Phenotypic Mutation 'sinecure' (pdf version)
Allelesinecure
Mutation Type missense
Chromosome11
Coordinate116,493,086 bp (GRCm39)
Base Change T ⇒ A (forward strand)
Gene Rhbdf2
Gene Name rhomboid 5 homolog 2
Synonym(s) cub, iRhom2, 4732465I17Rik, Rhbdl6, Uncv
Chromosomal Location 116,488,991-116,517,786 bp (-) (GRCm39)
MGI Phenotype PHENOTYPE: Mice homozygous for a null mutation display impaired TNF secretion and increased sensitivity to bacterial infection induced mortality. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_172572; MGI: 2442473

MappedYes 
Amino Acid Change Isoleucine changed to Phenylalanine
Institutional SourceBeutler Lab
Gene Model not available
AlphaFold Q80WQ6
SMART Domains Protein: ENSMUSP00000099317
Gene: ENSMUSG00000020806
AA Change: I387F

DomainStartEndE-ValueType
Pfam:Rhomboid_SP 98 306 1.8e-98 PFAM
transmembrane domain 376 398 N/A INTRINSIC
Pfam:Rhomboid 619 763 4.6e-31 PFAM
transmembrane domain 775 797 N/A INTRINSIC
Predicted Effect probably damaging

PolyPhen 2 Score 0.994 (Sensitivity: 0.69; Specificity: 0.97)
(Using ENSMUST00000103028)
SMART Domains Protein: ENSMUSP00000099318
Gene: ENSMUSG00000020806
AA Change: I387F

DomainStartEndE-ValueType
Pfam:Rhomboid_SP 98 304 4.7e-97 PFAM
transmembrane domain 376 398 N/A INTRINSIC
Pfam:Rhomboid 619 763 8.1e-31 PFAM
transmembrane domain 775 797 N/A INTRINSIC
Predicted Effect probably damaging

PolyPhen 2 Score 0.994 (Sensitivity: 0.69; Specificity: 0.97)
(Using ENSMUST00000103029)
Meta Mutation Damage Score Not available question?
Is this an essential gene? Non Essential (E-score: 0.000) question?
Phenotypic Category Autosomal Recessive
Candidate Explorer Status loading ...
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI

none

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL01069:Rhbdf2 APN 11 116492577 missense possibly damaging 0.80
IGL01464:Rhbdf2 APN 11 116491734 missense probably benign 0.18
IGL02060:Rhbdf2 APN 11 116491452 missense probably damaging 1.00
IGL02211:Rhbdf2 APN 11 116491261 missense possibly damaging 0.49
Lostnf UTSW 11 116490987 missense probably damaging 1.00
Lostnf2 UTSW 11 116491017 missense possibly damaging 0.94
Trapezoid UTSW 11 116491974 missense probably damaging 0.96
R0131:Rhbdf2 UTSW 11 116496170 missense probably damaging 1.00
R0399:Rhbdf2 UTSW 11 116494818 missense probably benign 0.00
R0739:Rhbdf2 UTSW 11 116490987 missense probably damaging 1.00
R1756:Rhbdf2 UTSW 11 116498092 missense probably benign
R1839:Rhbdf2 UTSW 11 116491017 missense possibly damaging 0.94
R2029:Rhbdf2 UTSW 11 116491974 missense probably damaging 0.96
R3833:Rhbdf2 UTSW 11 116495250 missense probably damaging 1.00
R4330:Rhbdf2 UTSW 11 116492782 missense probably benign
R4331:Rhbdf2 UTSW 11 116493122 missense probably damaging 1.00
R4872:Rhbdf2 UTSW 11 116492771 missense probably benign 0.04
R5530:Rhbdf2 UTSW 11 116491488 missense probably damaging 1.00
R5625:Rhbdf2 UTSW 11 116496203 missense probably damaging 0.99
R5841:Rhbdf2 UTSW 11 116493180 unclassified probably benign
R6579:Rhbdf2 UTSW 11 116495289 missense probably benign 0.02
R7047:Rhbdf2 UTSW 11 116494477 critical splice donor site probably null
R7403:Rhbdf2 UTSW 11 116491245 missense probably damaging 1.00
R7743:Rhbdf2 UTSW 11 116494775 missense probably benign
R7743:Rhbdf2 UTSW 11 116492427 missense probably benign 0.04
R7855:Rhbdf2 UTSW 11 116493066 nonsense probably null
R8055:Rhbdf2 UTSW 11 116498191 missense probably benign 0.01
R8700:Rhbdf2 UTSW 11 116498230 start gained probably benign
R9052:Rhbdf2 UTSW 11 116494758 missense probably benign 0.00
X0027:Rhbdf2 UTSW 11 116489919 missense probably benign
Mode of Inheritance Autosomal Recessive
Local Stock Live Mice, Sperm, gDNA
MMRRC Submission 034373-UCD
Last Updated 2017-04-06 3:27 PM by Katherine Timer
Record Created unknown
Record Posted 2012-05-23
Phenotypic Description
Figure 1. Responses of sinecure macrophages to different TLR signaling ligands. Homozygous sinecure macrophages produce greatly reduced amounts of TNF in response to the TLR2/6 ligands MALP2, PGN, and LTA, the TLR9 ligand CpG DNA, the TLR4 ligand LPS, and the TLR3 ligand poly(I:C). Responses are partially reduced to the TLR1/2 ligand Pam3CSK4 and the TLR7 ligand resiquimod. 

