Recent evolution of the NF-κB and inflammasome regulating protein POP2 in primates
- 3.8k Downloads
Pyrin-only protein 2 (POP2) is a small human protein comprised solely of a pyrin domain that inhibits NF-κB p65/RelA and blocks the formation of functional IL-1β processing inflammasomes. Pyrin proteins are abundant in mammals and several, like POP2, have been linked to activation or regulation of inflammatory processes. Because POP2 knockout mice would help probe the biological role of inflammatory regulation, we thus considered whether POP2 is common in the mammalian lineage.
BLAST searches revealed that POP2 is absent from the available genomes of not only mice and rats, but those of other domestic mammals and New World monkeys as well. POP2 is however present in the genome of the primate species most closely related to humans including Pan troglodytes (chimpanzees), Macaca mulatta (rhesus macaques) and others. Interestingly, chimpanzee POP2 is identical to human POP2 (huPOP2) at both the DNA and protein level. Macaque POP2 (mqPOP2), although highly conserved is not identical to the human sequence; however, both functions of the human protein are retained. Further, POP2 appears to have arisen in the mammalian genome relatively recently (~25 mya) and likely derived from retrogene insertion of NLRP2.
Our findings support the hypothesis that the NLR loci of mammals, encoding proteins involved in innate and adaptive immunity as well as mammalian development, have been subject to recent and strong selective pressures. Since POP2 is capable of regulating signaling events and processes linked to innate immunity and inflammation, its presence in the genomes of hominids and Old World primates further suggests that additional regulation of these signals is important in these species.
KeywordsInflammasome Activation Strong Selective Pressure World Primate Macaque POP2 Domestic Mammal
List of abbreviations
pattern recognition receptor
nucleotide-binding, leucine repeat receptor
caspase recruitment domain
C. jacchus POP2 pseudogene.
Initiation of innate immune/inflammatory responses by pathogens results in the secretion of cytokines that recruit phagocytes, increase phagocyte microbicidal activity, promotes antigen presentation and the development of adaptive immunity . To initiate these responses, pathogens must be sensed through one or more host pattern recognition receptors (PRR). PRRs include the Toll-like receptor (TLR), RIG-I helicase-like receptor, or nucleotide-binding, leucine repeat (NLR) receptor families. PRR engagement by pathogen-associated molecular patterns activates receptor-mediated signaling via MAPK, STAT, and/or NF-κB (reviewed in [1, 2, 3]). Activation of the MAPK and NF-κB pathways cooperate to drive the gene expression of proinflammatory cytokines such as IL-1β, IL-6, IL-8, and TNFα. Secretion of IL-1β and the IL-1β-related cytokine IL-18, requires processing of the respective pro-forms by caspase-1. Activation of caspase-1 occurs in the context of the dynamic multi-protein inflammasome complex through either direct or ASC (apoptotic speck-like protein containing a CARD)-mediated indirect recruitment via NLR proteins [4, 5].
While the molecular basis and regulation of NF-κB signal transduction downstream of PRR family members is well-studied [2, 6, 7], inflammasome function and regulation is poorly understood. Pyrin domain (PYD) and caspase recruitment domain (CARD) homodomain interactions are important for inflammasome formation, suggesting the potential for CARD-only proteins (COPs) and PYD-only proteins (POPs) to act as negative regulators. COPs (e.g. INCA, ICEBERG, and COP) inhibit Caspase-1 activation by preventing Caspase-1 recruitment to the inflammasome complex [8, 9, 10]. Two mammalian POPs have also been discovered. POP1 (ASC2) is highly similar to the PYD of ASC (PyCARD), the adaptor molecule that bridges the PYD of NLRPs to the CARD of Caspase-1 to facilitate inflammasome assembly. Although a potential function, POP1 has not yet been shown to inhibit inflammasome formation/activation . POP2 is more similar to NLR PYDs and effectively inhibits inflammasome activation by limiting the interaction of various NLRPs with ASC [12, 13]. Importantly, the inflammasomes influenced by POP2 include NLRP1, NLRP3, and NLRP12 which have been linked to specific inflammatory diseases including atopic dermatitis ; the cryopyrin-associated periodic syndromes [15, 16], and other hereditary periodic fevers . POP1 and POP2 are also capable of inhibiting NF-κB activation, although the mode of inhibition differs [11, 12]. Thus POP2 has the potential to function as a dual regulator of innate immune/inflammatory responses by influencing both inflammasome function and PRR signaling via NF-κB.
