The Bacteriophage VMY 22 Has Enhanced the Stability of Its Functional Proteins via Adaptive Evolution in a Temperature-Varying Environment

TL;DR

This study shows how a glacier bacteriophage adapts to temperature changes by evolving its functional proteins, improving stability in varying environments.

Contribution

The paper reveals a novel mechanism of temperature-driven functional protein evolution in cold-adapted bacteriophages.

Findings

Phage VMY22 showed temperature-dependent infectivity changes and genomic mutations in functional genes.

Protein structural adaptations were observed in response to different environmental temperatures.

Functional gene mutations suggest evolution primarily affects post-adsorption processes in phage.

Abstract

Temperature fluctuations strongly affect microbial viability, often inducing adaptive responses. In this study, we employed the psychrophilic bacterium Bacillus mycoides 41-22 and its associated phage VMY22, originally isolated from the Mingyong Glacier, to investigate phage adaptability under varied temperature conditions. Through selective enrichment at 4 °C, 15 °C, 28 °C, and 32 °C, we observed clear differences in phage infectivity, as assessed by plaque assays, along with genomic mutations and protein structural changes. Notably, mutations predominantly occurred in functional genes (ATPase, endolysin), while the examined structural loci remained conserved. Homology modeling revealed distinct adaptations in protein tertiary structures corresponding to environmental temperatures, suggesting that phage evolution mainly affects post-adsorption processes. Our findings elucidate a novel mechanism of temperature-driven functional protein evolution among cold-adapted bacteriophages (phage) and providing insights into their potential applications in microbial ecology and biotechnology.

Figures

Full text

1. Introduction

Microorganisms are the most abundant and widely distributed life forms on Earth, including bacteria, fungi, viruses, and certain protozoa. Despite their microscopic size, they play indispensable roles in global energy transfer and nutrient cycling, forming the foundation of the biosphere’s ecological networks [1,2,3]. Among the many environmental factors that influence microbial viability, temperature is particularly critical. It affects cellular metabolism, gene expression, and overall survival, compelling microbes to develop adaptive strategies in response to thermal fluctuations (e.g., diurnal or seasonal changes) and long-term environmental gradients (e.g., geographic or climatic differences) [4].

Scientific interest in temperature’s influence on microbial biology dates back to the 1970s, when Ashburner and colleagues reported that shifts in temperature can significantly impact bacterial transcriptional activity [5]. Since then, many studies have explored microbial responses at the transcriptional and proteomic levels, uncovering temperature-sensitive regulatory proteins and stress-response pathways that enable survival across diverse thermal conditions.

Viruses, as acellular entities composed solely of nucleic acids and proteins, present unique challenges in thermal adaptation due to their structural simplicity [6]. While the study of viruses began in the late 19th century, research into their temperature responsiveness is still developing. To date, most investigations have centered on thermophilic or mesophilic viruses, such as single-stranded RNA viruses, with work by Hossain and Jeffrey shedding light on viral evolution and adsorption kinetics at elevated temperatures [7,8]. However, cold-active (psychrophilic or psychrotolerant) viruses remain largely understudied, limiting our understanding of their adaptive mechanisms and ecological roles.

Despite progress in understanding microbial and viral responses to thermal stress, little is known about how cold-active phages adapt functionally and structurally to fluctuating temperatures. To address this gap, we employed Bacillus mycoides 41-22 and its lytic phage VMY22, originally isolated from the Mingyong Glacier, as a model system [9]. This study aimed to investigate how temperature-driven selective enrichment influences phage infectivity, genomic variation, and protein structural dynamics, thereby providing new insights into viral adaptation strategies and their implications for microbial ecology and biotechnology.

2. Materials and Methods

2.1. Host Bacteria and Phage

Bacilllus mycoides 41-22 and its associated phage MY22 were both previously isolated from the Mingyong Glacier in earlier experiments. Prior to use, frozen stocks stored at −80 °C were revived and cultivated in LB medium at 28 °C with shaking. When the bacterial culture reached an OD_600_ of approximately 0.7, phage VMY22 was added at a multiplicity of infection (MOI) of 0.1. The mixture of host and phage was incubated overnight, after which the supernatant containing enriched phages was collected and used in subsequent experiments.

