Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Feb 8.
Published in final edited form as: Cell Host Microbe. 2017 Jan 26;21(2):220–230. doi: 10.1016/j.chom.2017.01.002

Heme Oxygenase 2 Binds Myristate to Regulate Retrovirus Assembly and TLR4 Signaling

Yiping Zhu 1,#, Shukun Luo 2,#, Yosef Sabo 1, Cheng Wang 1, Liang Tong 2,*, Stephen P Goff 1,4,*
PMCID: PMC5300893  NIHMSID: NIHMS841903  PMID: 28132836

Summary

N-myristoylation is the covalent attachment of myristic acid to the N-terminus of proteins in eukaryotic cells. The Matrix domain (MA) of HIV-1 Gag protein is N-myristoylated and plays an important role in virus budding. In screening for host factors that interact with HIV-1 MA, we found that heme oxygenase 2 (HO-2) specifically binds the myristate moiety of Gag. HO-2 was also found to bind TRAM, an adaptor protein for Toll-like receptor 4 (TLR4), and thereby impact both virus replication and cellular inflammatory responses. A crystal structure revealed that HO-2 binds myristate via a hydrophobic channel adjacent to the heme-binding pocket. Inhibiting HO-2 expression, or blocking myristate binding with a heme analogue, led to marked increases in virus production. HO-2 deficiency caused hyperresponsive TRAM-dependent TLR4 signaling, and hypersensitivity to the TLR4 ligand lipopolysaccharide. Thus, HO-2 is a cellular myristate-binding protein that negatively regulates both virus replication and host inflammatory responses.

Graphical abstract

graphic file with name nihms-841903-f0001.jpg

Introduction

N-myristoylation is the covalent attachment of myristic acid, the 14-carbon saturated fatty acid, to the N-terminal glycine of proteins in eukaryotic cells. A large number of proteins of diverse functions are modified by N-myristoylation (Thinon et al., 2014). The addition is catalyzed by N-myristoyltransferases (NMTs), and two isoforms (NMT1 and NMT2) encoded by distinct genes have been identified in mammalian cells (Boutin, 1997; Giang and Cravatt, 1998; Wright et al., 2010). Myristoylation is generally permanent and irreversible. Myristoylated proteins are involved in a wide variety of physiological activities such as virus replication, cell signaling pathways, oncogenesis, and apoptosis (Wright et al., 2010). Examples of myristoylated proteins include the retrovirus Gag structural proteins (Henderson et al., 1983), tyrosine kinase Src and Src kinase family members (Cross et al., 1984), phosphatases such as calcineurin B (Aitken et al., 1982), the BH3 domain protein BID (a key mediator of apoptosis) (Zha et al., 2000), and TRAM (Toll-like receptor adaptor molecule, aka TICAM2), a mediator of TLR4 signaling (Rowe et al., 2006). Many, but not all, myristoylated proteins reside in intracellular membranes.

