Write a comment

This review resulted from the graduate-level course "How to Read and Evaluate Scientific Papers and Preprints" from the University of São Paulo, which aimed to provide students with the opportunity to review scientific articles, develop critical and constructive discussions on the endless frontiers of knowledge, and understand the peer review process.

Paper reviewers: Ester Alves Russo, Luan Andrade Lisboa, Yasmin Alvarenga

‘Interferon-Induced PARP14-Mediated ADP-Ribosylation in p62 Bodies Requires an Active Ubiquitin-Proteasome System’

Rameez Raja, Banhi Biswas, Rachy Abraham, Hongrui Liu, Che-Yuan Chang, Hien Vu, Anthony K. L. Leung

This interesting and innovative paper highlights the formation and potential activity of interferon-induced p62 bodies enriched with PARP14 and ADP-ribose. These condensates are formed when the gene encoding PARP14 is transcriptionally activated through cell treatment with interferon gamma. The inhibition of either the cell's transcription machinery or PARP14’s enzymatic activity, alongside the expression of the selective autophagy marker p62, results in the absence of ADPr enrichment in p62 bodies. The condensates described by the authors differ from canonical p62 bodies, which are linked to the maturation factor of autophagosomes, LC3, whereas the interferon-induced bodies are not, establishing the newly identified structures as independent from selective autophagy. Another important feature of the findings is the demonstrated dependence on an active ubiquitin-proteasome system (UPS) for the formation of the PARP14-dependent condensates, revealing a novel relationship between PARP14 activity and the UPS.

The work presented connects the actions of enzymes in a never-before-seen way, expanding our knowledge and bridging the pathways in which they act, thus opening up discussions about the role of not only the newly discovered structures as a whole but also each of their components. The authors' findings pave the way for future research in several areas, such as identifying the ubiquitinated protein substrates that are targeted for proteasomal degradation upon IFNγ treatment, elucidating the mechanism through which the ubiquitin-proteasome system regulates PARP14-mediated ADP-ribosylation within p62 bodies, and determining the implications of these IFN-induced ADPr condensates in cellular responses, including their role in immunity, immunotherapy, and viral infection.

The main findings of the work are supported by a robust series of experiments, but there are several experiments that show repeated results or could be grouped together to provide the reader with a clearer understanding and avoid cluttering the panels in this preprint. Interferon-induced cytoplasmic condensates were identified through immunofluorescence assays after A549 cell treatment with interferons α, β, and γ, alongside a control group that was not treated with these cytokines, as shown in figure 1. The samples were stained with a Pan-ADPr antibody, which revealed the appearance of condensates after various treatment times. There is no need to show the appearance of the condensates twice, as was done in both panels 1A and 1B. The transcriptional control of condensate formation was shown through treatment of A549 cells with a transcription inhibitor followed by IFNγ treatment, with controls lacking the inhibitor. However, the following experiments simply quantify PARP and PARP14 expression in the cell culture line. PARP14 knockout (KO), inhibition, and its colocalization with the condensates, along with controls with wild-type cells and the absence of the inhibitor, are shown in different images - these could be more effectively merged into a single figure. The redundancy and lack of grouping in similar experiments make the page feel cluttered and difficult to read.

The effect of inhibitors for different PARPs in the formation of condensates and the conditions required for their formation were analyzed in figure 2. The authors utilized inhibitors for various PARPs and compared them to effects in cells without inhibitors and without IFNγ treatment. Panels 2B and 2D show, respectively, the concentration of the PARP14 inhibitor and the time required for the inhibitor to take effect. These could be better grouped with panels 2E and 2F, which show the expression of PARP14 and its mRNA in the experimental conditions. Figure 2E becomes redundant with the presence of figure 2G - a different storytelling strategy could be used to avoid showing results twice. Grouping previous images could improve the spacing in figure 2H to more effectively display the immunoprecipitation assay in this series of experiments.

