NLRP7: From inflammasome regulation to human disease

Abstract

Nucleotide‐binding oligomerization domain (NOD) and leucine‐rich repeat (LRR)‐containing receptors or NOD‐like receptors (NLRs) are cytosolic pattern recognition receptors, which sense conserved microbial patterns and host‐derived danger signals to elicit innate immune responses. The activation of several prototypic NLRs, including NLR and pyrin domain (PYD) containing (NLRP) 1, NLRP3 and NLR and caspase recruitment domain (CARD) containing (NLRC) 4, results in the assembly of inflammasomes, which are large, cytoplasmic multiprotein signalling platforms responsible for the maturation and release of the pro‐inflammatory cytokines IL‐1β and IL‐18, and for the induction of a specialized form of inflammatory cell death called pyroptosis. However, the function of other members of the NLR family, including NLRP7, are less well understood. NLRP7 has been linked to innate immune signalling, but its precise role is still controversial as it has been shown to positively and negatively affect inflammasome responses. Inflammasomes are essential for homeostasis and host defence, but inappropriate inflammasome responses due to hereditary mutations and somatic mosaicism in inflammasome components and defective regulation have been linked to a broad spectrum of human diseases. A compelling connection between NLRP7 mutations and reproductive diseases, and in particular molar pregnancy, has been established. However, the molecular mechanisms by which NLRP7 mutations contribute to reproductive diseases are largely unknown. In this review, we focus on NLRP7 and discuss the current evidence of its role in inflammasome regulation and its implication in human reproductive diseases.

Abbreviations

AIM2

absent in melanoma 2

ALRs

AIM2‐like receptors

ASC

apoptosis‐associated speck‐like protein containing a CARD

BBox

B‐box‐type zinc finger domain

B30·2

B30·2 and SPRY (SPla and the RYanodine Receptor)

