I. Molecular evolution

Why is life on Earth the way it is? Molecular evolutionary study can help elucidate many aspects of this question. Study of molecular systems that constitute a living cell unravels lots of extraordinary, unexpected "approaches" nature takes in order for life to adapt and proliferate.

Amino acid sensor found across the Tree of Life

Amino acids are the building blocks of life. However, despite their common structure and indispensable nature no common mechanism of amino acid binding is known. Does a universal mechanism of amino acid binding exist? My research showed that it indeed does.

While studying bacterial chemoreceptors I have discovered a universal amino acid binding motif present in all cellular life forms, from bacteria to humans (link to the paper). This motif is present in a subclass of a sensor domain called dCache_1. In Bacteria and Archaea this motif exclusively binds amino acids, including gamma-aminobutyric acid (GABA), and it is present in all major signal transduction proteins. In humans, I found this motif in α2δ subunit of high voltage-gated calcium channels that are implicated in numerous diseases including neuropathic pain and neurodevelopmental disorders, and in a recently characterized CACHD1 protein that has been suggested to play a modulatory role in low voltage-gated calcium channels.
​
​Identification of the amino acid motif allowed me to get deep insights into the structure of α2δ and CACDH1 proteins and uncover extraordinary evolutionary events that gave rise to these eukaryotic proteins. It is truly amazing to "witness" how bacterial proteins shaped a particular aspect of the nervous system in Metazoa. Site-directed mutagenesis performed with bacterial and mammalian proteins showed that mutation of key positions of the motif leads to drastic reduction in binding of amino acids and GABA-derived drugs. My study with colleagues strongly suggests that GABA-derived drugs bind to the same motif in human α2δ subunits that binds natural GABA ligands and alpha amino acids in bacterial signal transduction proteins. The exact location of the binding site on a target protein and the mechanism of binding enable future improvements of drugs targeting pain and neurobiological disorders.

Gumerov V.M., Andrianova E.P., Matilla M.A., Page K.M. , Monteagudo-Cascales E., Dolphin A.C., Krell T., Zhulin I.B. (2022) Amino acid sensor conserved from bacteria to humans. PNAS, 119 (10) e2110415119. 

Amine sensor evolved from the amino acid sensor

Biogenic amines are of great physiological importance for microorganisms and  humans.  They  serve  as  substrates  for  aerobic  and  anaerobic  growth  and  play  a  role of neurotransmitters and osmoprotectants. Three bacterial chemoreceptors from closely related proteobacterial species were previously shown to bind and respond to quaternary amines and polyamines as signaling molecules. However, their relationships, evolutionary origins, binding determinants, and prevalence were unknown. 

In this study I identified thousands of bacterial and archaeal receptors containing dCache_1AM domains that bind various  biogenic  amines. I computed a conserved sequence motif signature for this class of sensory  domains, which was experimentally verified, and we showed that dCache_1AM sensors bind not only quaternary but also primary, secondary, and tertiary amines as well as the polyamine ethylenediamin. I also showed that amine sensors evolved from the universal amino acid receptor through  a  small  insertion  in  the  ligand- binding pocket and replacement of key ligand- binding residues. Structural analyses of bacterial and archaeal receptors showed that the ligand- binding interface is well conserved in phylogenetically distant species. By revealing that amine- sensing receptors originated from amino acid–sensing receptors, we show how receptors can change their specificity during evolution. This study  further  demonstrates  that  our  approach  is  applicable  to  characterizing other ligand- binding domain families thus leading to substantial gain in knowledge on signal transduction systems. 

*Cerna-Vargas J-P., *Gumerov V.M., Krell T., Zhulin I.B. (2023) Amine recognizing domain in diverse receptors from bacteria and archaea evolved from the universal amino acid sensor. PNAS, 120, e2305837120

Purine receptor shares a common ancestor with the amino acid sensor

Purines are central intermediates in the synthesis of nucleotides and nucleic acids and are among the most abundant metabolites of living organisms. Purine metabolites also provide cells with necessary energy and cofactors to promote cell survival and proliferation and act as signaling molecules in eukaryotes to coordinate multiple cell behaviors and physiological processes. However, their role as signal molecules in bacteria is largely unexplored.

In this study, we used computational and experimental approaches to identify sensors that specifically bind purines and, with lower affinity, pyrimidines. By combining analysis of the newly solved 3D structure of the McpH-LBD in complex with uric acid and comparative protein sequence analysis, we  identified a sequence motif for purine binding in thousands of dCache_1 domains. Purine binding dCache_1 domains were identified in four major receptor families - chemoreceptors, histidine kinases, diguanylate cyclases and phosphodiesterases, and Ser/Thr phosphatases - all of which play important roles in regulating bacterial virulence.  We further demonstrate the physiological relevance of purine sensing using a signaling protein, which modulates second messenger (c-di-GMP) turnover in a human pathogen V. cholerae. 