2.2. phageInfection Capability Under Temperature Stress

To obtain temperature-adapted phage populations, enrichment was performed at 4 °C, 15 °C, 28 °C, and 32 °C. Specifically, Bacillus mycoides 41-22 cultures were grown at the corresponding temperature until OD_600_ = 0.7, followed by infection with phage VMY22 at an MOI of 0.1. The host–phage mixtures were incubated overnight, and the resulting lysates were clarified by centrifugation. Phage titers were standardized according to an established calibration curve. These phages (P_4_, P_15_, P_28_′, and P_32_) were then used for infection assays as described below and phage titers were determined by plaque assays and expressed as plaque-forming units per milliliter (pfu/mL). All subsequent infections were performed after adjusting phage stocks to the same titer. We note that infections were conducted at 28 °C to ensure consistency across treatments; therefore, this assay primarily reflects post-adsorption events rather than adsorption kinetics at different temperatures. In addition, enrichment was attempted at 34 °C, generating a variant designated P_34_. However, P_34_ was unable to adapt and showed a rapid loss of infectivity. All infection and enrichment assays were performed in triplicate (n = 3). Data are presented as mean ± SD. Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc test (GraphPad Prism, Version 9.3.0).

2.3. Infection of the Same Phageon Host Bacteria at Different Temperatures

The host bacterium Bacillus mycoides 41-22 was cultured in a shaking incubator at 4 °C, 15 °C, 28 °C, and 32 °C until the OD_600_ = 0.7 (denoted as B_4_, B_15_, B_28_, and B_32_). Then, 1 mL of each culture was transferred to a 5 mL centrifuge tube, and VMY22 phage enriched at 28 °C was added (MOI = 0.1). The mixture was incubated for 30 min at the corresponding growth temperature of the host bacterium. After incubation, the titer of phage infecting the same host bacterium was determined using the double-layer agar method. The experiment was repeated three times.

2.4. phageBase Mutation Site Analysis

The target genes, including the ATPase gene, transcriptional regulator rho factor gene, capsid protein gene, tail fiber protein gene, and endolysin gene, from the temperature-adapted phage P_4_, P_15_, P_28_’, and P_32_, were amplified in vitro. Bioinformatics analysis of the sequencing data was performed using biological software such as EditSeq (Version DNAStar v7.1), SeqMan (Version DNAStar v7.1), and MegAlign (Version DNAStar v7.1).

2.5. phageAmino Acid Sequence Analysis

Based on the nucleotide sequences of phage P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures, the nucleotide sequences were translated into amino acid sequences using DNAman (Lynnon Biosoft, San Ramon, CA, USA). Oligo (Molecular Biology Insights, Cascade, CO, USA) was used only for sequence handling and format preparation. The translated amino acid sequences were then aligned and compared using DNAman to identify sequence variations among the phage variants.

2.6. Prediction of phageProtein Structure

Protein Structure Prediction Using SWISS-MODEL with Homology Modeling.

Template Identification: The amino acid sequence is input into the Swiss-Model search box for template identification. Based on sequence alignment and homology levels, suitable protein templates are selected. The template identified from the reference amino acid sequence serves as the control group, while the template from the mutated amino acid sequence serves as the experimental group.

Conformation Search: BLAST and HHblits are used to align the target protein with the initial template. The spatial structure of the target protein is constructed based on the alignment results.

Conformation Optimization: For unaligned regions of the target protein, the loop structure database is used to find the best structural fragments.

Conformation Evaluation: The QMEAN scoring function is used to evaluate the predicted protein conformation, and the optimal conformation is selected.

Conformation Comparison: The predicted conformations of the control and experimental groups are compared, and structural differences between the conformations are analyzed.

