Inflammation can help to eliminate infection, but excessive inflammation can cause damage to the body. The sensor proteins that trigger an inflammatory immune response must therefore be carefully regulated. Some intracellular immune-sensor proteins detect components in a cell that become abnormal or altered during a cellular crisis. Signs of cellular crisis are sometimes produced in the absence of an infection, so mechanisms are needed to prevent the proteins from triggering an inappropriate inflammatory response. Writing in Nature, Sharif et al.1 report a structural study that investigates an immune-sensor protein called NLRP3, revealing that a protein called NEK7 acts as a ‘licence’ that enables this protein to cause inflammation.
When an immune sensor recognizes a hallmark of infection in the cytoplasm, this can activate the protein and lead to the assembly of a multiprotein complex called an inflammasome. The activation of proteins that function downstream of an inflammasome can potently drive both inflammation and cell death2. Different types of inflammasome can form depending on the sensor components involved. Certain inflammasomes respond to a highly specific trigger: for example, those in mammalian cells containing the sensor protein NLRC4 respond to the presence of the bacterial protein flagellin3,4.
Proteins that are normally present in mammalian cells do not seem able to trigger the accidental formation of NLRC4-containing inflammasomes, given the lack of reports of such aberrant events. By contrast, inflammasomes that contain NLRP3 are activated when NLRP3 recognizes — by an as yet unknown mechanism — hallmarks of cellular catastrophe, such as extremely low concentrations of potassium in the cytoplasm, or signs of dysfunction in organelles called mitochondria2. Such events can arise from tissue damage that is unrelated to infection, and NLRP3 activation in such cases has been implicated as a possible cause of inflammatory diseases such as atherosclerosis.
It is widely accepted that the tightly regulated formation of NLRP3-containing inflammasomes occurs in two steps. In the first step, NLRP3 is primed for action by other immune-sensor proteins called TLRs, which can detect components of microorganisms. This priming step occurs in two ways2: NLRP3 can undergo a modification, such as the addition of a phosphate group or the removal of an attached ubiquitin protein. Further priming is achieved by a rise in expression of the gene that encodes NLRP3, increasing the chance that NLRP3 will detect any abnormalities. The second step, activation, then results in NLRP3 proteins binding together to form part of a disc-shaped inflammasome complex that is probably similar to those of other inflammasomes containing proteins of the NLR family (which includes NLRP3 and NLRC4)5,6. This activation step occurs during a cellular catastrophe, but the biochemical and structural mechanisms involved are unknown.
Researchers have long sought to determine the structure of NLRP3 as it forms an inflammasome, in the hope of gaining insights into how this protein functions. However, such efforts have been unsuccessful, perhaps because unknown protein partners that interact with NLRP3 were missing from earlier attempts.
The discovery7–9 that the enzyme NEK7 is essential for NLRP3 signalling provided a missing part of the puzzle. NEK7 regulates processes that occur during cell division, such as the breakdown of the nuclear-envelope structure10, so it was surprising to find that it has a separate role in inflammation. This suggested that NLRP3-containing inflammasome formation doesn’t occur during cell division because NEK7 is unavailable to aid inflammasome assembly8. NEK7 regulates NLRP3 signalling by binding to a region of the protein known as the LRR domain8,9. However, why such an interaction between NEK7 and NLRP3 is essential for inflammasome formation has remained elusive.
To tackle this question, Sharif et al. used cryo-electron microscopy to investigate the structure of a human NLRP3 interacting with NEK7. The authors’ structural data reveal that NLRP3 and NEK7 bind to form a dimer in which NLRP3 is in an inactive conformation. In this state, the LRR domain of NLRP3 has a lobed, semicircular shape (Fig. 1), and the carboxy-terminal region of NEK7 nestles in the inner curve of the LRR.
NLRP3 also contains a structure called the NACHT domain, and in the inactive NLRP3–NEK7 complex, this domain is structurally very similar to the NACHT domain11 of inactive NLRC4. It was previously shown5,6,11 that the NACHT domain of NLRC4 rotates as it transitions into an active conformation. This rotational-activation step uncovers part of the NACHT surface, enabling inflammasome formation through a protein-assembly process called oligomerization, and generating a disc-shaped NLRC4-containing inflammasome5,6. The authors used this information to model a hypothetical conformation for an NLRP3-containing inflammasome.
In Sharif and colleagues’ model, the hypothetical rotational activation of NLRP3 doesn’t affect NEK7 binding, and NEK7 still fits into NLRP3’s LRR domain in the same way as in the inactive structure. Furthermore, the authors made the surprising discovery that NEK7 provides a bridge between adjacent NLRP3 proteins, by forming an interface with the LRR of the adjacent NLRP3 in the inflammasome. The ability to form such an interface suggests that NEK7 provides a licence for NLRP3 to form part of the inflammasome. It seems that this licensing event occurs independently of both the priming and rotational-activation steps, because the authors did not include molecules that cause priming or add triggers for rotational activation in their structural studies. The results suggest a revised view of how the NLRP3-containing inflammasome is regulated, and put forward the idea that NLRP3 oligomerization requires NEK7 licensing. Taken together with evidence from earlier studies7–9, it seems likely that this licence is revoked during cell division.
The function of NEK7 in the NLRP3-containing inflammasome is interesting when considered in relation to the structure of the NLRC4-containing inflammasome. The LRR domain of NLRC4 is longer than that of NLRP3, and makes direct contact with the LRR of the adjacent NLRC4 protein in the inflammasome5,6. In an NLRP3-containing inflammasome, NEK7 fulfils a similar connecting role by making contact with adjacent LRR domains. This explains why NLRP3-containing inflammasomes require NEK7 licensing, whereas NLRC4-containing ones do not.
Many mysteries concerning the regulation of NLRP3-containing inflammasomes remain. Perhaps future structural studies will reveal how NLRP3 modification accomplishes the priming step.
Finally, we still don’t know the answer to perhaps the most important question of all: what direct interaction between NLRP3 and an unknown cellular factor results in the formation of the inflammasome? Perhaps the evidence that the NEK7 licence is revoked during cell division provides a clue. If inappropriate activation of NLRP3 is likely to occur during cell division, then having an NEK7-licensing step would help to combat this potential problem. Thus, one could imagine that the type of cellular catastrophe detected by NLRP3 also occurs during cell division but in a controlled manner.
Nature 570, 316-317 (2019)