Here we report that the genomes of mouse, rat, and a number of other domestic mammals with available complete genome sequence data lack POP2. While the available genomes of catarrhine primates (comprising both hominids and Old World monkeys) contain POP2, those of New World primates (platyrrhine) do not, strongly supporting the recent evolution of POP2. Our data also reveals an increasing number of NLRP2-related sequences during mammalian evolution. A functional analysis of macaque POP2 reveals a protein capable of both NF-κB and inflammasome inhibition, demonstrating that these functions likely coincide with the emergence of POP2 some time after the divergence of Old World and New World primates approximately 40 mya. The pattern of POP2 evolution and the significance of the recent emergence of both POP1 and POP2 as potential regulators of NF-κB signaling and inflammasome function are discussed.
Results and discussion
POP2is absent from the genomes of mice, rats, and other domestic mammals
POP2is an evolutionarily young gene present in hominid and Old World primates
POP2 likely resulted from a contraction of the NLRP2/7 paralog pseudogene CLRX/NOD24
POP2likely originated from a processed pseudogene (retrogene)
The marmoset (Callthrix jacchus) genome contains a pseudogene sequence representing a sister clade of POP2 (cjΨPOP2). The cjΨPOP2 pseudogene resides downstream of marmoset CCDC50 and upstream of marmoset Fgf12 in same location and orientation as higher primate forms of POP2. Further inspection of cjΨPOP2 reveals the presence not only of the PYD, but also a nucleotide binding domain, and evidence of leucine-rich repeat coding sequences. Introns between these domains are absent, strongly implicating an insertion of a retrogene copy (processed pseudogene) of one of the NLRs, likely the common ancestor of CLRX.1/NLRP2. The ATG start codon is missing and numerous stop codons are present in-frame with the residual PYD sequence. Genes expressed in reproductive tissues are believed to be more likely to generate a heritable processed pseudogene  and POP2 as well as a number of NLRs are expressed in testis and/or oocytes [12, 20, 21, 22, 23]. All of these observations are consistent with our interpretation that in marmoset, cjΨPOP2 is a processed pseudogene and suggest that in primate evolution, an older, functional NLR retrogene was acquired (likely after the divergence of haplorrhine and strepsirrhine primates), modified, and rapidly selected in Old world and hominid primates to produce POP2.
Macaque POP2 is a functional intermediate between NLRP2 and POP2
Both POP1 and POP2 inhibit NF-κB signaling induced by TNFα. While POP1 inhibits IKKα/IKKβ kinase activity upstream of IκBα phosphorylation and cannot inhibit transactivation by the active p65 subunit , POP2 inhibits transactivation by p65 . The isolated PYD of NLRP2 acts similarly to POP1 . Similar to huPOP2, mqPOP2 reduces TNFα-mediated NF-κB activation (Figure 5B). MqPOP2 also inhibits transactivation by NF-κB p65, demonstrating that like huPOP2, mqPOP2 is acting downstream of IκBα phosphorylation at the level of p65. However, the extent of NF-κB inhibition by mqPOP2 is less than that of huPOP2. This may result from the additional C-terminal sequence as its removal (mqPOP2(1-97)) results in a version of the protein with p65 inhibitory activity similar to that of huPOP2 (Figure 5C).
Neither POP1 nor NLRP2 have been demonstrated to prevent inflammasome activation, whereas POP2 is known to block NLR:ASC interaction and inhibits a variety of inflammasomes [12, 25]. MqPOP2 is less effective than human at inhibiting inflammasome activation mediated by ASC-overexpression in HEK293 cells (Figure 5D). Although it remains possible that mqPOP2 may more profoundly inhibit other specific inflammasomes or exhibit species specific inflammasome inhibition, these results demonstrate that mqPOP2 possesses the identified functions of its human ortholog.