The Gag and Gag-Pol precursor proteins of nearly all retroviruses are modified by the cotranslational addition of myristate to the amino-terminal glycine of the matrix domain (MA) (Gottlinger et al., 1989; Henderson et al., 1983; Pal et al., 1990). The avian alpharetroviruses and a few other retroviruses (equine infectious anemia virus, bovine immunodeficiency virus, and visna and foamy viruses) are exceptions to the rule. The Gag and Gag-Pol proteins of the avian viruses are modified by N-terminal acetylation (Palmiter et al., 1978; Pepinsky et al., 1996). For all the retroviruses with myristoylation modification on their Gag proteins, the N-myristoylation is essential for replication of these retroviruses, and inhibition of the NMT’s enzymatic activity or mutation of the Gag N-terminal glycine to alanine to prevent myristoylation blocks the spread of virus in host cells (Bryant and Ratner, 1990; Gottlinger et al., 1989; Rein et al., 1986). When Gag myristoylation is prevented, the Gag protein remains in the cytoplasm and is not properly delivered to the plasma membrane for virion assembly and budding (Bryant and Ratner, 1990; Ono and Freed, 1999). Mutational studies have revealed that the N-terminal myristate, and also a cluster of basic amino acids constituting a small basic patch on the surface of MA, are both required for membrane binding of Gag (Resh, 2005). The basic residues of Gag are thought to interact with the negatively charged phospholipids of the plasma membrane to promote its membrane association (Hill et al., 1996; Ono et al., 2000). It has been proposed that in the cytoplasm the N-terminal myristate of Gag is initially trapped by a hydrophobic pocket in the MA domain, limiting the interaction between Gag and endogenous membranes, and that conformational changes (a “myristoyl switch”) associated with virus maturation expose the myristate (Resh, 2004). The plasma membrane-specific lipid PI(4,5)P2 can compete with myristate for binding to the hydrophobic pocket, promoting the exposure and insertion of the myristate tail into the plasma membrane and thus facilitating virus budding (Bouamr et al., 2003; Saad et al., 2007; Zhou and Resh, 1996). The bulk of the MA domain is not absolutely required for membrane association and virion budding. An HIV-1 Gag mutant lacking most of MA and a portion of CA, but retaining the N-terminal myristoylation (so-called “miniGag”) can efficiently mediate virion assembly and release (Accola et al., 2000; Reil et al., 1998), suggesting that the exposure and insertion of the myristate tail is a primary determinant for the membrane association of Gag and virus budding.

It has long been supposed that there must be proteins that bind myristoylated substrates and regulate their localization and function, but few have been identified. UNC119 is a lipid-binding protein of photoreceptors (Higashide and Inana, 1999; Swanson et al., 1998) that interacts with acylated rod photoreceptor transducin α subunit (Tα) and myristoylated ciliopathy protein nephrocystin-3 (NPHP3) (Constantine et al., 2012; Wright et al., 2011; Zhang et al., 2011). Early studies revealed that a protein of 32 kDa, solute carrier family 25 member 5 (SLC25A5), bound to a myristoylated v-Src peptide (Resh and Ling, 1990; Sigal and Resh, 1993). In an effort to find new players in regulation of myristoyated substrates, we have now conducted a search for proteins that bind and modulate myristoylated HIV-1 Gag. We identified heme oxygenase 2 (HO-2), one of two heme oxygenase isoenzymes (HO-1 and HO-2) in mammalian cells (Maines, 1988).

Both HO-1 and HO-2 catalyze the metabolism of heme to form biliverdin, which is subsequently converted to bilirubin and carbon monoxide. The structures of HO-1 and HO-2 have been determined, and the site for binding and cleavage of heme has been located (Bianchetti et al., 2007; Lad et al., 2003a; Schuller et al., 1999). HO-2 is constitutively expressed in all tissues and cell types, while HO-1 is also ubiquitously expressed in most normal cells but induced to very high levels upon oxidative stress, such as treatment with heme or other inflammatory stimuli (Bellner et al., 2009; Prawan et al., 2005). There have long been suggestions that HO-2 plays a role in inhibition of inflammatory responses (Seta et al., 2006). HO-2 knockout mice display higher inflammatory cytokine levels and deficiency in wound healing (Bellner et al., 2009). Overexpression of HO-2 inhibits, while RNAi-mediated depletion of HO-2 enhances, the lipopolysaccharide (LPS)-induced inflammatory response in mouse cerebral vascular endothelial cells (Chen et al., 2014).

Here we report that HO-2 is a myristate-binding protein. The co-crystal structure at 1.9 Å resolution of HO-2 in complex with myristate reveals that HO-2 binds myristate via a long hydrophobic channel and that heme analogues block myristate binding to HO-2. We find that HO-2 negatively regulates the membrane association of HIV-1 Gag, and that knockdown of HO-2 or inhibition of HO-2’s myristate-binding activity, either by mutations altering the hydrophobic channel or by addition of a noncleavable heme analogue, significantly increases HIV-1 virion production. Furthermore, we show that HO-2 also binds to TRAM, the adaptor protein of TLR4, and inhibits the TRAM-dependent LPS-TLR4-induced immune response. Finally, we confirm that LPS induces the expression of HO-2, suggesting that HO-2 is involved in the LPS-TLR4 pathway as a negative feedback regulator. We propose that HO-2 traps many myristoylated proteins to inhibit their membrane association, and hence negatively regulate their functions.