Figure 3 investigates the essential role of p62 in the formation of PARP14 ADPr condensates. Panel 3A establishes the colocalization of p62, showing that only p62 colocalizes with ADPr condensates after IFNγ treatment among the tested cellular structures. Panel 3B visually and quantitatively complements this, confirming the presence of p62, MAR, and PARP14 in these IFNγ-induced structures compared to untreated cells. It is already established in the literature that PARP14 is a mono ADP-ribosyltransferase, and in this study, it is shown that PARP14 not only colocalizes with MAR but is also ADP-ribosylated, making the panel unnecessary. Panel 3C introduces the dynamics of p62 bodies with IFNγ treatment, suggesting modulation of their composition and organization. Panel 3D provides evidence for the prevention of p62 condensate formation upon comparison between p62 knockdown (KD) cells and wild-type (WT) cells. Panel 3E complements this by showing that the prevention of condensate formation is not due to lower levels of PARP14, emphasizing the role of p62 in the formation of these cytoplasmic bodies. Figure 3 presents a well-organized series of panels that collectively explain the composition and dynamics of interferon-induced p62 bodies, supported by concepts already established in the literature.

Figure 4 demonstrates that IFNγ treatment increases the association between PARP14 and p62, with p62 identified as a substrate for PARP14-mediated MARylation, a modification that intensifies after IFNγ treatment. The figure shows that the catalytic activity of PARP14 is essential for the co-condensation of ADPr and PARP14 with p62 within the p62 bodies induced by IFNγ, as the inhibition or absence of PARP14 prevents this process. FRAP analyses suggest that PARP14 activity modulates the dynamics of p62 within these structures. Finally, the expression of a PARP14 mutant without hydrolase activity demonstrates that increased levels of PARP14-mediated ADP-ribosylation are sufficient to induce the co-condensation of PARP14, ADPr, and p62 in p62 bodies, reinforcing the central role of PARP14's transferase activity in this process. Figure 4 uses different techniques to complement each other without redundancy, establishing the relationship between p62 and PARP14 and highlighting the importance of PARP14’s mono ADP transferase activity in the formation of the cytoplasmic condensates.

Figure 5 aims to characterize the ADPr-enriched p62 bodies formed due to IFNγ treatment and investigate the role of PARP14 in related signaling pathways. Initially, panel A presents a scheme of the domain structure of p62 to explain its diverse molecular functions. Panels B, C, and D then explore the involvement of PARP14 and p62 in the STAT1, NRF2, and NF-κB signaling pathways. The results suggest that the activation of these pathways does not depend on the presence or activity of PARP14 or p62 under the tested conditions. The figure then investigates the composition of these IFN-induced p62 bodies. Panel E demonstrates that these bodies do not colocalize with the autophagosome marker LC3B, indicating that autophagy may not be the primary degradation pathway involved. On the other hand, panels F, G, and H reveal that these condensates are enriched in ubiquitinated proteins, including K48 and K63 linkages, a feature shared with canonical p62 bodies. Figure 5 establishes that IFNγ-induced ADPr-enriched p62 bodies contain ubiquitin and NBR1 but not the autophagy marker LC3B, suggesting a distinct composition from canonical p62 bodies in the context of the IFNγ response. The figure is well-organized and contains important experimental data presented in small, digestible panels.

Figure 6 investigates the ubiquitin-proteasome system in the formation of IFNγ-induced ADPr-enriched p62 bodies. Initially, the results in panel A reveal that inhibition of autophagy does not prevent the formation of these condensates, suggesting that this degradation pathway is not connected to their regulation. On the other hand, inhibition of proteasome activity using different inhibitors in panels B, C, and E consistently demonstrates the elimination of ADPr condensate signals and a significant reduction of PARP14 in p62 bodies. Although panel D shows a slight reduction in PARP14 and p62 protein levels with proteasome inhibition, panel G reveals that inhibition of ubiquitin activation with TAK-243 also leads to the disappearance of ADPr condensates without changing the cellular levels of these proteins. These findings indicate that an active ubiquitin-proteasome system is required for the condensation of PARP14 and ADPr into p62 bodies, as summarized in the working model presented in panel H.

