Horizontal Cell Biology: Monitoring Global Changes of Protein Interaction States with the Proteome-Wide Cellular Thermal Shift Assay (CETSA).

The cellular thermal shift assay (CETSA) is a biophysical technique allowing direct studies of ligand binding to proteins in cells and tissues. The proteome-wide implementation of CETSA with mass spectrometry detection (MS-CETSA) has now been successfully applied to discover targets for orphan clinical drugs and hits from phenotypic screens, to identify off-targets, and to explain poly-pharmacology and drug toxicity. Highly sensitive multidimensional MS-CETSA implementations can now also access binding of physiological ligands to proteins, such as metabolites, nucleic acids, and other proteins. MS-CETSA can thereby provide comprehensive information on modulations of protein interaction states in cellular processes, including downstream effects of drugs and transitions between different physiological cell states. Such horizontal information on ligandmodulation in cells is largely orthogonal to vertical information on the levels of different proteins and therefore opens novel opportunities to understand operational aspects of cellular proteomes.

Modulation of Protein-Interaction States through the Cell Cycle.

Global profiling of protein expression through the cell cycle has revealed subsets of periodically expressed proteins. However, expression levels alone only give a partial view of the biochemical processes determining cellular events. Using a proteome-wide implementation of the cellular thermal shift assay (CETSA) to study specific cell-cycle phases, we uncover changes of interaction states for more than 750 proteins during the cell cycle. Notably, many protein complexes are modulated in specific cell-cycle phases, reflecting their roles in processes such as DNA replication, chromatin remodeling, transcription, translation, and disintegration of the nuclear envelope. Surprisingly, only small differences in the interaction states were seen between the G1 and the G2 phase, suggesting similar hardwiring of biochemical processes in these two phases. The present work reveals novel molecular details of the cell cycle and establishes proteome-wide CETSA as a new strategy to study modulation of protein-interaction states in intact cells.

Understanding transport by the major facilitator superfamily (MFS): structures pave the way.

Members of the major facilitator superfamily (MFS) of transport proteins are essential for the movement of a wide range of substrates across biomembranes. As this transport requires a series of conformational changes, structures of MFS transporters captured in different conformational states are needed to decipher the transport mechanism. Recently, a large number of MFS transporter structures have been determined, which has provided us with an unprecedented opportunity to understand general aspects of the transport mechanism. We propose an updated model for the conformational cycle of MFS transporters, the 'clamp-and-switch model', and discuss the role of so-called 'gating residues' and the substrate in modulating these conformational changes.

IMPRINTS.CETSA and IMPRINTS.CETSA.app: an R package and a Shiny application for the analysis and interpretation of IMPRINTS-CETSA data.

IMPRINTS-CETSA (Integrated Modulation of Protein Interaction States-Cellular Thermal Shift Assay) provides a highly resolved means to systematically study the interactions of proteins with other cellular components, including metabolites, nucleic acids and other proteins, at the proteome level, but no freely available and user-friendly data analysis software has been reported. Here, we report IMPRINTS.CETSA, an R package that provides the basic data processing framework for robust analysis of the IMPRINTS-CETSA data format, from preprocessing and normalization to visualization. We also report an accompanying R package, IMPRINTS.CETSA.app, which offers a user-friendly Shiny interface for analysis and interpretation of IMPRINTS-CETSA results, with seamless features such as functional enrichment and mapping to other databases at a single site. For the hit generation part, the diverse behaviors of protein modulations have been typically segregated with a two-measure scoring method, i.e. the abundance and thermal stability changes. We present a new algorithm to classify modulated proteins in IMPRINTS-CETSA experiments by a robust single-measure scoring. In this way, both the numerical changes and the statistical significances of the IMPRINTS information can be visualized on a single plot. The IMPRINTS.CETSA and IMPRINTS.CETSA.app R packages are freely available on GitHub at https://github.com/nkdailingyun/IMPRINTS.CETSA and https://github.com/mgerault/IMPRINTS.CETSA.app, respectively. IMPRINTS.CETSA.app is also available as an executable program at https://zenodo.org/records/10636134.