All structural analyses were performed in silico using homology modeling, and no experimental validation (e.g., CD spectroscopy or thermal stability assays) was conducted in this study.

2.7. Stability of Adaptive Evolution in Phage VMY22

The phages with an initial titer of 1 × 10^7^ pfu/mL, previously enriched, were placed at 4 °C, 15 °C, 28 °C, and 32 °C in separate incubators, labeled as P_4_, P_15_, P_28_, and P_32_. These represent the non-evolved parental phage populations that were stored directly at the indicated temperatures without prior temperature-specific enrichment. Simultaneously, the evolved phages were diluted to a uniform titer of 1 × 10^7^ pfu/mL and labeled as P_4_, P_15_, P_28_’, and P_32_. These phages were also placed in incubators at 4 °C, 15 °C, 28 °C, and 32 °C. These represent the non-evolved parental phage populations that were stored directly at the indicated temperature without prior temperature-specific enrichment. Phage titers were measured every 10 days using the double-layer agar method. The experiments were repeated three times.

3. Results and Analysis

3.1. Temperature Stress-Driven Phage Titer Variation Curve

Using enrichment time as the x-axis and phage titer as the y-axis, a curve was established to illustrate the dynamics of phages at different temperatures (Figure 1). For P_4_ and P_15_, titers initially decreased during continuous enrichment. However, as enrichment proceeded, adaptive evolution emerged and the values began to rise, eventually reaching a plateau. In P_28_’, the concentration increased by nearly two orders of magnitude after sustained enrichment before stabilizing. Similarly, P_32_ exhibited a marked elevation followed by a stable phase. In contrast, P_34_ failed to adapt to the temperature shift and consequently lost infectivity, showing a continuous decline.

3.2. Comparison of phageInfection Ability Under Temperature Stress

After standardizing the titers of P_4_, P_15_, P_28_’, and P_32_ to 2 × 10^8^ pfu/mL, they were used to infect host bacteria B_28_ cultured at 28 °C, with the results shown in Table 1. Compared to P_28_’, the titers of P_4_, P_15_, and P_32_ decreased by 48, 3.0, and 666.7-fold, respectively. This indicates infectivity varied noticeably in the infectivity of different phages at the same titer on the same host, demonstrating that phages enriched at different temperatures have undergone temperature-adaptive evolution.

3.3. phageInfection of Host Bacteria at Different Temperatures

Host bacteria B_4_, B_15_, B_28_, and B_32_, cultured at different temperatures, were standardized to a concentration of 1 × 10^7^ cfu/mL using the plate count method. Phage P_28_’, enriched at 28 °C, was diluted to a titer of 1 × 10^8^ pfu/mL and used to infect host bacteria B_4_, B_15_, B_28_, and B_32_ with the same colony count [9]. The results are shown in Table 2. Compared to the infection of B_28_, the titers of P_28_’ decreased by 5.2, 3.2, and 52.8-fold when infecting B_4_, B_15_, and B_32_, respectively. This demonstrates that the physiological state of the host bacterium, determined by its growth temperature, significantly affects the infectivity of the phage (Table 2). These results highlight a key environmental variable (host physiology) that may exert selective pressure during phage enrichment. The direct evidence for temperature-driven adaptation in the phages themselves comes from the comparison of differentially enriched phage populations under standardized infection conditions, as shown in Section 3.2 (Table 1).

3.4. Sequencing Analysis of Target Fragment

3.4.1. Sequencing Results of ATPase Gene from Different Phages

Using the genomes of phages P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures as templates, the ATPase gene was PCR-amplified, and the sequencing results were compared (Figure 2a). It is evident that the ATPase genes of phages P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures accumulated numerous mutations. Using the P_28_ gene sequence as the reference, sequence homology was 91.40% for P_4_, 49.38% for P_15_, 93.50% for P_28_’, and 89.19% for P_32_. The P_15_ sequence showed the lowest homology, indicating the highest number of mutation sites, while P_28_’ exhibited the highest homology due to its cultivation temperature being close to its original isolation temperature.