The recessive sinecure phenotype was identified in a screen of ENU-mutagenized G3 mice for altered responses to Toll-like receptor (TLR) ligands (TLR Signaling Screen).  Peritoneal macrophages isolated from homozygous sinecure mice produced severely reduced amounts of tumor necrosis factor (TNF)-α in response to the TLR2/6 ligands MALP2 (macrophage-activating lipopeptide 2), peptidoglycan (PGN) and lipoteichoic acid (LTA), the TLR9 ligand CpG DNA, the TLR4 ligand lipopolysaccharide (LPS), and the TLR3 ligand poly I:C (a dsRNA mimetic).  These macrophages also exhibited a partially reduced TNF-α response to the TLR1/2 ligand Pam3CSK4 (a triacyl lipopeptide), and to a low dose of the TLR7 ligand resiquimod (a ssRNA mimetic) (Figure 1).

Figure 2. (A) Graph showing the reduced response of homozygous sinecure macrophages to the TLR4 ligand LPS. (B) FACS analysis of TLR4 cell surface expression in wild type (C57BL/6), sinecure, and TLR4 mutant (lps-del) macrophages lacking TLR4. TLR4 expression is unaffected in homozygous sinecure macrophages. 

Although there was a reduced response to the TLR4 ligand, LPS, in the sinecure homozygous macrophages (Figure 2A), TLR4 is localized appropriately to the cell surface (Figure 2B).  

Figure 3. Mutant alleles of Rhbdf2 are non-complementary, and cause a specific block in TNFα secretion.
Thioglycollate-elicited peritoneal cells were isolated from mice of the indicated genotypes, and cultured in the presence of 200pg/ml MALP-2, 1ng/ml LPS, or media alone (-) for four hours. TNFα (A) and IL-6 (B) was then measured in the culture supernatant. (CRhbdf2sin/sin mutants and Rhbdf2+/sin sex-matched littermates were treated with 3% DSS in drinking water for seven days, with weight recorded on each day.  Obtained from (1).