Deep learning enables the discovery of a novel cuproptosis-inducing molecule for the inhibition of hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is one of the most common and deadly cancers in the world. The therapeutic outlook for HCC patients has significantly improved with the advent and development of systematic and targeted therapies such as sorafenib and lenvatinib; however, the rise of drug resistance and the high mortality rate necessitate the continuous discovery of effective targeting agents. To discover novel anti-HCC compounds, we first constructed a deep learning-based chemical representation model to screen more than 6 million compounds in the ZINC15 drug-like library. We successfully identified LGOd1 as a novel anticancer agent with a characteristic levoglucosenone (LGO) scaffold. The mechanistic studies revealed that LGOd1 treatment leads to HCC cell death by interfering with cellular copper homeostasis, which is similar to a recently reported copper-dependent cell death named cuproptosis. While the prototypical cuproptosis is brought on by copper ionophore-induced copper overload, mechanistic studies indicated that LGOd1 does not act as a copper ionophore, but most likely by interacting with the copper chaperone protein CCS, thus LGOd1 represents a potentially new class of compounds with unique cuproptosis-inducing property. In summary, our findings highlight the critical role of bioavailable copper in the regulation of cell death and represent a novel route of cuproptosis induction.

Recent advances in proteome-wide label-free target deconvolution for bioactive small molecules.

Small-molecule drugs modulate biological processes and disease states through engagement of target proteins in cells. Assessing drug-target engagement on a proteome-wide scale is of utmost importance in better understanding the molecular mechanisms of action of observed beneficial and adverse effects, as well as in developing next generation tool compounds and drugs with better efficacies and specificities. However, systematic assessment of drug-target engagement has been an arduous task. With the continuous development of mass spectrometry-based proteomics instruments and techniques, various chemical proteomics approaches for drug target deconvolution (i.e., the identification of molecular target for drugs) have emerged. Among these, the label-free target deconvolution approaches that do not involve the chemical modification of compounds of interest, have gained increased attention in the community. Here we provide an overview of the basic principles and recent biological applications of the most important label-free methods including the cellular thermal shift assay, pulse proteolysis, chemical denaturant and protein precipitation, stability of proteins from rates of oxidation, drug affinity responsive target stability, limited proteolysis, and solvent-induced protein precipitation. The state-of-the-art technical implications and future outlook for the label-free approaches are also discussed.

New ideas for non-animal approaches to predict repeated-dose systemic toxicity: Report from an EPAA Blue Sky Workshop.

The European Partnership for Alternative Approaches to Animal Testing (EPAA) convened a 'Blue Sky Workshop' on new ideas for non-animal approaches to predict repeated-dose systemic toxicity. The aim of the Workshop was to formulate strategic ideas to improve and increase the applicability, implementation and acceptance of modern non-animal methods to determine systemic toxicity. The Workshop concluded that good progress is being made to assess repeated dose toxicity without animals taking advantage of existing knowledge in toxicology, thresholds of toxicological concern, adverse outcome pathways and read-across workflows. These approaches can be supported by New Approach Methodologies (NAMs) utilising modern molecular technologies and computational methods. Recommendations from the Workshop were based around the needs for better chemical safety assessment: how to strengthen the evidence base for decision making; to develop, standardise and harmonise NAMs for human toxicity; and the improvement in the applicability and acceptance of novel techniques. "Disruptive thinking" is required to reconsider chemical legislation, validation of NAMs and the opportunities to move away from reliance on animal tests. Case study practices and data sharing, ensuring reproducibility of NAMs, were viewed as crucial to the improvement of non-animal test approaches for systemic toxicity.

The Cellular Thermal Shift Assay: A Novel Biophysical Assay for In Situ Drug Target Engagement and Mechanistic Biomarker Studies.

A drug must engage its intended target to achieve its therapeutic effect. However, conclusively measuring target engagement (TE) in situ is challenging. This complicates preclinical development and is considered a key factor in the high rate of attrition in clinical trials. Here, we discuss a recently developed, label-free, biophysical assay, the cellular thermal shift assay (CETSA), which facilitates the direct assessment of TE in cells and tissues at various stages of drug development. CETSA also reveals biochemical events downstream of drug binding and therefore provides a promising means of establishing mechanistic biomarkers. The implementation of proteome-wide CETSA using quantitative mass spectrometry represents a novel strategy for defining off-target toxicity and polypharmacology and for identifying downstream mechanistic biomarkers. The first year of CETSA applications in the literature has focused on TE studies in cell culture systems and has confirmed the broad applicability of CETSA to many different target families. The next phase of CETSA applications will likely encompass comprehensive animal and patient studies, and CETSA will likely serve as a very valuable tool in many stages of preclinical and clinical drug development.

A saposin-lipoprotein nanoparticle system for membrane proteins.