3.4.2. Sequencing Results of Lysin Gene from Different Phages

Using the genomes of phages P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures as templates, the endolysin gene was PCR-amplified, and the sequencing results were compared (Figure 3a). The results show that the endolysin-encoding genes of phages P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures underwent mutations. Using the P_28_ gene sequence as the reference, sequence homology was 55.26% for P_4_, 55.53% for P15, 92.95% for P_28_’, and 90.38% for P_32_. The P_4_ sequence showed the lowest homology, suggesting that the endolysin gene underwent the most mutations to adapt to low-temperature environments, thereby enhancing cold adaptability. Meanwhile, P_28_’ displayed the highest homology, as its cultivation temperature was close to its original isolation temperature.

3.5. Amino Acid Sequence Analysis of Selected Genes in Temperature-Adaptive Phage

3.5.1. Amino Acid Sequence Analysis of ATPase

The ATPase amino acid sequences of phages P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures were compared, as shown in Figure 2b. Using the amino acid sequence of P_28_ as a reference, homology was 83% for P_4_, 93% for P_15_, 95% for P_28_’, and 84% for P_32_. This comparison reveals that mutations occurred in the amino acid sequences of phages P_4_, P_15_, P_28_’, and P_32_, with P_4_ exhibiting the highest mutation rate and P_28_’ the lowest.

3.5.2. Amino Acid Sequence Analysis of Endolysin

The amino acid sequences of endolysin from phages P_4_, P_15_, P_28_’, and P_32_ enriched at different temperatures were compared, as shown in Figure 3b. Using the amino acid sequence of P_28_ as a reference, sequence homology was 70% for P_4_, 82% for P_15_, 92% for P_28_’, and 80% for P_32_. This comparison shows that amino acid mutations occurred in the endolysin sequences of phages P_4_, P_15_, P_28_’, and P_32_, with P_4_ exhibiting the highest mutation rate and P_28_’ the lowest.

3.6. Protein Structure Prediction of Mutant Phage Under Temperature Stress

3.6.1. ATPase Protein Structure Prediction

Protein Structure Prediction

According to protein structure predictions for the ATPase gene of phage VMY 22 enriched at different temperatures, generated using the Swiss-Model website, the results are shown in Figure 4a. Analysis indicates that this protein is a member of a subgroup within the large ASCE (Additional Strand Catalytic Glutamate) NTPase superfamily [10]. Members of this family participate in numerous macromolecular tasks, including chromosome segregation, DNA recombination, strand separation and binding, protein degradation, and the generation and maintenance of concentration gradients and electrostatic potentials. Structurally, the protein belongs to the Rossmann fold, consisting of five parallel β-strands interspersed with four α-helices. This arrangement leads to the C-terminus of each strand forming one edge of the β-sheet, with α-helices present on both faces of the sheet [11]. Predicted protein structures of ATPase are shown: (1) reference sequence, (2) ATPase-4, (3) ATPase-15, (4) ATPase-28, and (5) ATPase-32.

Protein Structure Alignment

The protein structures of ATPase-4, ATPase-15, ATPase-28, and ATPase-32 were superimposed on the ATPase protein structure for comparative analysis, as shown in Figure 4b. Modeling suggests that the C-terminal α-helix of the ATPase could potentially mediate interactions with host membrane components; however, this inference is based on in silico models and requires experimental validation. As temperature increases, the hydrophobicity and charge of the non-polar surface of this α-helix are enhanced, thereby improving its high-temperature adaptability [12].