To confirm that the sinecure phenotype was caused by Rhbdf2 mutation, Rhbdf2 null mice were generated from embryonic stem (ES) cells expressing the Rhbdf2tm1a(KOMP)Wtsi allele (MGI: 4362881(1).  The ES cells were injected into FVB blastocysts and transplanted into pseudopregnant females.  Mice heterozygous for the Rhbdf2tm1a(KOMP)Wtsi allele were subsequently mated to each other to generate Rhbdf2 knockout (Rhbdf2KO/KO) animals.  Similar to those of sinecure homozygotes, macrophages from Rhbdf2KO/KO animals secreted less TNF-α in response to MALP-2 and LPS (Figure 3A(1).  Compound heterozygous animals expressing one sinecure allele and one KO allele of Rhbdf2 had a similar phenotype. The sinecure animals did not have as severe a phenotype as the Rhbdf2KO/KO animals indicating that the sinecure mutation is hypomorphic.  The sinecure mutation does not affect IL-6 secretion by macrophages in response to either MALP-2 or LPS (Figure 3B(1).  

When tested for sensitivity to DSS-induced intestinal inflammation, during which processing and secretion of EGFR ligands are necessary for regeneration of the gut epithelium, sinecure homozygotes dipslayed responses similar to those of heterozygous mice, indicating that Rhbdf2 is unessential or redundant for the processing of EGFR ligands (Figure 3C(1).  Furthermore, the sinecure homozygotes did not display epidermal phenotypes consistent with EGFR or TACE mutations (1).

Nature of Mutation
The sinecure mutation was mapped to a region on Chromosome 11 bounded by the proximal marker c11.110.56 and the end of the chromosome, encompassing roughly 10Mb of DNA.  Candidate gene sequencing revealed an A to T transversion at position 1680 of the Rhbdf2 transcript, in exon 10 of 19 total exons.
 
1665 TTCGTTCACATCATCATCACCTTGCTGGTGATC
382  -F--V--H--I--I--I--T--L--L--V--I-
 
The mutated nucleotide is indicated in red lettering, and causes an isoleucine to phenylalanine mutation at amino acid 387 of the RHBDF2 protein.
Illustration of Mutations in
Gene & Protein
Protein Prediction

Figure 4. Structure of RHBDF2. (A) RHBDF2 contains a 6 + 1 transmembrane (TM) domain (TMD) membrane topology and has predicted TMDs occurring at amino acids 382-402, 628-648, 664-684, 687-707, 719-739, 745-765, and 774-794; the 381 N-terminal amino acids are predicted to be cytoplasmic. RHBDF2 has predicted phosphorylation sites at serines 60, 83, 87, 293, 295, and 298 as well as an N-glycosylation site at amino acid 555. The sinecure mutation causes an isoleucine to phenylalanine mutation at amino acid 387.  Modified from (1). (B) The iRhoms contain a 6 + 1 TMD membrane topology.  The N-terminus is cytoplasmic; the C-terminus is intracellular.  The sinecure missense mutation (I387F) is indicated by a red asterisk. Modified from (1).

Rhbdf2 or iRhom2 encodes an 827 amino acid rhomboid-like protein that is 91% identical to its 856 amino acid human homologue (Figure 4).  Two isoforms of the human RHBDF2 exist with the second isoform missing residues 51-79.  Rhomboids are a large family of intramembrane serine proteases present in most organisms that have been shown to activate membrane proteins by cleaving near or in their transmembrane helices and releasing soluble domains from the membrane [reviewed in (5)].  RHBDF2 is considered to be an iRhom, a conserved group of proteolytically inactive rhomboid-like proteins with proline residues in their active site and characteristic luminal loop domains (6)

Active rhomboid proteases can be subdivided into three classes based on membrane topology (6).  The first class containing a six-transmembrane domain (TMD) core is found in Escherichia coli protein GlpG and some eukaryotic rhomboids such as Saccharomyces cerevisiae Rbd2 (7;8).  The next class containing the classic Drosophila melanogaster protein Rhomboid-1, has an extra TMD fused to the C-terminus (6 + 1 topology) and a variable N-terminal domain, which brings the C-terminus to the extracellular or lumenal side of the membrane (9).  The third class is the presenilin-associated rhomboid-like (PARL) subfamily of mitochondrial proteases, which have their extra TMD fused to the N-terminus of the rhomboid core (1 + 6 topology).  This changes the position of their catalytic residues from TMD4 and TMD6 in other rhomboids to TMD5 and TMD7 and reverses their membrane topology (6).  Rhomboid proteases show substantial conservation only in a few regions, the active site formed by the serine protease motif (GxSx in TMD4 and an H in TMD6) and a domain of unknown function present in the L1 loop (between TMD1 and TMD2) and TMD2 that has a prominent tryptophan-arginine motif (WR) (6;9).