A limiting factor in membrane protein research is the ability to solubilize and stabilize such proteins. Detergents are used most often for solubilizing membrane proteins, but they are associated with protein instability and poor compatibility with structural and biophysical studies. Here we present a saposin-lipoprotein nanoparticle system, Salipro, which allows for the reconstitution of membrane proteins in a lipid environment that is stabilized by a scaffold of saposin proteins. We demonstrate the applicability of the method on two purified membrane protein complexes as well as by the direct solubilization and nanoparticle incorporation of a viral membrane protein complex from the virus membrane. Our approach facilitated high-resolution structural studies of the bacterial peptide transporter PeptTSo2 by single-particle cryo-electron microscopy (cryo-EM) and allowed us to stabilize the HIV envelope glycoprotein in a functional state.

Structural analyses of human thymidylate synthase reveal a site that may control conformational switching between active and inactive states.

Thymidylate synthase (TS) is the sole enzyme responsible for de novo biosynthesis of thymidylate (TMP) and is essential for cell proliferation and survival. Inhibition of human TS (hTS) has been extensively investigated for cancer chemotherapy, but several aspects of its activity and regulation are still uncertain. In this study, we performed comprehensive structural and biophysical studies of hTS using crystallography and thermal shift assay and provided the first detailed structural information on the conformational changes induced by ligand binding to the hTS active site. We found that upon binding of the antifolate agents raltitrexed and nolatrexed, the two insert regions in hTS, the functions of which are unclear, undergo positional shifts toward the catalytic center. We investigated the inactive conformation of hTS and found that the two insert regions are also involved in the conformational transition between the active and inactive state of hTS. Moreover, we identified a ligand-binding site in the dimer interface, suggesting that the cavity in the dimer interface could serve as an allosteric site of hTS to regulate the conformational switching between the active and inactive states. On the basis of these findings, we propose a regulatory mechanism of hTS activity that involves allosteric regulation of interactions of hTS with its own mRNA depending on cellular demands for TMP.

Structure of the Open Reading Frame 49 Protein Encoded by Kaposi's Sarcoma-Associated Herpesvirus.

Herpesviruses alternate between the latent and the lytic life cycle. Switching into the lytic life cycle is important for herpesviral replication and disease pathogenesis. Activation of a transcription factor replication and transcription activator (RTA) has been demonstrated to govern this switch in Kaposi's sarcoma-associated herpesvirus (KSHV). The protein encoded by open reading frame 49 from KSHV (ORF49KSHV) has been shown to upregulate lytic replication in KSHV by enhancing the activities of the RTA. We have solved the crystal structure of the ORF49KSHV protein to a resolution of 2.4 Å. The ORF49KSHV protein has a novel fold consisting of 12 alpha-helices bundled into two pseudodomains. Most notably are distinct charged patches on the protein surface, which are possible protein-protein interaction sites. Homologs of the ORF49KSHV protein in the gammaherpesvirus subfamily have low sequence similarities. Conserved residues are mainly located in the hydrophobic regions, suggesting that they are more likely to play important structural roles than functional ones. Based on the identification and position of three sulfates binding to the positive areas, we performed some initial protein-DNA binding studies by analyzing the thermal stabilization of the protein in the presence of DNA. The ORF49KSHV protein is stabilized in a dose-responsive manner by double-stranded oligonucleotides, suggesting actual DNA interaction and binding. Biolayer interferometry studies also demonstrated that the ORF49KSHV protein binds these oligonucleotides. IMPORTANCE: Kaposi's sarcoma-associated herpesvirus (KSHV) is a tumorigenic gammaherpesvirus that causes multiple cancers and lymphoproliferative diseases. The virus exists mainly in the quiescent latent life cycle, but when it is reactivated into the lytic life cycle, new viruses are produced and disease symptoms usually manifest. Several KSHV proteins play important roles in this reactivation, but their exact roles are still largely unknown. In this study, we report the crystal structure of the open reading frame 49 protein encoded by KSHV (ORF49KSHV). Possible regions for protein interaction that could harbor functional importance were found on the surface of the ORF49KSHV protein. This led to the discovery of novel DNA binding properties of the ORF49KSHV protein. Evolutionary conserved structural elements with the functional homologs of ORF49KSHV were also established with the structure.

Molecular insights into substrate recognition and catalytic mechanism of the chaperone and FKBP peptidyl-prolyl isomerase SlyD.

