvirus salinum). Cytochrome b5 genes were also detected in giant virus genomes assembled during the analysis of metagenomic data from terrestrial soils (satyrvirus, harvfovirus, hyperionvirus and terrestrivirus) and from marine sediments (marseillevirus and pithovirus)²¹. In addition, cytochrome b5 genes were retrieved from the genomes of NCLDV that infect the green algae Ostreococcus lucimarinus and O. tauri (the Phycodnavirus OtV-2 mentioned above)²².

**Fig. 3: Phylogenetic analyses of *Nucleocytoviricota* cytochrome b5 proteins.**

Table 1 Cytochrome b5 features

Full size table

In bacteria and eukaryotes there are multi-domain proteins that include a cytochrome b5 domain. A phylogenetic tree of protein families that include in their structure a cytochrome b5-domain-containing motif is presented in supplemental information (Supplemental Fig. 1), showing where viral cytochrome b5 proteins fit among the multi-domain proteins. Although some metagenomes have genes coding for multidomain proteins that include cytochrome b5 domains, the cytochrome b5 proteins described here are single domain proteins. Figure 3B shows a phylogenetic tree of cytochrome b5 proteins.

Mimiviridae cytochrome b5 genes

Searching the first giant virus genome sequence², Mimivirus bradfordmassiliense (Fig. 1A), revealed the presence of a gene (MIMI_L628) coding for a 124 amino acid protein containing the canonical eukaryotic cytochrome b5 heme domain (located between residues 49–124). Following this initial observation, we discovered putative cytochrome b5 proteins encoded in giant viruses in all three Mimiviridae genera, Mimivirus, Moumouvirus, and Megavirus (Table 1). Analysis of the viral cytochrome b5 sequences revealed the putative proteins to be either soluble in nature, existing as a single hydrophilic domain protein, or as presumed membrane bound proteins possessing a single N-terminal transmembrane anchor. This was true both within and between Mimiviridae sublineages. For example, in mimiviruses, the M. bradfordmassiliense cytochrome b5 (accession number Q5UR80) consists of a hydrophilic heme-bound globular domain with a predicted 16–18 amino acid N-terminal transmembrane anchor, while the hirudovirus cytochrome b5 gene encodes a smaller b5 domain without an encoded recognizable membrane anchor (Table 1; Fig. 2). Similar examples of membrane bound and soluble cytochromes b5 are observed in genera Megavirus, and Moumouvirus (Table 1; Fig. 3). The occurrence of an N-terminal membrane anchor is unique to viral cytochrome b5 proteins alone; all eukaryotic cytochrome b5 proteins examined to date have a C-terminal membrane anchor, e.g. human and viral amoeba host A. castellanii cytochromes b5 (Fig. 2).

There is a strikingly high degree of sequence identity (98–100%) between inferred cytochrome b5 protein homologues in the mimiviruses from different parts of the world. Similarly, there is >95% sequence identity among the cytochrome b5 proteins in Megavirus species, and among the cytochrome b5 proteins in Moumouvirus species, from different parts of the world. Between groups there is lower conserved sequence identity, with the mimiviruses cytochrome b5s having <60% identity with megavirus or moumouvirus cytochrome b5s. Moumouvirus and megavirus cytochrome b5s share a somewhat higher % identity ( ~ 75%). All of the Mimiviridae cytochrome b5 genes are AT-rich (69–75%), like the Mimiviridae genomes overall, and nucleotide sequences also are highly similar within subgroups.

To further understand how the Mimiviridae cytochrome b5 proteins utilize their N-terminal transmembrane anchor to interact and colocalize with cellular organelles, we performed a protein colocalization prediction. The Mimivirus bradfordmassiliense cytochrome b5 (Q5UR80) is predicted with high likelihood to colocalize with the ER through its N-terminal sequence (Fig. 4). Thus, in the case of mimivirus cytochrome b5 (Q5UR80) the likelihood for the endoplasmic reticulum was 0.8316, indicating an 83% probability of localization, while the likelihoods for other compartments were lower. (See Supplemental Table 2).