Extensive mutagenesis studies, as well as the solved structure of bacterial rhomboid proteases belonging to the first subclass, have further confirmed and defined the key catalytic and functional residues necessary for rhomboid activity (10-15).  The active site lies at the bottom of a cavity that is open to the aqueous environment outside the membrane.  The catalytic serine is located on top of a short TMD surrounded by longer TMDs that shield the serine from membrane lipids.  The catalytic serine and histidine are located near each other in the folded structure and form a hydrogen bond, enabling the histidine to activate the serine for direct nucleophilic attack on a substrate.  Further studies suggest that substrate sequences enter the active site between TMD2 and 5, with the repositioning of TMD5 or the L5 loop serving as potential gating mechanisms (16-19).  The L1 loop of rhomboid proteases lies partially within the membrane and plays an important regulatory role (16;18-20).  Generally, rhomboid proteases cleave type I membrane proteins with helix-destabilizing residues like glycine near their cleavage site (21), but other studies suggest rhomboids may be able to recognize diverse substrates (22).    

 

Like Rhomboid-1, the iRhoms contain a 6 + 1 TMD membrane topology (Figure 4) (6).  Mouse RHBDF2 has predicted TMDs occurring at amino acids 382-402, 628-648, 664-684, 687-707, 719-739. 745-765 and 774-794 with the first 381 N-terminal amino acids predicted to be cytoplasmic (UniProtKB).  This membrane topology has been experimentally determined for the related protein RHBDF1 (23), which has an N-glycosylation site at amino acid 583 (555 in RHBDF2) suggesting that the L1 loop is exposed to the lumen.  The iRhoms lack protease activity due to the presence of a proline in the active site before the expected location of the catalytic serine (6).  In RHBDF2, the catalytic serine has been replaced with an arginine at amino acid 692.  iRhoms also have a greatly extended N-terminus and an extended L1 loop that is highly conserved known as the iRhom homology domain (IRHD).  The IRHD contains 16 conserved cysteines that may form disulfide bonds to make a rigid globular domain (6).  RHBDF2 has predicted phosphorylation sites at serines 60, 83, 87, 293, 295, and 298 (UniProtKB). 

The related iRhom RHBDF1 is able to form dimers or oligomers independently of its N-terminal domain, and can undergo proteolytic cleavage to remove the N-terminal domain (23).  RHBDF2 may have similar properties, but these have not been investigated.  The sinecure mutation causes an amino acid change at amino acid 387 of the RHBDF2 protein, located in first predicted transmembrane domain. It is unknown if the mutated protein is expressed and localizes to the membrane normally.  According to MUpro, which predicts protein stability changes, this change will result in a highly unstable protein (24).

Expression/Localization

According to Symatlas, mouse Rhbdf2 is highly expressed in macrophages, B cells, mast cells, microglia and epidermis.  Other microarray experiments suggest Rhbdf2 expression can be found in adipose tissue, mammary gland, ovary, uterus, B cells, lung, lymph nodes, olfactory epithelium, snout epidermis, digits, and bone marrow (25).  Some Rhbdf2 mRNA has been found in the brain, particularly in the olfactory bulb (Allen Brain Atlas).

Human RHBDF2 expression is present in numerous tissues including B cells, dendritic cell (DC) subsets, NK cells, myeloid cells, monocytes, and some cancers. Some expression is found in T cells (Symatlas). 

Similar to many active rhomboid proteases (26), RHBDF2 is predicted to be a transmembrane protein localized to the endoplasmic reticulum (ER), Golgi apparatus or other components of the secretory pathway due to the presence of N-terminal targeting sequences (see the MultiLoc prediction tool). RHBDF1, which is highly related to RHBDF2, has been shown to be localized to the ER and Golgi in cell lines (23;27).