BACKGROUND: Peptidyl-prolyl isomerases (PPIases) catalyze cis/trans isomerization of peptidyl-prolyl bonds, which is often rate-limiting for protein folding. SlyD is a two-domain enzyme containing both a PPIase FK506-binding protein (FKBP) domain and an insert-in-flap (IF) chaperone domain. To date, the interactions of these domains with unfolded proteins have remained rather obscure, with structural information on binding to the FKBP domain being limited to complexes involving various inhibitor compounds or a chemically modified tetrapeptide. RESULTS: We have characterized the binding of 15-residue-long unmodified peptides to SlyD from Thermus thermophilus (TtSlyD) in terms of binding thermodynamics and enzyme kinetics through the use of isothermal titration calorimetry, nuclear magnetic resonance spectroscopy, and site-directed mutagenesis. We show that the affinities and enzymatic activity of TtSlyD towards these peptides are much higher than for the chemically modified tetrapeptides that are typically used for activity measurements on FKBPs. In addition, we present a series of crystal structures of TtSlyD with the inhibitor FK506 bound to the FKBP domain, and with 15-residue-long peptides bound to either one or both domains, which reveals that substrates bind in a highly adaptable fashion to the IF domain through ß-strand augmentation, and can bind to the FKBP domain as both types VIa1 and VIb-like cis-proline ß-turns. Our results furthermore provide important clues to the catalytic mechanism and support the notion of inter-domain cross talk. CONCLUSIONS: We found that 15-residue-long unmodified peptides can serve as better substrate mimics for the IF and FKBP domains than chemically modified tetrapeptides. We furthermore show how such peptides are recognized by each of these domains in TtSlyD, and propose a novel general model for the catalytic mechanism of FKBPs that involves C-terminal rotation around the peptidyl-prolyl bond mediated by stabilization of the twisted transition state in the hydrophobic binding site.

Crystal Structures of Type-II Inositol Polyphosphate 5-Phosphatase INPP5B with Synthetic Inositol Polyphosphate Surrogates Reveal New Mechanistic Insights for the Inositol 5-Phosphatase Family.

The inositol polyphosphate 5-phosphatase INPP5B hydrolyzes the 5-phosphate group from water- and lipid-soluble signaling messengers. Two synthetic benzene and biphenyl polyphosphates (BzP/BiPhPs), simplified surrogates of inositol phosphates and phospholipid headgroups, were identified by thermodynamic studies as potent INPP5B ligands. The X-ray structure of the complex between INPP5B and biphenyl 3,3',4,4',5,5'-hexakisphosphate [BiPh(3,3',4,4',5,5')P6, IC50 5.5 µM] was determined at 2.89 Å resolution. One inhibitor pole locates in the phospholipid headgroup binding site and the second solvent-exposed ring binds to the His-Tag of another INPP5B molecule, while a molecule of inorganic phosphate is also present in the active site. Benzene 1,2,3-trisphosphate [Bz(1,2,3)P3] [one ring of BiPh(3,3',4,4',5,5')P6] inhibits INPP5B ca. 6-fold less potently. Co-crystallization with benzene 1,2,4,5-tetrakisphosphate [Bz(1,2,4,5)P4, IC50 = 6.3 µM] yielded a structure refined at 2.9 Å resolution. Conserved residues among the 5-phosphatase family mediate interactions with Bz(1,2,4,5)P4 and BiPh(3,3',4,4',5,5')P6 similar to those with the polar groups present in positions 1, 4, 5, and 6 on the inositol ring of the substrate. 5-Phosphatase specificity most likely resides in the variable zone located close to the 2- and 3-positions of the inositol ring, offering insights to inhibitor design. We propose that the inorganic phosphate present in the INPP5B-BiPh(3,3',4,4',5,5')P6 complex mimics the postcleavage substrate 5-phosphate released by INPP5B in the catalytic site, allowing elucidation of two new key features in the catalytic mechanism proposed for the family of phosphoinositide 5-phosphatases: first, the involvement of the conserved Arg-451 in the interaction with the 5-phosphate and second, identification of the water molecule that initiates 5-phosphate hydrolysis. Our model also has implications for the proposed "moving metal" mechanism.

Structure of the C-Terminal Domain of the Multifunctional ICP27 Protein from Herpes Simplex Virus 1.