BIR

baculovirus inhibitor of apoptosis repeat

CAPS

cryopyrin‐associated periodic syndrome

CARD

caspase recruitment domain

CC

coiled‐coil

CHM

complete HM

COP

CARD‐only protein

DAMPs

danger‐associated molecular patterns

FGR

fetal growth restriction

FIIND

function to find domain

FRHM

familial recurrent HM

gDMR

germline differentially methylated region

GSDMD

gasdermin D

HM

hydatidiform mole

IFI16

interferon gamma‐inducible protein 16

IL

interleukin

KSHV

Kaposi sarcoma‐associated herpesvirus

LPS

lipopolysaccharide

LRR

leucine‐rich repeat

MAMPs

microbial‐associated molecular patterns

MDP

muramyl dipeptide

mtDNA

mitochondrial DNA

NAD

NACHT‐associated domain

NBD

nucleotide‐binding domain

NF‐κB

nuclear factor kappa B

NLR

NOD‐like, LRR‐containing receptor

NLRA

NLR family acidic domain containing

NLRB

NLR family BIR domain containing

NLRC

NLR family CARD containing

NLRP

NLR family PYD containing

NOD

nucleotide‐binding oligomerization domain

PAMPs

pathogen‐associated molecular patterns

PHM

partial HM

POP

PYD‐only protein

PRR

pattern recognition receptors

PYD

pyrin domain

ROS

reactive oxygen species

STAM

signal‐transducing adaptor molecule

STAMBP

STAM‐binding protein

STAND

signal‐transducing ATPase with numerous domains

TLR

Toll‐like receptor

TNF

tumour necrosis factor

TTSS

type III secretion system

UC

ulcerative colitis

YY1

ying yang 1

ZBTB16

zinc finger and BTB domain containing 16

INTRODUCTION

The innate immune system maintains homeostasis and responds to tissue damage and pathogen infections [1]. The presence of microbes and their microbial‐associated molecular patterns (MAMPs) and pathogen‐associated molecular patterns (PAMPs) or host‐derived danger signals or danger‐associated molecular patterns (DAMPs) is sensed by a wide variety of specialized pattern recognition receptors (PRRs) expressed on the cell surface or in the cytoplasm of phagocytes and other cells [2]. Nucleotide‐binding oligomerization domain (NOD)‐ and leucine‐rich repeat (LRR)‐containing receptors (NLRs) represent the largest cytosolic PRR family, which together with absent in melanoma 2 (AIM2)‐like receptors (ALRs), and Pyrin, are involved in inflammasome assembly and activation [3]. The NLR gene superfamily encodes 22 human and 34 mouse proteins, which are characterized by an evolutionarily conserved tripartite domain architecture with (i) an N‐terminal effector domain, (ii) a central nucleotide‐binding domain (NBD) – comprised of the NACHT domain (NAIP, CIITA, HET‐E and TP1) and NACHT‐associated domain (NAD) – and (iii) a C‐terminal autoregulatory LRR domain [4]. The five subfamilies are distinguished by the nature of the N‐terminal effector domain, containing either an acidic domain (NLRA), a baculovirus inhibitor of apoptosis repeat (BIR) domain (NLRB), a CARD domain (NLRC), a pyrin domain (PYD; NLRP) or an N‐terminal domain with no homology with other NLR members (NLRX) [4]. NLR activation triggers multiple innate immune signalling pathways to eradicate infections, promote wound healing and maintain tissue homeostasis, including MHC class I and II gene transcription, regulation of nuclear factor kappa B (NF‐κB), mitogen‐activated protein kinase, Akt, type I interferons, autophagy and developmental processes [5]. While most NLRs trigger a transcriptional response upon ligand sensing, several are capable of promoting the assembly of large multiprotein signalling platforms called inflammasomes (Figure 1).

FIGURE 1.

Inflammasome assembly. The activation of inflammasomes requires a ‘priming’ step resulting in activation of the transcription factor NF‐κB and, consequently, the expression of pro‐IL‐1β and pro‐IL‐18 and inflammasome components. In addition, inflammasome‐activating PRRs undergo various post‐translational modifications to confer a ready state, as well as the subcellular redistribution of inflammasome components. Inflammasome activation is initiated upon ligand sensing, resulting in the release of the inactive conformation of the inflammasome sensors and sensor oligomerization. The active, oligomerized PRR recruits ASC usually through PYD‐PYD interactions, which nucleates ASC polymerization and in turn caspase‐1 polymerization through CARD‐CARD interactions resulting in proximity‐induced caspase‐1 activation, while NLRC4 can also directly recruit caspase‐1. Active caspase‐1 proteolytically matures pro‐IL‐1β and pro‐IL‐18 into their bioactive form, which are then released. Caspase‐1 also proteolytically cleaves GSDMD, which polymerizes in membranes to generate pores resulting in pyroptosis

Inflammasomes are responsible for the recruitment and proximity‐mediated activation of the protease caspase‐1, which is required for proteolytic maturation of the pro‐inflammatory cytokines interleukin (IL)‐1β and IL‐18 [3, 6]. Upon recognition of their respective ligands and/or activating signals, NLR sensors oligomerize and subsequently recruit the adaptor protein apoptosis‐associated speck‐like protein containing a CARD (ASC) through homotypic PYD‐PYD or CARD‐CARD interactions, depending on the effector domain of the NLR sensor. Inflammasome assembly progresses with the self‐perpetuating polymerization of ASC prion‐like filaments, and the exposed CARD of ASC then recruits caspase‐1 through CARD‐CARD interactions, resulting in clustering and proximity‐mediated activation of caspase‐1 [7]. In macrophages, inflammasome activation also induces an inflammatory form of cell death called pyroptosis through the caspase‐1‐mediated proteolytic cleavage of gasdermin D (GSDMD), which removes its inhibitory C‐terminal domain and allows the N‐terminal fragment to bind to phosphoinositides and cardiolipin, insert into membranes, oligomerize and form a lytic pore [8, 9, 10]. Noticeably, inflammasome assembly can be fine‐tuned by small endogenous PYD‐only proteins (POPs) and CARD‐only proteins (COPs), which are incorporated into these structures by interfering with the PYD‐PYD and CARD‐CARD interactions [11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22].