**Fig. 4: Viral cytochrome b5 subcellular localization predictions.**

However, this was not the case for cytochrome b5s in all the Megaviricetes. Some of these cytochrome b5s, such those as in hirudovirus, Oyster virus, and Samba virus, were predicted to be in the cytoplasm, without association with any specific cellular organelle (Supplemental Table 2). This discrepancy could arise from the absence of a discernible membrane anchor in these proteins. Similar observations are noted for viruses in the Moumouvirus or Megavirus genera (Supplemental Table 2). This finding suggests that viral cytochrome b5 proteins strategically utilize their N-terminal membrane anchor to associate with the ER, potentially implicating these proteins in critical ER-related processes such as lipid and protein synthesis, which are vital for viral replication and assembly.

Tupanvirus cytochromes b5 genes

The tupanviruses (Fig. 1B), related to Mimiviridae, were discovered in a soda lake in Brazil (Tupanvirus salinum) and in the deep ocean (3,000-m depth) (Tupanvirus altamarinense) and are among the first viruses reported to possess genes for amino-acyl tRNA synthetases for all 20 standard amino acids²³. Both tupanviruses possess genes predicted to encode cytochrome b5 proteins, each containing a N-terminal transmembrane anchor (Supplemental Table 2; Fig. 4). While the tupanvirus cytochromes b5 share 78% percent amino acid identity, they share between 55–60% identity with cytochrome b5 proteins in the mimiviruses and megaviruses but only 45% identity with moumouvirus cytochrome b5 proteins. Subcellular colocalization predictions for both tupanviruses’ cytochrome b5 proteins showed a strong likelihood of association with the ER, possibly due to an ER-targeted sequence within their N-terminal region (Fig. 4). It is important to note, however, that these results represent predictions generated by algorithms and require experimental validation for confirmation. This observation nevertheless suggests that as with mimivirus, the tupanvirus cytochrome b5 proteins are likely to colocalize with the ER, suggesting its involvement in modulating ER function and potentially exerting influence over various metabolic pathways to enhance viral fitness.

Subcellular localization predictions for the cytochrome b5 proteins of O. lucimarinus virus, marseillevirus, and satyrvirus revealed a different pattern. These proteins did not show significant colocalization with the ER. Instead, our prediction results indicated their localization in the cytoplasm, with a weak likelihood of association with specific organelles (Supplemental Table 1). This discrepancy suggests that the subcellular localization of cytochrome b5 proteins in infected hosts vary among different viruses. One possible explanation for this difference could be the absence of a membrane anchor on these proteins, as indicated by our alignment result (Fig. 2). The lack of a membrane anchor could be expected to prevent their association with the ER and other organelles, leading to their predominant localization in the cytoplasm. These findings underscore the complexity of this protein localization, and the diverse strategies employed by viruses to interact with cellular machinery.

Gene synteny, transcriptome analysis and cytochrome b5 spectral analysis

Viruses in the Mimiviridae genera Mimivirus, Moumouvirus, and Megavirus exhibit high degrees of shared gene order (synteny) extending over 8 genes around the cytochrome b5 gene (Fig. 5). There is less shared synteny of these with tupanviruses (Fig. 5), and no evident sharing of synteny between the Mimiviridae and the Ostreococcus viruses (Supplemental Fig. 2). It is noteworthy that there is a prevalence of Fe(II)-2-oxoglutarate dependent dioxygenases among the genes flanking cytochrome b5 in the Ostreococcus viruses.

**Fig. 5: Synteny analysis of cytochrome b5 loci in *Mimiviridae* and tupanviruses.**

Transcriptomic analysis also revealed that, of the 90% of all mimivirus genes transcribed during infection, the mimivirus gene L628 (which putatively encodes cytochrome b5) is maximally transcribed at 6 h (Fig. 6A)²⁴. This expression pattern falls into the ‘late’ class of expression upon mimivirus infection, as do other genes encoding oxidoreductase enzymes, or structural components of the mimivirus particles²⁴. Heterologous expression of histidine-tagged mimivirus cytochrome b5 (MIMI_L628) in E. coli and purification using nickel affinity chromatography yielded a distinctive hemoprotein. Cell fractionation revealed MIMI L628 to be a membrane bound protein located in the membrane fraction following ultracentrifugation at 100 000 × g and solubilised by the use of detergents (1% (v/v) Tween 20 and 1.5% (w/v) sodium cholate). The absorbance spectra of oxidized recombinant MIMI L628 protein (Fig. 6B) shows maxima at 412, 528, and 567 nm. These spectral characteristics are typical of purified cytochrome b5s from eukaryotic organisms described in the literature (18).