UNLABELLED: Herpesviruses are nuclear-replicating viruses that have successfully evolved to evade the immune system of humans, establishing lifelong infections. ICP27 from herpes simplex virus is a multifunctional regulatory protein that is functionally conserved in all known human herpesviruses. It has the potential to interact with an array of cellular proteins, as well as intronless viral RNAs. ICP27 plays an essential role in viral transcription, nuclear export of intronless RNAs, translation of viral transcripts, and virion host shutoff function. It has also been implicated in several signaling pathways and the prevention of apoptosis. Although much is known about its central role in viral replication and infection, very little is known about the structure and mechanistic properties of ICP27 and its homologs. We present the first crystal structure of ICP27 C-terminal domain at a resolution of 2.0 Å. The structure reveals the C-terminal half of ICP27 to have a novel fold consisting of α-helices and long loops, along with a unique CHCC-type of zinc-binding motif. The two termini of this domain extend from the central core and hint to possibilities of making interactions. ICP27 essential domain is capable of forming self-dimers as seen in the structure, which is confirmed by analytical ultracentrifugation study. Preliminary in vitro phosphorylation assays reveal that this domain may be regulated by cellular kinases. IMPORTANCE: ICP27 is a key regulatory protein of the herpes simplex virus and has functional homologs in all known human herpesviruses. Understanding the structure of this protein is a step ahead in deciphering the mechanism by which the virus thrives. In this study, we present the first structure of the C-terminal domain of ICP27 and describe its novel features. We critically analyze the structure and compare our results to the information available form earlier studies. This structure can act as a guide in future experimental designs and can add to a better understanding of mechanism of ICP27, as well as that of its homologs.

VP22 core domain from Herpes simplex virus 1 reveals a surprising structural conservation in both the Alpha- and Gammaherpesvirinae subfamilies.

The viral tegument is a layer of proteins between the herpesvirus capsid and its outer envelope. According to phylogenetic studies, only a third of these proteins are conserved amongst the three subfamilies (Alpha-, Beta- and Gammaherpesvirinae) of the family Herpesviridae. Although some of these tegument proteins have been studied in more detail, the structure and function of the majority of them are still poorly characterized. VP22 from Herpes simplex virus 1 (subfamily Alphaherpesvirinae) is a highly interacting tegument protein that has been associated with tegument assembly. We have determined the crystal structure of the conserved core domain of VP22, which reveals an elongated dimer with several potential protein-protein interaction regions and a peptide-binding site. The structure provides us with the structural basics to understand the numerous functional mutagenesis studies of VP22 found in the literature. It also establishes an unexpected structural homology to the tegument protein ORF52 from Murid herpesvirus 68 (subfamily Gammaherpesvirinae). Homologues for both VP22 and ORF52 have been identified in their respective subfamilies. Although there is no obvious sequence overlap in the two subfamilies, this structural conservation provides compelling structural evidence for shared ancestry and functional conservation.

Structural insights into substrate recognition in proton-dependent oligopeptide transporters.

Short-chain peptides are transported across membranes through promiscuous proton-dependent oligopeptide transporters (POTs)--a subfamily of the major facilitator superfamily (MFS). The human POTs, PEPT1 and PEPT2, are also involved in the absorption of various drugs in the gut as well as transport to target cells. Here, we present a structure of an oligomeric POT transporter from Shewanella oneidensis (PepTSo2), which was crystallized in the inward open conformation in complex with the peptidomimetic alafosfalin. All ligand-binding residues are highly conserved and the structural insights presented here are therefore likely to also apply to human POTs.

The crystal structure of the DNA-binding domain of vIRF-1 from the oncogenic KSHV reveals a conserved fold for DNA binding and reinforces its role as a transcription factor.

Kaposi's sarcoma-associated herpesvirus encodes four viral homologues to cellular interferon regulatory factors (IRFs), where the most studied is vIRF-1. Even though vIRF-1 shows sequence homology to the N-terminal DNA-binding domain (DBD) of human IRFs, a specific role for this domain in vIRF-1's function has remained uncertain. To provide insights into the function of the vIRF-1 DBD, we have determined the crystal structure of it in complex with DNA and in its apo-form. Using a thermal stability shift assay (TSSA), we show that the vIRF-1 DBD binds DNA, whereas full-length vIRF-1 does not, suggesting a cis-acting regulatory mechanism in similarity to human IRFs. The complex structure of vIRF-1 DBD reveals interactions with the DNA backbone and the positioning of two arginines for specific recognition in the major grove. A superimposition with human IRF-3 reveals a similar positioning of the two specificity-determining arginines, and additional TSSAs indicate binding of vIRF-1 to an IRF-3 operator consensus sequence. The results from this study, therefore, provide support that vIRF-1 has evolved to bind DNA and plays a role in DNA binding in the context of transcriptional regulation and might act on some of the many operator sequences controlled by human IRF-3.