Background

Figure 6. EGFR ligand processing in the Drosophila secretory system.  In Drosophila, Rhomboid-1 is essential for the cleavage of membrane-bound EGFR ligands in the Golgi.  iRhom inhibits EGFR signaling by shunting EGF ligands from the ER to the ERAD pathway, counteracting the function of Rhomboid-1. Cell culture experiments with mammalian RHBDF2 indicate that the function of RHBDF2 remains conserved (not shown on figure; see text). Figure adapted from (2).

Rhomboids were first recognized as a distinct family of proteins with the identification of the gene altered in a Drosophila melanogaster embryonic mutant with pattern alterations in ventral ectodermal derivatives, known as rhomboid because of its effect on the embryonic head skeleton (28;29).  The effects of loss of rhomboid in multiple different tissues were similar to those caused by loss of epidermal growth factor receptor (EGFR) signaling, and rhomboid expression is required to activate EGFR signaling in Drosophila (Figure 6(29-33).  The EGFR is a receptor tyrosine kinase that is activated by growth factors and relays signals that control many aspects of Drosophila development [reviewed in (34)] (please see the record for Velvet).  Several other rhomboid proteases exist in Drosophila, some with roles that partially overlap with those of Rhomboid-1 (26;35).  Recent studies have found that Drosophila has a single iRhom (encoded by rhomboid-5) that is expressed predominantly in the ER of neuronal cells (2)Drosophila iRhom inhibits EGFR signaling by shunting EGFR ligands from the ER to the ERAD pathway, counteracting the function of active rhomboids (e.g. Rhomboid-1) that promote EGFR signaling (2).  In cell culture, coexpression of Drosophila iRhom with an EGFR ligand, Gurken, and Drosophila Rhomboid-1 led to an inhibition of the Rhomboid-1-induced release of soluble Gurken as well as a decrease in the total level of intracellular Gurken (2).  This study also found that the release of another EGFR ligand, Spitz, was blocked by iRhom coexpression (2).  This study determined that iRhom function occurs predominantly in the ER and that once the EGFR ligand is solubilized by an active rhomboid, it is no longer subject to iRhom-dependent ERAD (2).

Although EGFR signaling is critical for many Drosophila developmental processes, it is especially well-known to be critical for establishing the dorsoventral (DV) asymmetry in the egg chamber during Drosophila oogenesis, which is critical for subsequent DV patterning in the embryo (31;34).  During oogenesis, Rhomboid-1 is localized on the apical surface of the dorsal-anterior follicle cells surrounding the oocyte, thereby selectively activating the EGFR ligand Gurken in the dorsal follicular cells.  EGFR signaling in these cells represses the transcription of pipe in the dorsal half of the follicular epithelium.  Pipe, which encodes a heparan-sulfate 2-sulfotransferase involved in extracellular matrix modification, is then specifically expressed at the ventral side of the developing oocyte and activates a cascade of proteases culminating in the generation of the morphogen Spätzle, which binds to the transmembrane receptor Toll (Tl).  Toll signaling results in activation of the nuclear factor (NF)-κB transcription factor Dorsal (Dl), allowing it to enter cell nuclei and regulate zygotic genes that are important for embryonic patterning and development [reviewed in (36)].  Dl nuclear expression exists in a morphogenetic gradient, with the highest levels of Dl present in ventral nuclei, and no Dl present in dorsal nuclei.  Genes that are activated by high levels of Dl, contain low affinity Dl binding sites in their promoters and are specifically expressed ventrally, while genes containing high affinity Dl binding sites in their promoters are expressed more laterally in the embryo.  The rhomboid promoter contains high affinity Dl binding sites, and is specifically expressed in lateral stripes that encompass the ventral half of the presumptive neuroectoderm (37).  Thus, spatially restricted expression of Rhomboid-1 at different stages during Drosophila development is sufficient to restrict EGFR signaling to specific tissues (34)