**Fig. 6: *Mimivirus bradfordmassiliense* cytochrome b5 transcriptomic profile and protein spectrum characterization.**

Analysis of cytochrome b5 secondary structure and overall topology

All eukaryotic ER-bound microsomal cytochrome b5 proteins examined to-date are anchored to the ER lipid bilayer by a C-terminal transmembrane anchor. Mitochondrial cytochrome b5 proteins are also anchored in the membrane by a C-terminal transmembrane anchor. In the case of human and A. castellanii microsomal cytochrome b5s, each possesses a single 16–18 amino acid C-terminal region that anchors the protein to the ER membrane (Fig. 2 and Table 1). In contrast, analysis of the viral cytochrome b5 sequences reveals the putative proteins to be either soluble, existing as a single hydrophilic domain protein, or membrane bound apparently through a putative N-terminal transmembrane anchor sequence rather than a C-terminal anchor. Protein modelling demonstrates the high degree of tertiary structural homology between the crystalized house fly and Ostreococcus cytochrome b5 and the modelled mimivirus, megavirus, tupanvirus, and faunusvirus cytochrome b5s (Fig. 7). Notably, the heme-binding histidines are spatially conserved (Fig. 8), but there is some variability in the heme-adjacent residues: conserved Ostreococcus Phe44 is a tyrosine in faunusvirus; and Phe67 is a conserved methionine in mimiviruses, megaviruses, tupanviruses, and moumouviruses, but a phenylalanine in soil faunusvirus, hyperionvirus, and terrestrivirus cytochrome b5s.

**Fig. 7: Crystal structure of housefly cytochrome b5 compared with homology models of virus cytochrome b5s.**

**Fig. 8: Overlaid viral cytochrome b5 structures showing the positioning of the conserved histidine residues (red) binding the heme.**

Discussion

Our search of more than ten thousand completely sequenced and annotated viral genomes revealed cytochrome b5 genes in several virus taxa, including the AT-rich genomes in the Mimiviridae, tupanviruses, and Marseilleviridae, members of the class Megaviricetes, large double stranded DNA viruses. Additionally, we found cytochrome b5 genes in viruses infecting O. lucimarinus, similar to the virus infecting O. tauri (Table 1). The Ostreococcus viruses are also members of the Nucleocytoviricota, albeit much smaller than the giant mimiviruses, and lacking RNAPol2 genes. We assigned genes as encoding cytochrome b5 based on the presence of the conserved heme-binding consensus motif found in all cytochrome b5s described to date¹⁶. The Mimiviridae cytochrome b5 genes are not homologous with the Ostreococcus virus genes and appear to have been acquired separately, potentially an example of convergent evolution. While genome size is a clear and defining feature for giant viruses, it is interesting to note that OtV2 infects one of the world’s smallest free-living eukaryotes, suggesting an important metabolic role(s) for cytochrome b5 in overcoming the extremes of scale in both the virus and host.

The earlier expression of a recombinant cytochrome b5 from the OtV2 virus, and analysis of activity, showed that at least this virus cytochrome b5 gene encodes a functional protein¹⁹. Whether cytochrome b5 genes in the other Ostreococcus phage and the giant viruses also encode functional proteins is not yet known, although based on structural and transcriptional features the virus cytochrome b5 proteins are almost certainly functional. Larger questions regarding functional role(s) these enzymes play in the viruses, or hosts, and the origin of the virus cytochrome b5 genes are unanswered.

Increasingly, metabolic enzymes encoded by genes in the giant viruses have been found to be functional in vitro and to be expressed in vivo upon infection of a host (10,11). These include genes encoding enzymes involved in glycolysis, gluconeogenesis, tricarboxylic acid cycle, photosynthesis, and β-oxidation. One of the important functions of cytochrome b5 is providing reducing equivalents during the catalytic cycle of cytochrome P450 proteins. This could be considered as a possible function in the Mimiviridae and the tupanviruses, where there are multiple P450 genes, and those genes are also expressed in the infected host, shown at least for mimivirus¹⁰. However, the Marseilleviridae, which do have cytochrome b5 genes, do not appear to have genes that encode any cytochrome P450. Moreover, our search revealed cytochrome b5 genes only in the AT-rich Mimiviridae, Marseilleviridae and the tupanviruses but not in the other branches of the Megaviricetes including in the GC-rich pandoraviruses, mollivirus, kaumoebavirus or the orpheovirus, all of which contain cytochrome P450 genes. This would suggest that if cytochrome b5 does support virus P450 function, that so far it would seem to be restricted to AT-rich virus clades.