High-throughput analytical gel filtration screening of integral membrane proteins for structural studies.

BACKGROUND: Structural studies of integral membrane proteins (IMPs) are often hampered by difficulties in producing stable homogenous samples for crystallization. To overcome this hurdle it has become common practice to screen large numbers of target proteins to find suitable candidates for crystallization. For such an approach to be effective, an efficient screening strategy is imperative. To this end, strategies have been developed that involve the use of green fluorescent protein (GFP) fusion constructs. However, these approaches suffer from two drawbacks; proteins with a translocated C-terminus cannot be tested and scale-up from analytical to preparative purification is often non-trivial and may require re-cloning. METHODS: Here we present a screening approach that prioritizes IMP targets based on three criteria: expression level, detergent solubilization yield and homogeneity as determined by high-throughput small-scale immobilized metal affinity chromatography (IMAC) and automated size-exclusion chromatography (SEC). RESULTS: To validate the strategy, we screened 48 prokaryotic IMPs in two different vectors and two Escherichia coli strains. A set of 11 proteins passed all preset quality control checkpoints and was subjected to crystallization trials. Four of these crystallized directly in initial sparse matrix screens, highlighting the robustness of the strategy. CONCLUSIONS: We have developed a rapid and cost efficient screening strategy that can be used for all IMPs regardless of topology. The analytical steps have been designed to be a good mimic of preparative purification, which greatly facilitates scale-up. GENERAL SIGNIFICANCE: The screening approach presented here is intended and expected to help drive forward structural biology of membrane proteins.

Structural basis for PTPA interaction with the invariant C-terminal tail of PP2A.

Protein phosphatase 2A (PP2A) is a highly abundant heterotrimeric Ser/Thr phosphatase involved in the regulation of a variety of signaling pathways. The PP2A phosphatase activator (PTPA) is an ATP-dependent activation chaperone, which plays a key role in the biogenesis of active PP2A. The C-terminal tail of the catalytic subunit of PP2A is highly conserved and can undergo a number of posttranslational modifications that serve to regulate the function of PP2A. Here we have studied structurally the interaction of PTPA with the conserved C-terminal tail of the catalytic subunit carrying different posttranslational modifications. We have identified an additional interaction site for the invariant C-terminal tail of the catalytic subunit on PTPA, which can be modulated via posttranslational modifications. We show that phosphorylation of Tyr307(PP2A-C) or carboxymethylation of Leu309(PP2A-C) abrogates or diminishes binding of the C-terminal tail, whereas phosphorylation of Thr304(PP2A-C) is of no consequence. We suggest that the invariant C-terminal residues of the catalytic subunit can act as affinity enhancer for different PP2A interaction partners, including PTPA, and a different 'code' of posttranslational modifications can favour interactions to one subunit over others.

Deep learning based CETSA feature prediction cross multiple cell lines with latent space representation.

Mass spectrometry-coupled cellular thermal shift assay (MS-CETSA), a biophysical principle-based technique that measures the thermal stability of proteins at the proteome level inside the cell, has contributed significantly to the understanding of drug mechanisms of action and the dissection of protein interaction dynamics in different cellular states. One of the barriers to the wide applications of MS-CETSA is that MS-CETSA experiments must be performed on the specific cell lines of interest, which is typically time-consuming and costly in terms of labeling reagents and mass spectrometry time. In this study, we aim to predict CETSA features in various cell lines by introducing a computational framework called CycleDNN based on deep neural network technology. For a given set of n cell lines, CycleDNN comprises n auto-encoders. Each auto-encoder includes an encoder to convert CETSA features from one cell line into latent features in a latent space [Formula: see text]. It also features a decoder that transforms the latent features back into CETSA features for another cell line. In such a way, the proposed CycleDNN creates a cyclic prediction of CETSA features across different cell lines. The prediction loss, cycle-consistency loss, and latent space regularization loss are used to guide the model training. Experimental results on a public CETSA dataset demonstrate the effectiveness of our proposed approach. Furthermore, we confirm the validity of the predicted MS-CETSA data from our proposed CycleDNN through validation in protein-protein interaction prediction.