As this is a preprint on a relatively new topic, subject to changes. However, on parts, the content is outdated. Several topics presented treat concepts as if they were new to the literature or use outdated terminology. The "condensates" mentioned were recently described and named 'ICABs - interferon-induced cytosolic ADPr bodies' (RIBEIRO, 2025). Although the manuscript includes a series of experiments that may justify the existence of these condensates, there are weak points that undermine the impact of the presented data. A more detailed analysis of the structures, both from a morphological and quantitative perspective, could strengthen conclusions and increase the overall impact of the work. The results indeed include microscopy data that could be relevant for observing phenomena in these cytosolic bodies, but many graphs show non-significant statistics, as seen in figure 4B, indicating low reliability or statistical power of these experiments. Presenting these graphs with non-significant results in the manner discussed could mislead readers into thinking there is an important relationship, when in reality, there is no solid evidence to support such a conclusion. 

At the start of the results section, the authors clarify that only 1 hour of exposure to IFN-γ is required to generate ICABs. However, it is noteworthy that there is a lack of investigation into the presence of other proteins in these cytosolic bodies. It has already been shown that other proteins, such as the PARP9/DTX3L complex and the recruitment of LC3 by P62 (RIBEIRO, 2019), can colocalize with PARP14, one of the main proteins in this preprint. Therefore, further investigation into other proteins within this condensate is needed to clarify its functionality and composition. This uncertainty could potentially be addressed by evaluating the colocalization of proteins in the cytosolic bodies and/or assessing protein inhibition to better understand their role.

Additionally, RBN012579 was used as an inhibitor of PARP14, and the results presented were statistically non-significant, raising doubts about the veracity and quality of the data. Furthermore, the preprint states that ADPr is not dependent on PARP14, whereas the literature suggests the opposite (DUCIK, 2023; RIBEIRO, 2019).

Lysines 63 and 48 were highlighted in the results, and it is reasonable to suppose that they are associated with ADPr via serine linkages. However, there is evidence suggesting that lysine 11 might also be involved with ADPr, especially in ester linkages with aspartate/glutamate (BEJAN, 2025). Although we understand that this information was published after the preprint, it would be valuable to investigate other lysines related to this underexplored area of ADPr.

In Figure 5, while the graphs show low reliability, the authors employed interesting strategies to evaluate the process of IFN-γ induction by measuring STAT-1 activation. In this context, IFNs bind to specific membrane receptors and activate kinases from the JAK family (Janus kinases), which then phosphorylate transcription factors from the STAT family (Signal Transducer and Activator of Transcription). This process leads to the translocation of STATs into the cell nucleus, resulting in the expression of a range of interferon-stimulated genes, among which PARPs are included (SCHOGGINS, 2019).

Although a series of experiments were conducted, we believe testing these experiments across different cell lines could yield more robust results on the existence of this new structure.

The conclusion text could have been more organized. While the colocalization of PARP9/DTX3L was mentioned, the interaction of these components with PARP14 in the cytosolic bodies was not shown. Had this interaction been demonstrated, it could have added more robustness to the results.

However, the idea of detailing the methods suggests that the experiments were conducted with a high level of care and precision, which is essential for ensuring the robustness of the results obtained. A well-structured methodology, with a clear and logical sequence, is crucial for the reproducibility of experiments, allowing other researchers to follow the same protocol and obtain consistent results. The meticulous presentation of the methods not only reinforces the credibility of the study but also provides a solid foundation for future investigations. Clarity in describing each experimental step, from reagent preparation to analysis methods, makes the work more transparent and accessible to other scientists, which is essential for the advancement of scientific research.

Furthermore, the inclusion of the list of antibodies and primers is particularly valuable, as it provides a level of transparency that allows for the verification of the tools used in the study. This type of information makes it easier for other research groups to replicate the experiments and ensures the validity of the results, as the diagnostic tools and reagents are clearly specified. An organized and sequential methodology contributes to building a robust study, where each phase of the experiment is properly justified and can be validated by both reviewers and other researchers wishing to replicate the findings. This strengthens the confidence in the validity of the data and, ultimately, contributes to the credibility and impact of the scientific work.

Competing interests

The authors declare that they have no competing interests.