In addition to its role in in providing electrons to support the mono-oxygenation by cytochrome P450, cytochrome b5 plays pivotal roles in cellular metabolism, particularly in fatty acid elongation, desaturation, and cholesterol biosynthesis²⁵. These metabolic pathways are essential for maintaining cellular homeostasis and protecting against metabolic disorders²⁶. In the context of viral infection, the presence of cytochrome b5 in giant viruses like mimiviruses and tupaniviruses suggests its potential involvement in similar metabolic processes. Our in silico predictions of subcellular colocalization indicate cytochrome b5’s likely localization within the endoplasmic reticulum (ER), consistent with its predicted transmembrane domain, and suggests that cytochrome b5 expression in the ER may be related to the ER involvement in lipid metabolism. The ER serves as a hub for lipid metabolism, housing enzymes crucial for triglyceride and cholesterol biosynthesis as well as ergosterol biosynthesis in amoebae²⁷. During the replication cycles of viruses like the mimiviruses, there is a notable close association between membranes derived from the endoplasmic reticulum (ER) and sites where viral assembly occurs²⁸. This suggests that the ER is effectively “hijacked” for lipid synthesis, a process crucial for the formation of virus particle envelopes as shown in some DNA viruses²⁹. Importantly, no viral cytochrome b5 reductase has yet been discovered indicating that the viral cytochrome b5s described herein would function in tandem with host cytochrome 5 reductase. Viral cytochrome b5 may serve a function within the ER, participating in the desaturation and elongation of fatty acids. Such a role would ensure the availability of an adequate lipid supply, which is essential for the overall fitness and successful replication of the virus. Moreover, in this case, we speculate that cytochrome b5, by transferring electrons to host enzymes, may facilitate the synthesis of polyunsaturated fatty acids by transferring electrons to both host and virally encoded enzymes involved in lipid metabolism. We note a slight caution regarding the localization studies, however, and suggest that experiments are needed to validate the bioinformatic results, for example, fluorescence microscopy to validate the colocalization studies.

Another potential role of cytochrome b5 in giant viruses is as an antioxidant. Viral replication can stress the host cell, leading to reactive oxygen species (ROS) accumulation³⁰ Studies have shown that overexpression of cytochrome b5 in mammalian cells reduces oxidative stress³¹ implying that cytochrome b5 in viruses might similarly safeguard the viral replication apparatus from ROS-induced damage. Nevertheless, the roles of cytochrome b5 in giant viruses likely span beyond a single function. Given its multifaceted nature, comprehensive molecular and biochemical studies are needed to decode specific contributions of cytochrome b5 within giant viruses, underscoring the need for further investigative work to clarify its exact function and role.

In a study of another gene family, the cytochrome P450s, genetically unrelated to the cytochrome b5s, we found it impossible to conclude that the P450 genes in the giant viruses were acquired from any known host²⁰. However, some other viruses have P450s that appear to have been acquired from a known eukaryote host. Thus, hokovirus CYP5724A1 and Ranid herpesvirus 3 CYP5723A1 share approximately 30% sequence identity to lipid and steroid metabolizing P450s present in their hosts, and the P450 102L1 we reported in the mycophage Mycobacterium phage Adler was acquired from the Mycobacterium abscessus host²⁰. A similar picture appears with the cytochrome b5 genes. The cytochrome b5s found in the Ostreococcus phages do bear significant similarities to the cytochrome b5s in their algal hosts, while the cytochrome b5 genes in other viruses in the Nucleocytoviricota do not appear to have been derived from any known hosts. It would be informative to determine whether there is similarity to cytochrome b5 in any new hosts identified. Barring that, the giant virus cytochrome b5 genes do not appear to have originated through gene transfer from a contemporary host, but rather, as appears to be the case for some other Megaviricetes genes, the cytochrome b5 genes may have been transferred from ancient hosts no longer extant.