To date, thirteen inflammasomes have been identified (Table 1) [23]. NLRP1 was the first NLR sensor identified to form an inflammasome and together with NLRC4, Pyrin, AIM2 and NLRP3 is among the best studied inflammasomes [24, 25, 26, 27, 28, 29]. NLRP3 has initially been identified as the causative gene for a group of autoinflammatory diseases referred to as Cryopyrin‐associated periodic syndrome (CAPS) and has a unique position by sensing not only infections, but primarily sterile tissue damage, and its activation is now well defined and requires a two‐step mechanism (Figure 1) [28, 30]. The first signal (‘priming’) induces the transcription and various post‐translational modifications of NLRP3 to prepare the sensor for activation while remaining inactive. The second and activating signal is unique for NLRP3, as it senses a diverse spectrum of signals culminating in an altered intracellular milieu caused by (i) changes in the ionic balance (potassium efflux, calcium influx), (ii) lysosomal and mitochondrial damage, (iii) pore forming toxins, cell death‐induced pores and other cellular stress signals, and (iv) changes in the subcellular localization of NLRP3 inflammasome components [28]. In addition to these established inflammasome sensors, several other proteins have been recognized as inflammasome sensors, but their mechanisms are less well understood, including interferon gamma‐inducible protein 16 (IFI16), NLRP2, NLRP6, NLRP9b, NLRP12, NLRC5, CARD8 and NLRP7(Table 1).

TABLE 1.

Inflammasomes and their activators. The protein structure and the ligands, NCBI Accession numbers and activating signals for known inflammasomes are shown

Sensor	Protein domains	Accession #	Activators/PAMPs/DAMPs
NLRP1		(NP_127497.1)	Bacillus anthracis lethal toxin [117, 118, 119, 120] K+ efflux [121] MDP [118, 119] dsRNA [122]
NLRP2		(NP_001167552.1)	ATP [111, 112]
NLRP3		(NP_004886.3)	Pathogen mRNA [118, 123] Extracellular ATP [124, 125, 126] Crystals [b‐amyloid [127], asbestos [128], silica [128, 129], MSU [130], cholesterol [131] mtDNA [132, 133] cAMP [134] K+ efflux/Ca2+ influx [121, 135, 136] Lysosomal proteases [129] ROS [137] Ceramide [138] Cardiolipin [139] Aluminium salts [140]
NLRP6		(NP_612202.2)	Lipoteichoic acid [65]
NLRP7		(NP_631915.2)	Gram‐positive bacteria [39] Mycoplasma spp. [39] Bacterial acylated lipopeptides [39] Mycobacterium bovis [63] K+ efflux? Lysosomal proteases?
Nlrp9b		(NP_918947.2)	short double‐stranded RNA [141]
NLRP12		(NP_001264055.1)	Acylated lipid A [142]
NLRC4		(NP_067032.3)	Flagellin [143, 144, 145, 146, 147] Bacterial TTSS [148] Gram‐negative bacteria [147, 148, 149, 150] K+ efflux/Ca2+ influx [151] Lysosomal proteases [152]
NLRC5		(NP_001371879.1)	Unknown
CARD8		(NP_001171829.1)	Inhibition of dipeptidyl‐peptidases [153, 154]
AIM2		(NP_004824.1)	cytosolic dsDNA [155, 156, 157, 158]
IFI16		(NP_001193496.1)	KSHV DNA [159]
Pyrin		(NP_000234.1)	Loss of Rho GTPase‐mediated PKN1/2 phosphorylation [160]

Advances have been made to better understand the role of inflammasomes during pathogen infection and sterile injury, and there is compelling evidence for their central role in maintaining tissue homeostasis and human health. Hence, a tight control and proper regulation of inflammasome activation is crucial [11, 31]. The malfunction, dysregulation or the presence of mutations or allelic variations in genes encoding inflammasome sensors, including NLRP1, NLRP3, NLRP7, NLRP12, NLRC4, AIM2, CARD8 and Pyrin, results in either gain or loss of function, and is associated with various pathologies [31, 32, 33, 34, 35]. In this review, we focus on one of the less well‐characterized NLRPs, NLRP7. We will discuss the controversial findings regarding its role in inflammasome signalling, and we will give an update on the current knowledge on its implication in human reproductive diseases.