Results

HO-2 is a myristate-binding protein

The Matrix domain (MA) of HIV-1 Gag protein is N-myristoylated and plays an important role in HIV-1 virus budding (Bryant and Ratner, 1990; Gottlinger et al., 1989). To screen for host factors that interact with HIV-1 MA, proteins that bound to MA were isolated by immunoprecipitation and identified by mass spectrometry (Figure S1A). Heme oxygenase 2 (HO-2), which catalyzes the metabolism of heme, was recovered among these proteins. Coimmunoprecipitation (Co-IP) assays showed that ectopically expressed HO-2 efficiently bound wild-type MA, but not an MA mutant with the N-terminal glycine substituted with alanine (G2A), indicating that the interaction between MA and HO-2 was strictly dependent on the N-terminal myristate (Figure 1A). Wild-type MA also bound endogenous HO-2, and again the G2A mutation of MA completely abolished the binding (Figure 1B).

Figure 1.

Figure 1

Open in a new tab

HO-2 is a myristate-binding protein.

(A) Proteins from cells expressing Myc-HO-2 and the indicated forms of MA-flag were immunoprecipitated using anti-flag antibody, and Myc-HO-2 and MA-flag were detected by Western blot.

(B) Endogenous HO-2 interacts with wild-type HIV-1 MA (WT), but not G2A mutant. Proteins of cells expressing indicated MA-flags were immunoprecipitated with anti-flag antibody. Endogenous HO-2 was detected by HO-2 specific antibody.

(C) HO-2 interacts with various myristoylated proteins. Proteins of cells expressing empty vector (EV), or flag-tagged versions of indicated variants of HIV-1 MA or MLV MA, or v-Src and Myc-HO-2, were immunoprecipitated using anti-flag antibody. Myc-HO-2 and myristoylated proteins were detected by Western blot.

(D) Myristic acid competes with HIV-1 MA for binding to HO-2. Proteins of cells expressing Myc-tagged HO-2 and MA-flag as indicated were incubated with indicated concentrations of myristic acid and immunoprecipitated using anti-flag antibody. Myc-HO-2, MA-flag, and control RRS proteins were detected by Western blot.

Also see Figure S1.

The strongly myristate-dependent interaction between HO-2 and MA suggests that HO-2 interacts with MA by binding to the N-terminal myristate directly, and thus might bind to other myristoylated proteins. To test this, we examined the interaction between HO-2 and other proteins known to be myristoylated. HO-2 bound both Moloney leukemia virus (MLV) MA and the v-Src tyrosine kinase of Rous sarcoma virus (Figure 1C). Mutation of the N-terminal glycine of MLV MA or v-Src to prevent N-myristoylation again eliminated their interactions with HO-2 (Figure 1C). We also confirmed that HO-2 bound HIV-1 Gag in addition to MA, and that this interaction was again dependent on myristoylation (Figure S1C). HO-2, however, did not bind to all myristoylated proteins; a survey for binding to the Src family kinases showed strongest binding to c-Src itself, weaker binding to several kinases, and no detectable binding to others (Figure S1B). ). If HO-2 is a myristate binding protein, then myristic acid or other compounds containing a myristate group (e.g. Phorbol 12-myristate 13-acetate, PMA) might block HO-2’s binding to HIV-1 MA. The addition of small molecules containing a myristate group (such as myristic acid or PMA) to the cell lysates indeed decreased the interaction between HO-2 and HIV-1 MA in a dose-dependent manner (Figures 1D and S1D). As a control, the interaction between HIV-1 MA and a known binding protein, arginyl-tRNA synthetase (RRS), whose interaction with MA is independent of its myristoylation, was not affected by addition of myristic acid (Figure 1D). These findings that HO-2 bound many different myristoylated proteins and that free myristic acid could compete with HIV-1 MA for binding to HO-2, suggest that HO-2 is a myristate-binding protein.