Rhomboid proteases have now been found to be involved in many different biological processes in diverse organisms [reviewed in (26)].  These include EGFR signaling in Caenorhabditis elegans (38) and Drosophila (2;29-33), activating a protein translocase involved in exporting proteins across membranes in bacteria (39), mitochondrial morphology and function in multiple organisms (40-42), regulation of apoptosis (43;44), and host cell invasion and immune evasion by parasites (21;45;46).  Apicomplexan parasites are obligate intracellular parasites.  Invasion of host cells by these parasites occurs when parasite cell surface proteins known as adhesins bind to host cell receptors, followed by a process in which actin-based motility drives the adhesins to locate to the posterior end of the parasite, causing invagination of the host plasma membrane around the parasite.  For invasion to be complete, the adhesins are cleaved by rhomboid proteases (21;45).  Rhomboid protease activity is also essential for immune evasion by the extracellular parasite, Entamoeba histolytica (46).  Immune evasion is achieved by surface receptor capping, during which surface receptors that have been recognized by the host immune system are rapidly polarized to the posterior of the cell and released.  Proteolytic cleavage of parasitic lectin by the rhomboid protease, EhROM1, is necessary for this process to occur.      

In mammals, four rhomboid proteases have been partially characterized: RHBDL1 (also known as RRP1) (47), RHBDL2, the mitochondrial rhomboid protease presenilin associated rhomboid-like protein (PARL) (41), and RHBDL4 (also known as Ventrhoid for its ventrally restricted expression pattern) (48).  Membrane-tethered EGFR ligands in mammals are cleaved by the ADAM (a disintegrin and metalloprotease domain) family of metalloproteases (see wavedX for the record on Adam17(49).  However, in vitro studies suggest that B-type ephrins can be efficiently cleaved by RHBDL2.  B-type ephrins activate Eph family receptor tyrosine kinases to guide cell migration and patterning.  RHBDL2 was also found to cleave the anticoagulant thrombomodullin (49).  In vivo, only the mitochondrial protease PARL has been characterized.  Parl knockout mice develop normally until four weeks of age and then display severe growth retardation and loss of muscle mass, with death occurring between eight and twelve weeks old.  These animals displayed severe thymus and spleen atrophy (49).  Mitochondrial respiratory function was normal, but mice displayed abnormal mitochondrial cristae remodeling and excessive cytochrome c release due to abnormal processing of the GTPase OPA1.  Loss of one copy of the human OPA1 gene causes dominant optic atrophy (Optic Atrophy 1; OMIM #165500(50).  PARL was also found to be critical for processing the HtrA2 (high-temperature regulated A2, also known as Omi) protease, which regulates apoptosis in lymphyocytes by preventing accumulation of activated Bax, a Bcl-2-family-related protein (43).  Mutations in HtrA2 are associated with susceptibility to Parkinson’s disease in humans (OMIM #168600).  Rhomboid-7, which encodes a Drosophila mitochondrial rhomboid protease, interacts genetically with the Parkinson’s disease factors, PTEN-induced putative kinase (pink1), and the gene encoding the ubiquitin-protein ligase, Parkin (51).  In cell culture, RHBDF1 and RHBDF2 promote the ERAD of EGFR ligands (e.g. EGF, TGFα, Epiregulin, Amphiregulin, Betacellulin, and Neuregulin 4) (2); the physiological role of RHBDF2 in EGFR signaling in vivo is not known.

 
 

 

Figure 7.  RHBDF2 is essential for TACE to exit the ER.  RHBDF2 associates with TACE in the ER to facilitate the folding and/or trafficking of TACE to the Golgi.  Within the Golgi, TACE is processed by furin to the mature form.  At the plasma membrane, TACE cleaves the membrane-bound form of TNF to generate the soluble TNF that can bind the TNFR.  This image is interactive. Click to image to view additional TACE-related pathway mutations.  Modified from Lichtenthaler, S.F. (2012).