IDENTIFICATION AND ORIGIN OF NLRP7

NLRP7 is located on the long arm of chromosome 19q13·4, and its first link to disease was recognized in 1999 when an NLRP7 variant was identified in a woman with a history of recurrent molar pregnancy [36]. Taxonomic analyses suggest that the human NLRP7 gene may have originally evolved in Simiiformes (i.e. Simians) nearly 43 million years ago and then evolved in Catarrhini, Hominoidea, Hominidae, Homininae and finally Homo sapiens. Further, phylogenetic analyses revealed that NLRP genes are conserved and that all NLRPs are monophyletic, except for NLRP2, which encompasses the primate‐specific NLRP7, suggesting a gene duplication of the NLRP2/7 ancestor in primates [37]. Significantly, this study also revealed a clade of NLRPs that are related to the reproductive system, including NLRP2, NLRP4, NLRP5, NLRP7, NLRP8, NLRP9, NLRP11, NLRP13 and NLRP14, which are all expressed in human oocytes and embryos [38]. Like all NLRP family members, NLRP7 is composed of an N‐terminal PYD, a central NACHT domain followed by a NAD and several C‐terminal LRRs (Table 1; Figure 2). Multiple isoforms have been described and predicted with varying LRR numbers due to alternative splicing (Figure 2). Whether the number of LRRs or the amino acid composition of these additional or lacking domains have consequences on NLRP7 function/pathway is a question largely remaining to be answered. However, the analysis of two NLRP7 isoforms with 5 and 7 LRRs revealed a higher pro‐inflammatory activity of the longer isoform, suggesting a potential functional consequence based on the number of LRRs [39]. Functional consequences of alternative splicing of LRR‐encoding exons have also been demonstrated for human NLRP3, where exons 5 and 7 are frequently skipped in human macrophages. The variant lacking exon 5 is inactive due to the absence of specific surface amino acids present in the missing LRRs, which impairs the essential interaction with NEK7 [40, 41].

FIGURE 2.

NLRP7 structure and splice forms. NLRP7 exon organization of the 3 main domains: PYD, NACHT and LRRs (predicted by InterPro, https://www.ebi.ac.uk/interpro/) [113] and the main 3 annotated isoforms present in the NCBI protein database are indicated with their Accession numbers

NLRP7 AND INFLAMMASOME SIGNALLING

NLRP7: a negative regulator of inflammasome signalling

NLRP7, also known as NALP7, PAN7, NOD12, CLR19·4, PYPAF3 and HYDM, has the necessary domain architecture to function as an inflammasome sensor. However, in 2002, a functional analysis based on the overexpression of several PYPAF family members, including NLRP7 (PYPAF3) in non‐immune HEK293 cells, demonstrated the lack of NF‐κB and caspase‐1 activation [42]. Another overexpression study in HEK293 cells observed that NLRP7 inhibited NLRP3‐mediated IL‐1β release without impacting NF‐κB activation and the authors hypothesized that this inhibition is a consequence of the interaction between NLRP7, pro‐caspase‐1 and pro‐IL‐1β, which could block pro‐IL‐1β maturation [43]. Reduced IL‐1β release was also observed in some LPS‐treated peripheral blood mononuclear cells isolated from hydatidiform mole (HM) patients carrying disease‐associated NLRP7 variants (as discussed further below), compared with cells from healthy individuals. However, in most HM patients analysed, the presence of NLRP7 mutations was not associated with IL‐1β levels [44]. Transient overexpression of wild‐type or HM‐associated mutant NLRP7 in HEK293 cells decreased the protein levels of both IL‐1β precursor and mature form, suggesting effects on transcription/translation rather than maturation of IL‐1β [45]. Moreover, this phenotype of reduced pro‐IL‐1β and IL‐1β was abolished in HM‐associated NLRP7 mutants that result in a premature stop codon after the PYD and therefore generate a truncated protein, reminiscent of POPs [11, 12, 45]. A similar phenotype was observed with mutants harbouring deletions of each of the main functional domains, except the PYD [45]. Consequently, NLRP7 was viewed to have anti‐inflammatory activity, reminiscent of other NLRs that can dampen pro‐inflammatory signalling under certain conditions when expressed in HEK293 cells, including NLRP2, NLRP4, NLRP6, NLRP10, NLRP11, NLRP12, NLRX, NLRC3 and NLRC5 [46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60]. However, this anti‐inflammatory activity requires further validation, as contradictory results have been reported in fibroblasts, THP‐1 macrophages and primary macrophages, where NLRP7 did not inhibit NF‐κB, NF‐κB‐mediated cytokine production or inflammatory responses [39, 42, 61, 62, 63].