Given the high degree of similarity to the NLRP2 PYD, mqPOP2 retains more sequence identity with the NLRP2 PYD than with human or chimp POP2 and likely represents a form of POP2 preceding refinements now fixed in the more broadly functional human POP2. The high degree of conservation between human and chimp POP2 suggests that POP2 refined these functions at some point following the divergence of macaques with purifying selection acting at that point to completely conserve POP2 (Figure 5E) prior to the divergence of humans and chimps (~ 6 mya). The divergence of orangutans from other hominid primates occurred around 14 mya  and orangutan POP2 is not identical to human and chimp POP2 revealing that as recently as 14 mya, this selection was still acting upon the POP2 locus. The selective pressures driving the appearance of both NLRP2 and POP2 are unknown, however the recent emergence of POP2 suggests a very strong selective pressure impacting reproductive success.
Of all extant species, chimpanzees and other primates are most similar to humans at the protein and genomic levels. Rodent species however, are widely used as models for biomedical research. Since arrival in the genomic age has accelerated the pace of discovery and increased our knowledge of comparative genomics, the differences between humans and those species used as disease models as well as the need to understand important differences has become increasingly apparent . It has become clear that humans possess greater diversity in protein families involved in inflammation than rodent models used to model inflammatory disease. This is most evident in members of the IL-1/IL-1R family where both agonist and antagonist members are more abundant in humans than mice . Not surprisingly, humans and mice also show differences in the number of NLR proteins potentially involved in inflammasome activity . In this report we have examined the evolutionary history of pyrin-only proteins (POPs) implicated in the regulation of inflammation and find that the emergence of POPs in the mammalian genome is a very recent event occurring roughly at the divergence of Old and New World primates. Further, our evidence suggests that POP2, which can inhibit inflammasome activity, arose from gene duplication events that first gave rise to multiple paralogs of an ancestral Nlrp2/7-like gene, followed by diversification of these genes, retrogene insertion of an NLRP2-like paralog, and rapid loss of the signature NBD and LRR encoding regions to yield a functional POP2 gene. MqPOP2 approximates the most distant intact ancestor of POP2, predating the divergence of humans and chimps by approximately 5-10 million years, and possesses both the NF-κB p65 and inflammasome inhibitory properties reported for the human protein. Collectively, these data suggest a strong selective pressure driving the recent emergence of a small Pyrin-only protein inhibiting both NF-κB signaling and the activation of multiple inflammasomes that corresponds with the emergence of hominid and Old World primates.
The marmoset genome clearly reveals the prior insertion of a retrogene copy of an NLRP2-like transcript at the developing POP2 locus. By the emergence Old World primates, traces of the non-PYD coding portions of the retrogene are no longer apparent. In humans, a functional polyadenylation sequence is present within the 3' UTR of POP2 . Similar sequences are present in the 3' UTR of chimp and mqPOP2. Beyond reflecting the selective pressures acting on the POP2 gene, these features lend additional support to the emerging hypothesis that retrogenes are frequently functional, add to the complexity of the genome and may confer important regulatory functions as recently demonstrated for fibroblast growth factor 4 .
Do the recent emergence of POP2 and other inflammatory regulators (e.g. POP1, COP, and INCA) reflect an increased need to control inflammation during the most recent stages of primate evolution? This is a difficult question requiring further investigation. However, considering that a strong selective pressure likely drove the evolutionary development of POP2, it is reasonable that POP2 might act by offsetting inflammatory events that decrease reproductive success. Inflammation of the reproductive organs would be a direct example. Recently, mutations in NLRP14 (although not yet demonstrated to initiate an inflammasome) have been implicated in failed spermatogenesis and may dysregulate inflammation or promote apoptosis [21, 22]. Interestingly, although inducible in monocytic cells, POP2 is expressed constitutively in the testis  and could thus potentially have a role in modulating NLRP14 function. As a more extreme example, NOMID, one of the most severe autoinflammatory diseases, is one outcome of mutations in NLRP3, an inflammasome-initiating protein modulated by POP2 [12, 25]. Approximately 20% of individuals with NOMID die before adulthood. Identification of mutations or deletions in POP2 correlating with male reproductive system failure would lend support to this idea. The involvement of NLRP2 and NLRP7 in inflammasome inhibition  and a connection between their presence among maternal RNAs in oocytes and the formation of hydatiform moles [31, 32, 33], a form of reproductive failure, further suggest that some NLRs may have roles in both inflammation and reproductive success. Nevertheless, as no specific disease associations have been identified for the POP2 locus, these ideas remain speculative and await the results of further studies exploring the role of POP2.