Crystal structure of myristate-bound HO-2

To define the molecular details of the interactions between HO-2 and myristate, we determined the crystal structure of their complex at 1.9 Å resolution (Figure 2A; Table S1). Electron density for myristate was observed in a deep hydrophobic channel in all four molecules of HO-2 in the crystallographic asymmetric unit. The myristate has especially close contacts with the side chains of Phe53, Phe57 and Phe234 (Figure 2B and 2C). The aliphatic chain of myristate chain is strongly bent or curved. The carboxylate is located near a large opening of the pocket (Figure 2B) and has very weak electron density (Figure 2D), suggesting that that it is mostly disordered, and is more accessible to solvent, providing space and flexibility in the positioning of a polypeptide attached to it by an amide bond. We also determined the crystal structure at 2.1 Å resolution of human HO-2 in complex with the C12 fatty acid laurate (Figure 2E, Table S1). The crystal was isomorphous to that of the myristate complex. Good electron density for laurate was observed in two of the four HO-2 molecules. Laurate is positioned similarly to myristate (Figure 2E), but the carboxylate group is ordered in the complex (Figure 2F). In addition, we obtained the crystal structure of free HO-2 at 2.0 Å resolution (Table S1), without adding laurate or myristate during crystallization. There is no electron density in the hydrophobic pocket in this structure, confirming that the observed electron density was truly due to the laurate or myristate that was introduced during crystallization. Attempts to cocrystallize HO-2 with a myristoylated MA peptide failed due to the low solubility of the peptide.

Figure 2.

Figure 2

Open in a new tab

Crystal structures of HO-2 in complex with myristate and laurate

(A) Schematic drawing of the structure of HO-2 in complex with myristate. HO-2 is shown as ribbons (light cyan) and myristate as spheres (black for carbon atoms).

(B) Molecular surface of myristate binding site of HO-2, colored by electrostatic potential.

(C) Structural image showing detailed interactions between myristate (black) with HO-2 (light cyan).

(D) Omit Fo–Fc electron density at 1.9 Å resolution for myristate in (a), contoured at 2.5σ. The density for the carboxylate group becomes visible at 2σ.

(E) Structural image showing detailed interactions between laurate with HO-2.

(F) Omit Fo–Fc electron density at 2.1 Å resolution for laurate in (a), contoured at 2.5σ.

(G) Residues in HO-2 essential for myristate-binding activity. Proteins from cells expressing MA-flag and WT or indicated mutant Myc-HO-2 were immunoprecipitated with anti-flag antibody. Myc-HO-2 and MA-flag were detected by immunoblot.

Also see Figure S2.

To assess the importance of the hydrophobic pocket residues for myristate binding, we mutated each of selected amino acids to alanine, and tested the mutant HO-2 proteins for binding HIV-1 MA. While the wild type HO-2 efficiently bound HIV-1 MA, mutation of Phe53, Phe57, Phe234, or Arg156 completely eliminated binding activity, while mutation of Leu74, Tyr134 or Ile233 reduced the binding and mutation of Asn230 had only a small effect on binding (Figure 2G). To assess whether the basic residues of MA or the negatively charged residues of HO-2 contribute to their interaction, we generated mutations altering all the exposed basic amino acid residues (lysine and arginine) on the surface of MA (mutant KRA), or the acidic residues (Asp46 and Glu49) on the surface of HO-2. These mutations had no effects on the binding of HO-2 to MA (Figure S2A and S2B). These experiments suggest that all the important interactions of HO-2 with myristoyl MA are with the lipid.