NLRP7: a sensor for the inflammasome

Contrary to the above‐described anti‐inflammatory responses, evidence also supports the formation and activation of an NLRP7 inflammasome upon bacterial infection in human macrophages [39]. An RNAi screen identified NLRP7 as a component required for IL‐1β release in response to Mycoplasma spp. infection in human macrophages. The NLRP7 inflammasome was also activated in response to infection with Gram‐positive bacteria Staphylococcus aureus and Listeria monocytogenes, as silencing of NLRP7 prevented IL‐1β and IL‐18 release and phenocopied NLRP3 silencing. Furthermore, both NLRs restricted bacterial replication, although only silencing of NLRP3, but not NLRP7, reduced pyroptosis. In addition, even though both NLRs sensed Gram‐positive bacteria, only NLRP7 specifically sensed bacterial di‐ and triacylated lipopeptides, reminiscent of Toll‐like receptor (TLR) 2, and accordingly, TLR2 and NLRP7 were required for IL‐1β release and restriction of replication of L. monocytogenes [39]. In contrast to earlier studies in HEK293 cells, silencing of NLRP7 did not affect NF‐κB activation and therefore did not alter the transcription of pro‐IL‐1β, pro‐IL‐18 and other inflammatory cytokines, but supported an inflammasome role by binding to ASC and assembling with caspase‐1 into a high molecular weight complex exhibiting caspase‐1 activity [39]. The NLRP7 inflammasome was recently confirmed upon Mycobacterium bovis infection in THP‐1 cells, as RNAi‐mediated silencing of NLRP7 reduced pro‐IL‐1β processing and IL‐1β release in response to M. bovis infection. However, the authors also observed decreased IL1B, but upregulated IL18 mRNA levels that were not observed in the earlier study, which will require further investigation. TNF and CCL3 mRNA levels were also reduced, which may be a feedback from autocrine IL‐1β signalling [63]. NLRP7 belongs to the signal‐transducing ATPase with numerous domain (STAND) family, and ATP binding and hydrolysis by the NLRP7 NACHT domain is required for assembly of the NLRP7 inflammasome in response to S. aureus infection [62]. Complex post‐translational modifications regulate inflammasome sensors, and accordingly, ubiquitination of NLRP7 regulates inflammasome activity [61]. Lysine residues 288 and 290 are K‐63‐polyubiquitinated in resting macrophages, and the STAM‐binding protein (STAMBP) deubiquitinates NLRP7 in response to TLR activation to prevent trafficking of NLRP7 to lysosomes and degradation. Accordingly, STAMBP silencing or inhibition impairs TLR‐mediated increase in NLRP7 expression levels and dampens IL‐1β release, providing evidence for ubiquitin‐mediated NLRP7 inflammasome regulation [61]. The direct interaction between NLRP7 and STAM or STAMBP remains to be determined, as is the identity of the NLRP7 ubiquitin‐conjugating enzyme. Further studies will also be necessary to elucidate additional factors controlling NLRP7 inflammasome activity, based on the complex interplay between transcriptional and post‐translational modifications regulating NLRP3 and other NLRs [64]. Finally, while lipoteichoic acid (LTA) from Gram‐positive bacteria binds to the LRR of NLRP6 and activates the NLRP6 inflammasome [65], it is still unclear whether there is any direct interaction between NLRP7 and acylated lipopeptides.

Taken together, these studies provided evidence for a role of NLRP7 in regulating inflammation, particularly in macrophages. While NLRP7 inflammasome activation has been demonstrated in response to bacterial infection in macrophages, its anti‐inflammatory role has been largely observed by overexpression in non‐immune cells. Therefore, elucidating the precise mechanism will require additional studies due to the conflicting activities reported for inflammasome activation and/or NF‐κB regulation.