The model predicts that compared to the reference ATPase, ATPase-4 may stabilize its structure at 4 °C by increasing the charge on its polar surface, potentially facilitating the activation of water molecules for hydrolysis (Figure 4b(1)) [13,14]. The predicted conformational change in ATPase-15 might affect adjacent subunits, and the observed increase in hydrophobicity of the helical region could contribute to enhanced structural stability at 15 °C (Figure 4b(2)) [15]; the structure of ATPase-28 did not undergo large-scale rearrangements in the model. Its stability at 28 °C may be enhanced by substitutions that increase the hydrophobicity of the helical surface (Figure 4b(3)) [16]; in the ATPase-32 model, Lys166 is position such that it could interact with β/γ phosphates and form a putative salt bridge, and the model also predicts a possible Mg^2+^ coordination (carbonyl oxygens of Ser101, Glu111, Asn132, and His157). These features may contribute to altered surface charge and stability, but biochemical assays are needed to confirm their functional relevance. Stability at 32 °C is thus maintained primarily by increasing the charge on the non-polar surface of the helical region (Figure 4b(4)) [17,18].

3.6.2. Endolysin Protein Structure Prediction

Protein Structure Prediction

According to protein structure predictions for the endolysin gene of phage VMY22 enriched at different temperatures, generated using the Swiss-Model website, the results are shown in Figure 5a. This type of protein exhibits the typical α-helical structure of the lysozyme family, consisting of seven α-helices. Its active site is formed by a conserved catalytic triad: Glu64, Asp73, and Thr79, with Glu64 forming a salt bridge with a conserved Arg185 located on an α-helix [19].

Protein Structure Alignment

The protein structures of endolysin-4, endolysin-15, endolysin-28, and endolysin-32 were superimposed on the endolysin structure for comparative analysis, as shown in Figure 5b. Compared to the control, endolysin-4 displayed the lowest protein homology, while endolysin-28 had the highest. The random coil regions within the endolysin protein structure exhibit considerable flexibility, allowing rapid responses to environmental changes. In our models, random coil regions were predicted in endolysin and ATPase variants. Although such regions are often considered less stable than α-helices at higher temperatures [20,21], our results are based on sequence alignment and in silico modeling only. Experimental confirmation will be required to establish whether these structural tendencies occur in this phage system [22].

The model predicts numerous amino acid substitutions in endolysin-4. These substitutions are associated with increased hydrophobicity and charge redistribution, which could promote the transition of random coil structures from a disordered to a more ordered state, thereby potentially enhancing protein stability at 4 °C (Figure 5b(1)) [23]; the predicted amino acid substitutions and synonymous mutations in endolysin-15 might effectively reduce the likelihood of random coils forming α-helices. The presence of additional hydrophobic residues could help stabilize the protein structure at 15 °C (Figure 5b(2)) [24,25]; the polypeptide sequence of endolysin-28 is highly conserved. The model suggests that the high level of conservation reduces the propensity for structural fluctuations, thereby potentially enhancing the stability of the protein structure at 28 °C (Figure 5b(3)) [26]. With a further increase in temperature, the model predicts that maintaining the stability of the endolysin-32 structure requires significant energy, potentially achieved through reinforced interactions between amino acid residues within the protein structure (Figure 5b(4)) [27,28].

It is crucial to note that the functional interpretations of flexibility and stability are based solely on sequence alignment and in silico modeling. Experimental confirmation (e.g., via circular dichroism spectroscopy or differential scanning calorimetry) is required to establish whether these predicted structural tendencies occur in this phage system.

3.7. Stability of Adaptive Evolution in Phage VMY22

The evolved phages VMY22, specifically P_4_, P_15_, P_28_’, and P_32_, were stored in incubators at 4, 15, 28, and 32 °C, respectively, with titers measured at different time intervals using the double-layer agar method (Figure 6). The results show that the titers of the adapted phages P_4_, P_15_, P_28_’, and P_32_ remained stable, indicating good stability suitable for long-term storage at their respective enrichment temperatures. In contrast, the titers of the non-adapted phages P_4_, P_15_, P_28_, and P_32_, which were the parental population stored at corresponding temperatures but without undergoing the prior adaptive enrichment process, decreased noticeably after approximately 30 days of storage. The titer reduction was smaller for P_4_* and P_15_* but notably larger for P_28_* and P_32_*, with decreases close to an order of magnitude. This suggests that adaptive evolution enhances phage storage stability, with lower temperatures being particularly favorable for phage preservation.

