Say What? E1 and E2 not observed in Ubiquitination

Tyler Florio’s Blog Spotlight – Ubiquitination independent of E1 and E2 enzymes by bacterial effectors
Link: http://www.nature.com.muhlenberg.idm.oclc.org/nature/journal/vaop/ncurrent/full/nature17657.html

Ubiquitination is a vital cellular process universally observed in eukaryotes and is responsible for tagging protein substrates for various cellular functions such as degradation, modulation, transport, etc. This process is named after the 76 amino acid polypeptide called ubiquitin which is attached to designated protein substrates in three steps (activation, conjugation, and ligation respectively) via three specific enzymes (E1, E2, and E3 respectively). It has been demonstrated that the ubiquitination pathway is responsible for regulating vital cellular processes such as immune response.

Although bacteria does not utilize the ubiquitination pathway, many infectious bacteria species such as E. coli and Salmonella exploit ubiquitination’s role in immune response by developing toxic “ubiquitin-like” proteins called effectors. These effectors can function as either E3-ligases, deubiquitinases, and even attack ubiquitin attached to protein substrates which allows the bacteria to suppress the host’s immune system and allow it to survive. The classical method of ubiquitination involves the systematic cascade of E1, E2, and E3 to attach ubiquitin to its target; however, a new study by Qiu et al (doi: 10.1038/nature17657) demonstrated that the infectious bacteria L. pneumophila express an effector protein SdeA which performs a substrate ubiquitination independent of E1 and E2 enzymes.

The authors were originally investigating the SidE effector protein family (SdeA, SdeB, SdeC, and SdeD) for L. pneumophila which are the virulent effectors and discovered a conserved mono-ADP ribosyltransferase (mART) domain through protein-sequence analysis. Upon further investigation, the authors determined that mART domain in critical for SdeA-mediated toxicity, but surprisingly did not demonstrate ADP-ribosylation activity. The lack of ADP-ribosylation activity suggests that a different biochemical mechanism than previously thought was occurring

SdeA induces a posttranslational modification on multiple ER-associated Rab proteins.
A) An SDS-PAGE experiment with incubated Sea with Rab proteins. This demonstrates molecular mass shifts for Rab1 and Rab33b B) Immunoblotting with antibodies for flagged-Rab33b. This shows a clear shift in molecular mass for flagged-Rab33b with SdeA and the mART domain importance. C-F) SdeA induces Rab33b ubiquitination with SDS-PAGE and mass spectrometry analysis. G) Overexpression of Rab33b restricts intracellular bacterial growth.

Perplexed, the authors performed a co-expression of WT and mART-mutated SdeA with ER-associated Rab proteins Rab1, Rab6A, Rab30, and Rab33b (known L. pneumophila target) and discovered Rab1 and Rab33b experienced a molecular mass shift with WT SdeA and not the mutant. Through mass spectrometry analysis, the authors determined that altered Rab1 and Rab33b had ubiquitin polypeptides attached to them while the unaltered Rab1 and Rab33b did not which suggests that SdeA performs a ubiquitination reaction with Rab1 and Rab33b. Then when the authors attempted an in vitro ubiquitination with WT SdeA and Rab33b in the presence of E1, ATP, and several E2’s and found Rab33b had not been ubiquitinated; however, the addition of cofactor NAD resulted in ubiquitination activity (Fig 1). The lack of ubiquitinated Rab33b protein with E1, ATP, and several E2’s suggests that the classical ubiquitination pathway does not apply to SdeA. This is the first reported case where a protein substrate is ubiquitinated independent of E1 and E2 dependence and suggests a currently unknown mechanism is occuring with SdeA which results in ubiquitination. This opens the door to determine if these effectors have other unknown substrates and if there exists other proteins which ubiquitinate independent of E1 and E2. Future investigations into these E1-E2 independent ubiquitinases with mART domains can expand our understanding of the cellular processes regulated by ubiquitination.

Reference:  Qiu, Jiazhang, Michael J. Sheedlo, Kaiwen Yu, Yunhao Tan, Ernesto S. Nakayasu, Chittaranjan Das, Xiaoyun Liu, and Zhao-Qing Luo. “Ubiquitination Independent of E1 and E2 Enzymes by Bacterial Effectors.” Nature, 2016. doi:10.1038/nature17657.

 

7 Replies to “Say What? E1 and E2 not observed in Ubiquitination”

  1. Very interesting article as it challenges the canonical idea of ubiquitinization involving E1, E2 & E3. It appears that the bacterial cell used NAD and Ub from the host cell in order work. So, in eukaryotic cells it works as a NAD dependent ubiquitinase. Is this is only mechanism of action or does it have a role such as ADP-ribosylation or conjugation of other ubiquitin-like molecules towards other pathogenic bacteria?

  2. Great article. It is very interesting to see ubiquitination without the classic uses of E1 and E2 that we have seen in the past. I am curious about the ubiquitination processes of Rab1 and Rab33b that occurs independently of E1 and E2 but with the addition of NAD. You mentioned that the exact mechanism is unknown but have there been any studies that show how NAD influences the specific Rab proteins ubiquitination? If so, have any shed light on possible reasons for why these specific Rab proteins are caused to be ubiquitinated in a different way than the other Rab proteins?

  3. Really interesting study with really intimidating gels! I am wondering about the canonical function of the mART domain and the mechanism by which it catalyzes ADP-ribosylation. Maybe this could provide some insight into how these motifs are also able to perform ubiquitination. Great summary of the paper.

  4. This is crazy! Great article Tyler. Your explanation was very clear and helpful in understanding the authors’ approaches in the paper. One thing I’m curious about is their results from the experiment with the different Rab proteins when co-expressing with WT and mART-mutant SdeA. The authors noticed Rab1 and Rab33b were affected, but why these Rab proteins as opposed to Rab6A and Rab30? It’s understandable that their lack of results with Rab6A and Rab30 caused them to not continue with them in their investigation, but did they ever find out why these Rab proteins are affected differently by the SdeA?

  5. Great job on finding a really cool paper, and your blog post really helped to clarify a lot! Did you get the sense that these authors will continue with their research to determine a mechanism for which SdeA results in ubiquitination? I’m also curious about how determining this new mechanism would open doors for drug therapies to combat infection in humans. If there was a way to block SdeA from suppressing immune response via this pathway, then perhaps new treatments could be developed. Did you come across any mention of the benefits involved in determining this new mechanism in regards to human health throughout your research process?

  6. Great Summary! It was really interesting that this alternate mechanism for ubiquitination exists, but I was a little confused. You state that the mechanism for ubiquitination with SdeA is independent of E1 or E2 but from the sentence ” Then when the authors attempted an in vitro ubiquitination with WT SdeA and Rab33b in the presence of E1, ATP, and several E2’s and found Rab33b had not been ubiquitinated; however, the addition of cofactor NAD resulted in ubiquitination activity” it suggests that the mechanism is still dependent on E1 and E2 but also requires NAD. Did they perform experiments in which they excluded E1 and/or E2’s to demonstrate the mechanisms independence of these enzymes?

  7. Very cool, and great summary! Bacterial pathogens are known to interfere with eukaryotic host signaling through a variety of mechanisms, including secreted DUBs, and while this is a new discovery, it is not so surprising ^_^. Good job bacteria, they strike again!
    Zachary’s question seems to be answered by revisiting Figure 4…but my question is curiosity about the prevalence of these mART domains across other bacteria and perhaps even within humans. Human PARPs have been studied for years, with some family members having no demonstrated ADP-ribosylating activity…I wonder if they are also ubiquitin-conjugators in disguise!

Comments are closed.