NLRP7 AND REPRODUCTIVE DISEASES

NLRP7 expression in placental development and early pregnancy

NLRP7 belongs to a reproduction‐associated human gene cluster whose members are highly expressed in reproductive organs [66]. Interestingly, NLRP7 mutations in males do not have any adverse effects on reproductive outcomes, indicating that NLRP7 specifically regulates female reproduction [67, 68]. While NLRP2 and NLRP7 expression decreases during oocyte maturation but then increases 3‐5 days after fertilization in the blastocyst stage of embryonic development, the expression of other reproduction‐associated NLRs decreases irreversibly. Hence, NLRP2 and NLRP7 seem to have a unique function [38]. NLRP7 expression is also very critical in the early stages of placenta development, and particularly in trophoblasts, which provide critical nutrients for the embryo and develop into a large part of the placenta, and in fibroblast‐like endometrial stromal cells, which differentiate into epithelioid‐like decidual cells during decidualization. Trophoblasts undergo a hypoxic period during placental development and express NLRP7 in a hypoxia‐inducible manner. Furthermore, NLRP7 expression increases trophoblast proliferation but negatively regulates trophoblast differentiation [69]. Yet, another study confirms that impaired NLRP7 expression is associated with excessive trophoblast differentiation, which was dependent on bone morphogenetic protein 4 signalling [70, 71]. The importance of controlled NLRP7 expression during placental development is further underscored by the fact that NLRP7 expression is elevated in the placenta of pregnancies suffering from fetal growth restriction (FGR). Interestingly, FGR pregnancies are often characterized by increased inflammation, hinting that the NLRP7 inflammasome could be involved. Indeed, expression of other NLRP7 inflammasome components, including ASC, cleaved caspase‐1 and mature IL‐1β, is also increased in FGR placentas, and circulating IL‐1β, but not IL‐18 levels, is significantly increased in the sera from FGR patients [69]. NLRP7 is also important during decidualization, a process that involves the crucial remodelling of endometrial tissue and cellular differentiation at the beginning of a pregnancy, which is necessary for embryo implantation and placental development. Here, NLRP7 is mainly expressed in the cytosol and nucleus of swollen decidualized human endometrial stromal cells, and in an in vitro decidualization model, NLRP7 is upregulated and translocated to the nucleus after decidualization. Accordingly, NLRP7 silencing impairs, but exogenous NLRP7 expression enhances, decidualization [72]. NLRP7 is also abundantly expressed in M2‐polarized decidual macrophages from first‐trimester pregnancies. While NLRP7 overexpression suppresses M1, it increases M2 macrophage marker expression, which may contribute to endometrial haemostasis and reproductive success [73]. Overall, there is growing evidence that NLRP7 plays an important role during early pregnancy by affecting oocyte maturation, endometrial remodelling and placental development.

NLRP7 and hydatidiform mole

Molar pregnancy or HM is an abnormal pregnancy characterized by trophoblast neoplasia, which leads to defective embryonic development, reproductive wastage and/or spontaneous abortion. Complete HM (CHMs) and partial HM (PHMs) are the two subtypes distinguished by histopathological features and the genetic origin of the molar tissue [74]. The risk of recurrence is relatively low, except for women suffering from rare autosomal‐recessive familial recurrent HM (FRHM) [75, 76]. 5‐25% of women with CHM and 2‐3% of women with PHM may develop an aggressively spreading gestational choriocarcinoma [77]. In the last 15 years, ample evidence has been collected to convincingly link HM to NLRP7, as numerous NLRP7 gene variants are clearly associated with reproduction and imprinting defects [44, 66, 78, 79, 80, 81, 82, 83]. In particular, non‐synonymous variants of NLRP7 are linked to HM. The first variants of NLRP7 located on the long arm of chromosome 19q13·4 were observed in women with a history of recurrent HM, and more than 200 sequence variants (https://infevers.umai‐montpellier.fr) have now been reported in 48‐80% of recurrent HM patients [36, 67, 68, 84, 85, 86, 87, 88]. Mutations include substitutions, insertions, deletions and duplications, which affect most NLRP7 exons and directly alter NLRP7 (Figure 3). A contributing factor may be the high frequency of Alu repeats in NLRP7 introns, which are susceptible to recombination and are known to promote genomic instability.