The major distinction in the presence of cytochrome P450 genes and cytochrome b5 genes among the giant viruses either indicates that these genes were acquired separately, or that if they were acquired together, that the cytochrome b5s were lost from the GC-rich clades. Alternatively, if the cytochrome b5 genes are de novo creations of the giant viruses, then that process would have been restricted, or the genes may have been lost along the way from the GC-rich clades. For both cytochrome b5 and cytochrome P450 there is an absolute requirement for heme cofactor, intimating that both viral gene families were transferred from a cellular host as no complete heme biosynthetic enzyme machinery has been found to be encoded in a giant virus genome to-date. Alternatively, viruses in the Megaviricetes may originally have possessed heme biosynthetic machinery but then lost it as it became possible to dependent on a host pathway for heme supply. In either case, the presence of genes for cytochromes b5 in the giant viruses adds to the growing number of genes in the giant viruses that are not expected in any virus, informing the discussion of the origin of the giant viruses and the genes they contain. In particular this study raises adds to questions about the origin of genes for heme proteins and their role in the giant viruses.

Methods

Viral cytochrome b5 sequence datasets

The cytochrome b5 domain is readily identifiable bioinformatically. Candidate viral cytochrome b5 sequences were retrieved from the NCBI sequence databases (online version as of December 15, 2024). Sequences homologous to mimivirus cytochrome b5 were identified in the environmental metagenomic database (env_nr) and the non-redundant database (nr).

The Genbank accession numbers for viruses here are given in Supplemental Table 1). Inferred protein sequences were aligned to the PfAM b5-domain hidden Markov model (PF00173) using ClustalOmega³², and scored with T-Coffee³³. Alignments were visualized in BioEdit³⁴. Virus names are those in the original reports or described in Genbank. Pairwise amino acid identities reported in the text are based on the length of the shortest protein in the alignment. Identity values used in the tables are based on multiple sequence alignments and are usually not more than 1% different from stated values.

Phylogenetic analyses of viral cytochrome b5

Phylogenetic trees of cytochrome b5 were inferred using maximum likelihood (ML) approaches (RAxML-ng v1.2.2) with the VT + I + G4 model of amino acid substitution as selected by ModelTest³⁵. Twenty replicate ML searches were conducted, and the best-scoring tree was subjected to rapid bootstrapping in RAxML to assess bifurcation support using the autoMRE bootstopping criteria, which produced 1000 bootstrap replicates. Virus phylogenies were adapted from the concatenated seven conserved gene set identified by Aylward et al (2021) by extensively pruning the published phylogeny²⁰. Figures were constructed using iToL³⁶.

Modelling and structural analysis

Homology modelling of viral b5 proteins was performed using Modeller (v9.16)³⁷. 100 independent models were produced based on rat, housefly, and Ostreococcus virus cytochrome b5 structures (PDB codes 4B8N,1EUE, 3MUS, 2IBJ). The model with the highest DOPE-HR (high resolution discrete optimized potential energy) score was selected and visualized in Pymol. Further structural analysis was performed using Thesesus³⁸.

Subcellular colocalization of viral cytochrome b5 proteins was predicted using the DeepLoc 1.0 and DeepLoc 2.0 algorithms^39,40 These algorithms differ slightly. By using both versions, we ensured that any observed differences were not due to algorithmic changes but instead gave meaningful biological insights. The significance of the predictions was determined based on the likelihood values provided by the DeepLoc algorithms. These values range from 0 to 1, where 0 indicates no likelihood of colocalization and 1 indicates absolute certainty of colocalization. Higher likelihood values suggest a stronger probability of colocalization with a given compartment.

Mimivirus cytochrome b5 transcriptomic analysis and protein characterisation

The mimivirus cytochrome b5 gene (MIMI_L628) was synthesised by Eurofins MWG Operon and gene integrity confirmed by DNA sequencing. A unique NdeI restriction site was generated prior to the ATG start codon and a HindIII site following the TGA stop codon to facilitate cloning into the Escherichia coli expression plasmid, pET17b. Additionally, a polyhistidine tag, to allow purification of the expressed protein by Ni²⁺-NTA affinity chromatography, was engineered into the C-terminus of the protein. MIMI L628 was expressed and purified using similar conditions described for Ostreococcus virus cytochrome b5¹⁹. Cytochrome b5 content was quantified from the reduced versus oxidized difference spectrum of purified protein using a differential absorption coefficient Δ_{424 – 409} of 185 mM⁻¹ cm⁻¹⁴¹.