In summary, the recent emergence of the highly selected and functional POP2 gene in higher primates, apes, and humans suggests a strong selective pressure among these species for the functions of the POP2 protein. Although our understanding of the biological role of POP2 is in its infancy, its apparent biochemical roles in regulating NF-κB activity and inflammasome formation suggest a variety of possibilities that may shed light on important differences between higher primates and other mammalian species.
Database search strategy, and sequence prediction
TBLASTN searches of the NCBI genome databases for human (build 36.3, March 2008), chimp (Build 2.1, October 2006), macaque (Build 1.1 June 2006), dog (build 2.1, September 2005), cow (Btau_4.0, August 2008), chicken (build 2.1, November 2006), and mouse (build 37.1, July 2007); Washington University genome databases for orangutan (Pongo pygmaeus abelii 2.0.2 contigs) and marmoset (Callithrix jacchus 2.0.2 contigs) were performed without the low complexity filter and with default expect (E) values using the huPOP2 (PYDC2, AY858112.1) as the query. Subsequent analysis of the marmoset genome used the Ensembl database (Callithrix jacchus 3.2). TBLASTN searches of the nr/nr and non-human, non-mouse EST databases were also performed as above.
Phylogenetic analysis of nucleotide and protein sequence
Predicted or known nucleotide and protein sequences for all the identified loci were aligned with each other using CLUSTALX  or MUSCLE . Phylogenetic trees were constructed using the MEGA 3.1  or MEGA 4  software packages. For nucleotide alignments trees were constructed based on the Maximum Likelihood method (Tamura-Nei model ) with gaps and missing data eliminated and bootstrapped with 1000 repetitions. Protein alignments were analyzed and neighbor-joining trees were constructed based on the amino acid: number of differences algorithm with pairwise deletion of gaps and bootstrapped with 1000 repetitions.
The kidney epithelial fibroblast cell line HEK293T was cultured in DMEM with 10% FBS, 1% L-glutamine and 0.1% penicillin/streptomycin cocktail at 37°C, 5% CO2. Peripheral blood leukocytes from Rhesus macaque were the kind gift of Drs. Deborah H. Fuller and Michael Murphey Corb.
RNA isolation and cloning
RNA was isolated from Rhesus macaque peripheral blood leukocytes using RNEasy (Qiagen) reagents. RNA was treated with DNAaseI and the macaque POP2 specific primers 5'-AA GAATTC ATG GCA TCT TCT GCA CAG CTG G-3' and 5'-AA CTCGAG TCA ATA TAC TGG TGA TAT AGA TAT TTC-3' were used with the One-Step RT-PCR kit (Qiagen) to amplify mqPOP2 cDNA. The cDNA product was digested with EcoRI and XhoI and ligated into pcDNA3 (Invitrogen). Two independently isolated clones were sequenced (Genewiz) and returned identical nucleotide sequences. The sequence of this clone of macaque POP2 has been deposited at GenBank (Accession Number: JF327668).
NF-κB reporter assays
NF-κB Luciferase reporter assays were performed as previously described . Briefly, HEK293T (2 × 105 cells/well) were seeded in 6-well plates, transfected with 100 ng 3 × NFκB-luciferase reporter and 50 ng NFκB-p65 with 1 μg myc-huPOP2, myc-macaque POP2, or empty vector using Fugene6 (Roche) as described by the manufacturer (2.5:1 ratio Fugene6:ug DNA. Cells were harvested 18 hours post-transfection, lysed in 1 × reporter lysis buffer (Promega), and luciferase activity was assayed using a Victor3V luminometer (Perkin-Elmer). For TNFα-induced NFκB activation, cells were stimulated with TNFα (10 ng/ml) 3 hrs prior to harvesting. All relative light unit values were normalized to total protein as described .