FIGURE 3.

Mapping and pathogenicity of NLRP7 variants. Usual human genome variation society (HGVS) names of the variants impacting the NLRP7 protein sequence and their pathogenicity are shown and mapped along the NLRP7 gene and NLRP7 protein. Grey, substitutions; red, deletions; blue, duplication; green, insertions. Database: Infevers: an online database for autoinflammatory mutations. https://infevers.umai‐montpellier.fr/ Accessed June 2020 [88, 114, 115, 116]

While the link between NLRP7 genomic variants and HM is compelling, the molecular mechanism and the functional consequences of these mutations are still poorly understood. An emerging hypothesis is that inflammation may contribute to HM pathogenesis, as placental tissues from HM patients exert histopathological features of chorioamnionitis (inflammation of the fetal membranes) and chorangiosis (tissue hypoxaemia leading to an excessive chorion neoangiogenesis and necrosis) [89, 90]. The impact of NLRP7 mutations on inflammatory signalling, the aetiology of HM, and whether the mutations confer gain or loss of function are still elusive. Attenuated NLRP7 may render the host more susceptible to infections by commensal or pathogenic microorganisms, and increasing infections in the lower reproductive tract and/or the loss of immune surveillance could lead to the development of moles. In contrast, hyperactive NLRP7 may exacerbate infection‐triggered inflammation or promote sterile inflammation [39, 62, 63]. Supporting a gain‐of‐function phenotype, three independent HM‐associated mutations displayed elevated basal activity and were hyperactive in promoting inflammasome responses, even in the absence of microbial triggers [39]. These mutations are reminiscent of the NACHT and LRR mutations present in other NLRs in patients with autoinflammatory diseases [31, 33, 34].

NLRP7 and genomic imprinting

Genomic imprinting is an epigenetic mechanism governing imprinted gene expression from only one of the parental alleles [91]. At the chromosomal level, both PHMs and CHMs have an androgenetic origin, which means that two copies of the paternal genome are transmitted, and this results in the overexpression of genes from the paternally inherited allele [92]. Indeed, 75% of HMs are CHMs and have a diploid androgenetic origin, while the 25% of PHMs have a triploid origin [89, 90, 93]. The importance of the androgenetic origin of HMs suggests genomic imprinting defects. Studies carried out on embryonic tissues from women carrying NLRP7 variants revealed abnormal DNA methylation of imprinted germline differentially methylated regions (gDMRs), and abortion is linked to defective placental imprinting [85, 86, 94, 95, 96]. Accordingly, NLRP7 has been implicated in proper DNA methylation and interacts with DMRs in a methylation‐dependent manner [97]. However, NLRP7 variants were absent from women suffering from the imprinting disorder Beckwith–Wiedemann syndrome [98, 99]. A relation between NLRP7, HM and the establishment and/or maintenance of maternal genomic imprinting is emerging, but not every patient suffering from FRHM harbours NLRP7 variants, which is consistent with a multifactorial aetiology [100]. Truncating K homology domain containing 3 like mutations were observed in patients negative for NLRP7 mutations in FRHM, and both proteins colocalized in lymphoblastoid cells [101, 102], suggesting an intrinsic link between these two proteins, but any biological significance of this interaction is unknown. Studying NLRP7 binding partners involved in chromatin remodelling, and DNA methylation identified the transcription and chromatin‐binding factor ying yang 1 (YY1) [97], which binds to imprinted gDMRs in a methylation‐dependent manner to regulate imprinted gene expression. Reduced NLRP7 expression affected DNA methylation and accelerated trophoblast lineage differentiation, but the significance and impact of its interaction with YY1 on genomic imprinting is still unclear, as NLRP7 binding did not affect YY1 expression nor its DNA binding affinity in gDMRs [97]. NLRP7 also binds to the transcriptional repressor zinc finger and BTB domain containing 16 (ZBTB16), which is involved in chromatin remodelling and transcriptional silencing [103, 104]. NLRP7 overexpression increased the cytoplasmic presence of ZBTB16, and NLRP7 may retain ZBTB16 in the cytoplasm, disrupt its chromatin remodelling function and promote the proliferation and differentiation of myeloid progenitors in HM.