Our analysis showed that purine sensing domains share a common ancestor with the ubiquitous amino acid sensors.

​*Monteagudo-Cascales E., *Gumerov V.M., Fernández M., Matilla M.A., Gavira J.A., Zhulin I.B., Krell T. (2024) Ubiquitous purine sensor modulates diverse signal transduction pathways in bacteria. Nature Communications, 15, 5867 (2024)

A new class of metallo-β-lactamases 

Many bacteria contain cytoplasmic chemoreceptors that lack sensor domains. In the collaborative work we demonstrated that a subclass of such cytoplasmic receptors found in 8 different bacterial and archaeal phyla genetically couple to a specific group of metalloproteins.

While studying these metalloproteins I discovered that they all have the same unique metal-binding motif and represent a new previously unknown class of metallo-β-lactamases (link to the paper). They are related to the family of flavodiiron proteins (FDPs) and serve as oxygen and/or nitric oxide reductases. I have established that this protein co-evolves with specific cytoplasmic chemoreceptors and was recruited to signal transduction as sensor module.
​
Based on the experiments with the human pathogen Treponema denticola the protein was found to bind both iron and oxygen and oxygen binding is reversible. Furthermore, the protein was found to mediate chemotaxis toward iron and away from oxygen.

Muok A.R., Deng Y., Gumerov V.M., Chong J.E., DeRosa J.R., Kurniyati K., Coleman R.E., Lancaster K.M., Li C., Zhulin I.B, Crane B.R. (2019) A di-iron protein recruited as an Fe[II] and oxygen sensor for bacterial chemotaxis functions by stabilizing  an iron-peroxy species. PNAS, 116: 14955-14960.

Molecular evolution of the NusG transcription factor paralog RfaH

Housekeeping regulators and their specialized paralogs are the only universally conserved transcription factors. They are represented by well-studied NusG and RfaH. Despite their ubiquity, little information was available on the evolutionary origins, functions, and gene targets of the NusG family members. In contrast to NusG which is universally present in prokaryotes RfaH is mostly limited to Proteobacteria and lacks common gene neighbors. It activates only a few xenogeneic operons that are otherwise silenced by NusG and Rho.

Phylogenetic reconstructions revealed extensive duplications and horizontal transfer of rfaH genes. My computational work helped establish that NusG is not only ubiquitous in Bacteria but also common in plants and Chromista, in which it likely modulates the transcription of plastid genes (paper).

Markov clustering and phylogenetic analysis of RfaH sequences allowed to propose a step-wise conversion of a NusG duplicate copy into a sequence-specific regulator which excludes NusG from its targets but does not compromise the regulation of house-keeping genes. Gene duplication and lateral transfer gave rise to a surprising diversity within this family of transcription factors.

Wang B., Gumerov V.M., Andrianova E.P., Zhulin I.B., Artsimovitch I. (2020) Origins and Molecular Evolution of the NusG Paralog RfaH. mBio, 11: e02717-20.

II. Computational pipelines and databases

Computational tools and databases are the hallmarks of bioinformatics and computational biology. They allow for large-scale studies that lead to the discovery of new phenomena and initiate new research directions.

Microbial Signal Transduction database (MiST)

With the team of talented bioinformaticians I have created a comprehensive online database of microbial signal transduction systems called MiST (https://mistdb.com/). The database includes more than 200,000 sequenced genomes and metagenome-assembled genomes and their signal transduction profiles. The database is used in numerous projects around the world.

Gumerov V.M., Ortega D.R., Adebali O., Ulrich L.E., Zhulin I.B. (2020) MiST 3.0: an updated microbial signal transduction database with an emphasis on chemosensory systems. Nucleic Acids Res., 48: D459-D464.
​&Gumerov V.M., Ulrich L.E., &Zhulin I.B. (2024) MiST 4.0: a new release of the microbial signal transduction database, now with a metagenomic component. Nucleic Acids Research, 52: D647–D653.

Computational platform for protein evolution and function analysis (TREND)

Computational analysis of protein evolution and function can lead to deep insights and unexpected discoveries. However a comprehensive analysis is extremely time-consuming and error-prone. To help biologists perform evolutionary analysis more efficiently I have created a computational platform called TREND (http://trend.evobionet.com/). The platform automates many steps in the computational analysis of proteins and puts all the information obtained into a phylogenetic context.