4. Discussion

In this study, we demonstrated that phage VMY22 undergoes temperature-driven adaptive evolution, primarily through mutations in functional proteins such as ATPase and endolysin, while structural proteins like capsid and tail fiber remain highly conserved. This evolutionary strategy allows the phage to maintain virion stability and host recognition while fine-tuning post-adsorption processes in response to fluctuating thermal conditions [29,30,31,32,33].

Our findings align with previous work on mesophilic and thermophilic phages. For instance, work on single-stranded RNA phage Qβ and Listeria phages has shown that elevated temperatures can considerably alter adsorption kinetics, replication efficiency, and evolutionary paths [7,8]. Similarly, research on phage phi29 and related DNA packaging motors has highlighted the dynamic role of ATPase domains in environmental adaptation [34,35,36,37]. However, unlike these well-studied systems, psychrophilic phages—particularly those from glacial environments—remain underexplored. Our study helps bridge this gap by revealing that cold-active phages like VMY22 also employ targeted functional protein remodeling to cope with thermal stress, underscoring a universal yet flexible adaptive mechanism across thermal niches [38,39,40,41,42,43].

The concentration of mutations in ATPase and endolysin is mechanistically consistent with their critical roles in the viral life cycle. ATPase drives DNA packaging and energy transduction, while endolysin facilitates host cell lysis—both processes are highly sensitive to temperature-dependent biochemical dynamics [44,45,46,47]. Notably, the pattern of mutations within these genes revealed a temperature-dependent strategy: the P_15_ variant (adapted to 15 °C) accumulated the highest number of nucleotide substitutions in the ATPase gene, yet its amino acid sequence remained highly conserved, suggesting a predominance of synonymous mutations or those with minimal structural impact. In contrast, the P4 variant (adapted to 4 °C) exhibited the highest rate of amino acid changes, indicating a greater reliance on non-synonymous mutations to remodel protein structure under extreme cold stress. This distinction implies that adaptive evolution may proceed through fine-tuning of gene expression or translation at moderate low temperatures (e.g., 15 °C), while requiring substantial protein remodeling near the lower thermal limit (4 °C). Homology modeling suggested that these mutations, particularly the non-synonymous ones in P_4_, enhance protein stability through mechanisms such as altered hydrophobicity, charge redistribution in helical regions of ATPase, and increased cysteine content in endolysin. These modifications appear to improve protein performance under stress without compromising structural integrity, reflecting an evolutionary trade-off that maximizes adaptability while preserving infectivity [48,49].

From an ecological perspective, this adaptive mechanism may be critical for phage persistence in extreme habitats like glaciers, where temperatures fluctuate widely [50]. By stabilizing functional proteins, cold-active phages such as VMY22 can sustain infection and contribute to microbial turnover, nutrient cycling, and community stability in cold ecosystems [51,52]. Beyond ecological relevance, these findings have biotechnological implications, particularly for applications requiring phage stability under low-temperature conditions, such as biocontrol in refrigerated foods or environmental monitoring in cold climates [53].

Several limitations of this study should be acknowledged. First, our structural inferences are based solely on in silico modeling without experimental validation (e.g., circular dichroism spectroscopy or thermal shift assays). Second, we did not perform synchronized adsorption assays or one-step growth curves across temperatures, which would help decouple adsorption from post-adsorption effects. Third, our conclusions are drawn from a limited set of PCR-amplified loci rather than whole-genome sequencing. Thus, while we observed conservation in the structural genes examined, we cannot rule out mutations in other genomic regions. Future studies integrating biophysical assays, mutagenesis, and full-genome sequencing will be essential to validate and expand upon these findings.

In conclusion, our study reveals a conserved yet flexible evolutionary strategy in which phage VMY22 maintains structural gene stability while functionally remodeling ATPase and endolysin to cope with thermal stress. These results not only expand our understanding of phage adaptation to extreme environments but also provide a framework for future ecological and applied research on cold-active viruses.