HO-1 is highly similar to HO-2, with 45% amino acid sequence identity (Figure S2C) and a similar overall structure (Lad et al., 2003a; Lad et al., 2003b; Rahman et al., 2008) and hydrophobic pocket (Bianchetti et al., 2007). However, HO-1 does not interact with HIV-1 MA (Figure S2D). Superimposition of the two structures revealed residues in the HO-1 hydrophobic pocket that are predicted to be unfavorable for myristate binding (Figure S2E). Mutation of the corresponding residues in HO-2 to those residues present in HO-1 (V54M/A70V) significantly reduced HO-2’s myristate-binding activity (Figure 2G). Overall, these mutagenesis studies confirm the importance of the residues predicted by the crystal structure to contact with myristate.

Knockdown of HO-2 enhances production of HIV-1 and MLV virions

The interaction of HO-2 with HIV-1 MA raised the potential for a role in HIV-1 replication. Knockdown of HO-2 expression by siRNAs had no measurable effect on the early phase of virus infection (Figure 3A) but markedly increased the yield of infectious virus released into culture medium, by approximately seven fold (Figure 3B). To test whether the increase in yield of infectious virus was due to an increase in the levels of virion particles produced, or to an increase in the specific infectivity of the virus, the levels of viral Gag protein in the viral harvests were assessed by Western blot. Cells depleted of HO-2 showed a dramatic increase in the levels of CA protein in the culture supernatant, comparable to the increase in yield of infectious virus, indicating that virion particle production per se, and not specific infectivity, was affected (Figure 3C). Knockdown of HO-2 had no effect on the levels of Pr55gag precursor protein in the cell lysates (Figure 3C). Cells depleted of HO-1 showed no change in CA levels from the control (Figure 3C). Probing for the levels of HO-2 and HO-1 confirmed that the knockdowns were efficient, and specific to the targeted gene product (Figure 3C).

Figure 3.

Figure 3

Open in a new tab

HO-2 inhibits retrovirus production.

(A) HO-2 does not affect infection by HIV-1. 293A cells were transfected twice with control non-targeting siRNA (NT) or an siRNA pool targeting HO-2 (siRNA HO-2) and then infected with VSV-G-pseudotyped NL4-3luc virus at 1:10 or 1:100 dilution. Luciferase activities were measured 48 hours after infection. The luciferase activity from cells transfected with control siRNA (NT) and infected with 1:10 diluted virus was set as 100. The data are means +/− SD from three independent experiments.

(B) Knockdown of HO-2 enhances the production of infectious HIV-1 virus. 293A cells were transfected twice with the control non-targeting siRNA (NT) or an siRNA pool targeting HO-2 (siRNA HO-2) and then transfected with pNL4.3luc and pVSVG to package virus. 48 hours after plasmids transfection, equal amounts of the supernatant from transfected cells were used to infect 293A cells. Luciferase activities were measured 48 hours after infection. The luciferase activity from cells infected with virus packaged from control siRNA (NT) transfected cells was set as 1. The data are means +/− SD from three independent experiments.

(C) Knockdown of HO-2, but not HO-1, increases release of HIV-1 VLPs. Cells were transfected with control non-targeting siRNA (NT) or siRNA pools targeting HO-2 or HO-1, and then with viral DNA genomes. Gag protein in the cells (cell lysate) and CA in released virion particles were detected by HIV-1-specific antibody.

(D) Myristate-binding activity of HO-2 is required to inhibit VLP release. Cells expressing WT HO-2, or myristate nonbinding mutant F53A HO-2, were treated to knock down endogenous HO-2, transfected to produce virus, and VLPs were collected 48h later. Gag expression in the cells (cell lysate) and CA in the VLP were detected by HIV-1-specific antibody.