Gumerov V.M. and Zhulin I.B. (2020) TREND: a platform for exploring protein function in prokaryotes based on phylogenetic, domain architecture and gene neighborhood analyses. Nucleic Acids Res., 48: W72-W76.

III. Extremophilic Bacteria and Archaea

It has been estimated that bacteria account for ~60% of Earth’s biomass. Many bacteria cause disease in humans and other animals. The analysis of whole genomes, boosted by next-generation sequencing, has had profound effects on our understanding of bacteria and archaea.

Genomes

Comparative genomics and phylogenetics help discover numerous new phenomena: regulatory elements, protein domains, recruitment of domains to new biological processes, viral elements, sequence motifs, etc. Microorganisms that live in extreme environments have developed various unique molecular strategies to adapt to such conditions. These mechanisms and in particular enzymes encoded in the genomes of these organisms find important industrial, medical and research application (for example, thermostable Taq polymerase)  

As a graduate student I went on an expedition to Kamchatka to collect samples from the habitats with extreme conditions to study how genomes of the extremophilic microorganisms are organized. I studied genomes and encoded metabolic pathways of archaea and bacteria isolated from such harsh environments as boiling acidic springs (ref1, ref2, ref3, ref4) and hypersaline soda lakes (ref5, ref6) and uncovered unusual metabolic adaptations of these organisms. In one of such projects my analysis enabled identification and description of two new classes of bacteria, Chitinivibrionia (ref5) and Chitinispirillia (ref6), within the Fibrobacteres phylum.

Genomes deposited to NCBI:
Vulcanisaeta moutnovskia 768-28
Chitinivibrio alkaliphilus ACht1
Chitinispirillum alkaliphilum Acht6-1
Pyrobaculum ferrireducens 1860
Thermoproteus uzoniensis 768-20
Fervidicoccus fontis Kam940

Enzymes

Enzymes from bacteria and archaea have a wide application in various industrial processes. Extremophilic microorganisms represent a rich source of enzymes with unusual properties. Enzymes isolated from extremophyles, or extremozymes, are able to catalyze chemical reactions under extreme conditions, similar to those found in industrial processes. Due to their stability under such conditions, extremozymes offer new catalytic alternatives for industrial applications.

My analysis of extremophilic archaea resulted in identification and biochemical characterization of several thermostable enzymes with unique properties: multifunctional beta-glycosidase (ref1), lipase (ref2), alcohol dehydrogenase (ref3, ref4, ref5), superoxide dismutase (ref6), M42 aminopeptidase (ref7), and prolidase (ref8).
​
Structures deposited to PDB:
Beta-glycosidase from Acidilobus saccharovorans:
    - in complex with Tris: PDB 4HA3
    - in complex with glycerol: PDB 4HA4
Alcohol dehydrogenase from  Thermococcus sibiricus: PDB 3TN7
Aldehyde dehydrogenase from Pyrobaculum ferrireducens 1860:
    - in apo form: PDB 5EEB
​    - in complex with NADP+: PDB 5F2C, 4NMJ, 4NMK, 4H73
Iron superoxide dismutase from Acidilobus saccharovorans: 4FFK
Prolidase from Thermococcus sibiricus: PDB 4FKC

 Microbial communities

Investigation of the microbial communities associated with thermal ecological niches is of great interest for evolutionary biology as many microorganisms living in these conditions belong to ancient evolutionary lineages of thermophilic bacteria and archaea. 

Hot springs have been investigated since the XIX century, but isolation and examination of their thermophilic microbial inhabitants did not start until the 1950s. Many thermophilic microorganisms and their viruses have since been discovered. 16S rRNA-based studies subsequently revealed that microbial diversity was much broader than suggested by culture-dependent techniques.
​
I investigated microbial communities of hot and acidic springs of Kamchatka volcanic regions. One of my findings is that phylogenetically diverse microbial communities consisting of various metabolic groups of prokaryotes can thrive even under the most extreme conditions of low pH and high temperatures (ref1, ref2, ref3). Remarkably, in the thermal groundwater of the Uzon volcano crater Archaea are the predominant component of microbial communities (ref4)! Another interesting finding is that in several acidic thermal pools Nanoarchaeota - an obligate symbiont on another archaeon Ignicoccus - comprise a significant part of the communities (ref1). It is interesting to note, that the genome of this organism lacks essentially all the genes required for the synthesis of nucleotides, amino acids, cofactors, and lipids, but encodes all the genes necessary for repair and replication.