Taken together, support exists for a role of NLRP7 in HM, decidualization and regulation of genomic imprinting, and one could envision distinct, cell type‐specific NLRP7 roles in the nucleus and the cytosol, such as imprinting and inflammasome responses, respectively. Deciphering the NLRP7 interaction network and the biological significance of these interactions will help to understand the roles of NLRP7 in inflammasome regulation, genomic imprinting and molar pregnancy. Appropriate NLRP7 expression and NLRP7 inflammasome activity seem to be essential during the early stages of pregnancy. However, further investigation is required to establish how HM‐associated NLRP7 variants might affect NLRP7 inflammasome activity and lead to reproductive wastage.

CONCLUDING REMARKS

A role of NLRP7 in the reproductive system and establishment of pregnancy is strongly supported, but mechanistic details remain elusive. However, NLRP7 expression requires a precise regulation during oocyte maturation and preimplantation of the embryo. NLRP7 is also crucial in decidualization, that is the preparation of the endometrium for and during pregnancy, and is involved in placental development by regulating trophoblast differentiation and proliferation. Excessive trophoblast proliferation is a hallmark of HM and an oncogenic role of NLRP7 in the pathological proliferation of germline cells in testicular seminomas [105], and NLRP7 expression in endometrial cancer tissues [106] has been proposed. Furthermore, the impact of NLRP7 mutations observed in HM patients suffering from autoimmune disorders, such as Hashimoto disease, lupus, vitiligo and Crohn's disease, remains to be explored [89, 107]. More recently, whole‐exome sequencing identified a significant association of the low‐frequency NLRP7 S361L variant with an increased risk of developing ulcerative colitis (UC) [107]. However, any functional insights on how NLRP7 variants contribute to immune system pathologies are still elusive and will require further investigations, but inflammasomes and NF‐κB are crucial for immune homeostasis. While the link of NLRP7 and NLRP7 mutations to reproductive wastage is now established, the precise function of NLRP7 in pre‐embryo and placental development and the causative link between HM‐associated NLRP7 variants and reproductive disorders remain to be discovered. Furthermore, the involvement of the NLRP7 inflammasome in these processes is still elusive. While the first studies proposed an anti‐inflammatory role of NLRP7 in non‐immune cells, further studies showed the assembly of a functional NLRP7 inflammasome particularly in response to pathogenic bacteria in macrophages. Growing evidence suggests a connection between inflammasome regulation in normal pregnancy and supports the possibility of inflammation‐triggered reproductive wastage [108]. The NLRP7 inflammasome could also be involved in the inflammation of fetal membranes and in inflammation‐mediated placental defects leading to fetal growth restriction, but further studies are required. However, the rarity of human oocytes and embryos available for research, as well as the absence of NLRP7 in mice, complicates these studies. In addition to in vitro fertilization and maturation of oocytes, leveraging humanized mouse models could advance this field. Interestingly, NLRP7 likely originated from a gene duplication of NLRP2, which also has a role in embryonic development. Like NLRP7, NLRP2 mutations are also associated with recurrent miscarriage and NLRP2 deficiency is associated with maternal age‐associated fertility loss in mice [109, 110]. Similar to NLRP7, NLRP2 has also been implicated in inflammasome activation and inhibition of NF‐κB [54, 111, 112], but the precise mechanisms are elusive. Exciting opportunities therefore exist to further precisely define the role of NLRP7 and NLRP2, not only in inflammasome signalling during inflammatory responses but also during reproduction, and to determine the consequences for physiology and pathology.

CONFLICT OF INTEREST

None.

AUTHOR CONTRIBUTION

All authors contributed to the concept and writing of this article.

Carriere J, Dorfleutner A, Stehlik C. NLRP7: From inflammasome regulation to human disease. Immunology. 2021;163:363–376. 10.1111/imm.13372

DATA AVAILABILITY STATEMENT

No data were generated for this article.