Inflammasome inhibition assays
HEK293T cells were seeded (5 × 104 cells/well) in 24 well plate a day before the experiment. For inflammasome reconstitution, cells were transfected with plasmids encoding Pro-caspase1 (50 ng), Pro-IL1β (200 ng), and ASC (400 ng) in presence or absence of POP1, full-length huPOP2 or macaque POP2 (500 ng). At 18 hrs post-transfection, culture supernatants were harvested, centrifuged briefly to remove any cellular debris and immediately used for the measurement of secreted IL-1β by human IL-1β ELISA kit (eBiosciences) as per manufacturer's instructions or stored at -20°C for later use. For ASC over-expression, similar experiments were performed by using 400 ng of ASC instead of 20 ng. Note that in ASC over-expression experiments there was no transfection of any NLR.
HEK293T cells were transfected with 1 μg of myc-tagged huPOP2, mqPOP2 WT or mqPOP2 1-97. Cells were lysed 24 hr post-transfection using 1% NP-40 lysis buffer [50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40 (v/v), 2 mM EDTA, 2 mM DTT with protease inhibitors (Roche)], run on SDS-PAGE (Bio-Rad) and transferred to PVDF membrane. After blocking with 3% milk/PBS, membrane was probed using mouse anti-myc antibody (1:1000, Millipore) overnight, followed by HRP-conjugated goat anti-mouse antibody (1:5000, Sigma) for 1 hr. Protein bands were detected using SuperSignal West Pico detection reagents (Thermoscientific).
Experiments were repeated at least three times unless indicated otherwise. Statistical significance between experimental groups were measured by Student's t-test with p < 0.05 considered significant.
The authors thank Dr. Deborah H. Fuller (Albany Medical College, Albany, NY) and Dr. Michael Murphey Corb (University of Pittsburgh, Pittsburgh, PA) for providing macaque peripheral blood and Dr. Ellen B. Duffy for critical reading and assistance in finalizing the manuscript. This work was supported in part by National Institutes of Health grant AI072259 (to J.A.H.).
- 22.Horikawa M, Kirkman NJ, Mayo KE, Mulders SM, Zhou J, Bondy CA, Hsu SY, King GJ, Adashi EY: The mouse germ-cell-specific leucine-rich repeat protein NALP14: a member of the NACHT nucleoside triphosphatase family. Biol Reprod. 2005, 72 (4): 879-889. 10.1095/biolreprod.104.033753.CrossRefPubMedGoogle Scholar
- 23.Ponsuksili S, Brunner RM, Goldammer T, Kuhn C, Walz C, Chomdej S, Tesfaye D, Schellander K, Wimmers K, Schwerin M: Bovine NALP5, NALP8, and NALP9 genes: assignment to a QTL region and the expression in adult tissues, oocytes, and preimplantation embryos. Biol Reprod. 2006, 74 (3): 577-584. 10.1095/biolreprod.105.045096.CrossRefPubMedGoogle Scholar
- 25.Dorfleutner A, Bryan NB, Talbott SJ, Funya KN, Rellick SL, Reed JC, Shi X, Rojanasakul Y, Flynn DC, Stehlik C: Cellular PYRIN domain-only protein (cPOP) 2 is a candidate regulator of inflammasome activation. Infect Immun. 2006Google Scholar
- 29.Parker HG, Von Holdt BM, Quignon P, Margulies EH, Shao S, Mosher DS, Spady TC, Elkahloun A, Cargill M, Jones PG, et al: An expressed fgf4 retrogene is associated with breed-defining chondrodysplasia in domestic dogs. Science. 2009, 325 (5943): 995-998. 10.1126/science.1173275.CrossRefPubMedPubMedCentralGoogle Scholar
- 30.Kinoshita T, Wang Y, Hasegawa M, Imamura R, Suda T: PYPAF3, a PYRIN-containing APAF-1-like protein, is a feedback regulator of caspase-1-dependent interleukin-1beta secretion. J Biol Chem. 2005Google Scholar
- 31.Wang CM, Dixon PH, Decordova S, Hodges MD, Sebire NJ, Ozalp S, Fallahian M, Sensi A, Ashrafi F, Repiska V, et al: Identification of 13 novel NLRP7 mutations in 20 families with recurrent hydatidiform mole; missense mutations cluster in the leucine-rich region. J Med Genet. 2009, 46 (8): 569-575. 10.1136/jmg.2008.064196.CrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.