Two recent studies independently examined the role of RHBDF2 in innate immunity using Rhbdf2 knockout (KO) mouse models (Rhbdf2-/-(52;53).  In one study, similar to the findings stated above (Figure 3), Adrain et al. found that ablation of Rhbdf2 had no effect on the induction of IL-1β, IL-6, and IL-12 by peritoneal macrophages upon LPS stimulation (52).  Also, a reduced amount of TNF-α was secreted by LPS-stimulated macrophages, although TNF-α mRNA and protein levels were upregulated normally; TNF was expressed appropriately at the plasma membrane of Rhbdf2-/- macrophages.  This study proposed that changes in the expression of Rhbdf2 does not alter the induction or trafficking of TNF (see the record for PanR1), but may cause a defect in the shedding of the ectodomain of TNF, a process necessary for activation (52).  The study also examined the expression levels and activity of the protease TNF-α converting enzyme (TACE), the protein encoded by Adam17; TACE activity is essential for the cleavage of TNF-α to yield the bioactive form in response to numerous stimuli including PMA and LPS (54-56). Maturation of TACE occurs in the Golgi by proteolytic cleavage mediated by furin proprotein convertases before TACE is trafficked to the cell surface (57). In the Rhbdf2-/-macrophages the levels of TACE were comparable to levels observed in the Rhbdf2+/+ cells, but there was a complete loss of TACE activity (52).  Further examination of the Rhbdf2-/- macrophages found that TACE was absent from the cell surface due to an inability of the protein to traffic from the ER to the trans-Golgi network (TGN).  Adrain et al. proposed that RHBDF2 is either involved in the folding and/or maturation of TACE in the ER, or it is a cargo receptor that assists in the trafficking of TACE (52) (Figure 7).  Additional findings in this study indicate that TACE is unable to access furin in the Golgi.

In the second study, McIlwain et al. demonstrated an association between RHBDF2 and the pro- and mature forms of TACE by immunoprecipitation from RHBDF2-overexpressing cells (53). Peritoneal macrophages from the Rhbdf2-/- mice generated by these researchers upregulated TNF-α and TACE mRNA upon LPS stimulation; however, significantly less TNF-α was secreted into the cell culture supernatant relative to wild type supernatant. The secretion of other cytokines (i.e. IL-5 and IL-12) by Rhbdf2-/- macrophages was normal (53).  In addition, examination of granulocytes, CD4+ T cells, and B cells from the Rhbdf2-/- model revealed impaired surface downregulation of L-selectin (CD62L), another TACE substrate, upon PMA stimulation (53).  McIlwain et al. proposed that RHBDF2 is required for TACE-mediated shedding of multiple surface molecules (e.g. TNF-α and CD62L) from immune cells.  Similar to the findings of Adrain et al., McIlwain et al. found that RHBDF2 is necessary of the maturation and trafficking of TACE.

Putative Mechanism

Figure 8.  The sinecure mutation prevents RHBDF2-mediated trafficking of TACE from the ER.  In normal conditions (semi-transparent), RHBDF2 assists in the trafficking of TACE from the ER to the Golgi wherein it is further processed to generate a mature peptide before reaching the plasma membrane (see Figure 7).  The sinecure mutation in RHBDF2 results in an inability of TACE to traffic from the ER.  Modified from (2).

RHBDF2 is a member of a rhomboid-like family, now known as the inactive Rhomboids (iRhoms) due to the lack of a functional active site.  Mutagenesis experiments with active rhomboids have demonstrated that the presence of an active site proline in these proteins blocks proteolytic activity (6).  As mentioned in the "Phenotypic Description" section, although in vitro studies have shown a possible role of RHBDF2 in ERAD of EGF (2), changes in EGFR signaling were not detected in the sinecure mice (1(Figure 8).  As described above, RHBDF2 appears to be highly expressed in macrophages and other immune cell types (Expression/Localization).  Additionally, macrophages stimulated with LPS further upregulated RHBDF2 (58), while dendritic cells (DCs) showed increased expression in response to double-stranded DNA or CpG DNA.  In response to the TLR5 ligand, flagellin, intestinal epithelial cells upregulate Rhbdf2 (59).  Following B cell receptor activation, B cells downregulate Rhbdf2.  The findings of Adrain et al. (52) and McIlwain et al. (53), and the sinecure phenotype (1) all indicate that RHBDF2 is essential for TNF secretion.  The Adrain et al. and McIlwain et al. studies point to a role for RHBDF2 in TACE maturation (52;53).  Loss of RHBDF2 function, as in sinecure mice, would lead to an inability of TACE to exit the secretory pathway and a subsequent failure in cleavage and release of TNF from the cell surface. 

Primers Primers cannot be located by automatic search.
Genotyping
Sinecure genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide change. 
 
Primers for PCR amplification
Sin(F): 5’- GGTTGACAGTGTGACCTGATAGGC -3’
Sin(R): 5’- CGAGAACTTTGCTCCCAGGTGAATG -3’
 
PCR program
1) 94°C             2:00
2) 94°C             0:15
3) 56°C             0:30
4) 72°C             1:00
5) repeat steps (2-4) 35X
6) 72°C             10:00
7) 4°C               ∞
 
Primers for sequencing
Sin_seq(F): 5’- TCTTCACACAGTAGGCAGTG -3’
Sin_seq(R): 5’- CTAGAAAGAGGCTATGAGCCTC -3’
 
The following sequence of 1160 nucleotides (from Genbank genomic region NC_000077 for linear DNA sequence of Rhbdf2) is amplified:
 
24138                   ggt tgacagtgtg acctgatagg cctcggactc caaattacac
24181 aaagtaagaa agggcaggag tcgaagtaag aaagggcagg agtggtggag tgggtgggga
24241 gcgggtgggg gacttttggg atagcattgg aaatgtaaat gaaataaata cctaataaat
24301 aaagtaaaga aaaagaaagg gcaggagtgg tgggcaagcg gctagggtgg aggggtgctg
24361 ttagggtggg atgccgtgag cagaggtgtg ggccccccca ggtgatgtga agtgtgcaca
24421 aaggtcctgg ggcttcaggg gccgtggcct ggtccaggaa gcacaagcag aggaggtgag
24481 cagcaattgc aggtgagcac aggcaggctt gccagttatc agcaggggat ggctttgatc
24541 ttcacacagt aggcagtggg aagccatcgt gggctttgaa ctgagagaga cactggggat
24601 gcaccttcca agtttctgtg gggaaggctg agctggggga atgcctgcca tgcccaggtg
24661 aagggcgtga ggagcggggc atgagctctg ccaggctgac tggaagcctc ccccggcagg
24721 ccctacttca cctactggct gacgttcgtt cacatcatca tcaccttgct ggtgatctgc
24781 acctatggca tcgcacctgt gggctttgcc cagcacgtta ccacccagct ggtgagtagg
24841 gttcctcctg gggggtcccc ggcctctccc aaggagcttt ggcacagttg gcaccaagta
24901 tctcccacca cagtaccctg gcccaagttg gagatgcctg aggctcatag cctctttcta
24961 gaggcccttt tctggggatg cccccacccc cgtccctttc tcttctcacg ccgagggctc
25021 tggctcctct aatgccacaa actaccttct tgataggtgc tgaagaacag aggcgtgtat
25081 gagagcgtga agtacatcca gcaggagaac ttctggattg gccccagctc ggtgaggccc
25141 aggatgcccg ggagccctgt atcctgccac tccacactgg gcagaggggt ggatggtagg
25201 gtgccccgcc cactgcttcc gagtatgggg cactggctca cagtcccccc cccccccact
25261 ccagattgac ctcattcacc tgggagcaaa gttctcg 
 
PCR primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated A is shown in red text.
References
Science Writers Nora G. Smart, Anne Murray
Illustrators Diantha La Vine, Victoria Webster, Katherine Timer
AuthorsNengming Xiao, Owen Siggs, Bruce Beutler