(E) Knockdown of HO-2 increases the release of HIV-1 Gag from Jurkat T cells. JTag-SCR, JTag-HO-2i253, JTag-HO-2i257 cells were transfected with pNL4.3luc. 48 hours after transfection, virus like particles (VLP) were pelleted from supernatant of cells. Gag expression in the cells (cell lysate) and CA in the VLP were detected by HIV-1 p24 antibody.

(F) HO-2 knockdown increases MLV release. Cells were transfected with control non-targeting siRNA (NT) or siRNA pools targeting HO-2, and then with MLV DNA. Gag protein in the cells (cell lysate) and in released VLPs were detected by specific antibody against MLV Gag.

(G) Knockdown of HO-2 has no effect on the release RSV Gag from 293A cells. 293A cells were transfected twice with the control non-targeting siRNA (NT) or the siRNA pool targeting HO-2 (siRNA HO-2) and then transfected with pCMV-RSVGag. 48 hours after transfection, virus like particles (VLP) were pelleted from supernatant of cells. Gag expression in the cells (cell lysate) and CA in the VLP were detected by specific antibodies against RSV Gag.

Also see Figure S3.

To confirm that the increased HIV-1 virus production was attributable to the knockdown of HO-2 and not to off-target effects, we expressed an RNAi-resistant version of HO-2 in the knockdown cells and again measured the yield of virus produced after transfection with viral DNAs (Figure 3D). Re-expression of HO-2 reversed the increase in virion particle yield back to normal levels produced by control cells. Expression of a non-binding mutant of HO-2 (F53A) did not change the increased virion yield of the knockdown cells, indicating that HO-2’s ability to limit virus production is dependent on its myristate-binding activity. We note that even though the levels of HO-2 expression in these experiments were much higher than the endogenous levels of HO-2, in no case were the levels of CA reduced below the levels seen in wild-type cells. We conclude that depletion of HO-2 allows for abnormally high levels of virion production, while the endogenous levels or overexpressed levels of HO-2 repress production equally to a similar basal level in unmanipulated cells. We speculate that knockdown of HO-2 may expose some new machinery or pathway for HIV-1 budding, and that overexpression of HO-2 blocks this but cannot inhibit the basal pathway for HIV-1 virus production.

We tested the effect of HO-2 on virus yield in several other settings. Knockdown of HO-2 in the Jurkat T cell line expressing either of two short hairpin RNAs again resulted in large increases in the yield of HIV-1 virion particles (Figure 3E). The effect was not limited to HIV-1. Knockdown of HO-2 in 293A cells resulted in a significant increase in the levels of MLV virions (Figure 3F), but had no effect on the yield of avian Rous sarcoma virions, whose Gag precursor lacks the N-terminal myristate (Figure 3G). We also examined the effects of HO-2 knockdown on replication of wild-type MLV in permissive Rat2 cells as judged by appearance of reverse transcriptase activity in the culture medium. MLV replicated more efficiently in HO-2 knockdown cells compared to control cells (Figure S3A and S3B).

The association of HIV-1 Gag with the plasma membrane and subsequent virus production is dependent on both the N-terminal myristoylation of Gag and also a cluster of basic amino acids near the amino-terminal region of MA (Hill et al., 1996; Resh, 2004). Mutation of certain residues of MA (I19K/L21K, K29T/K31T, and Y86G) can inhibit membrane association, or can redirect HIV-1 Gag to endogenous membranes (Ono et al., 2000). These sequences of MA are not absolutely required for virus production, and a mutant HIV-1 Gag lacking nearly all of MA but retaining only the first 7 amino acids (miniGag) can still support virus production (Accola et al., 2000). HO-2 knockdown resulted in a similar increase of HIV-1 virus production for virus with any of several MA mutations (Figures S3C) or a large MA deletion (Figure S3D, miniGag). In some cases the knockdown of HO-2 increased the virus production from nearly undetectable to readily detectable levels (Figure S3C and S3D). The increase in HIV-1 virus production was not dependent on virus maturation, as virus with a mutation in the PR protease that blocked cleavage of the Pr55gag precursor still showed increased virion production upon HO-2 knockdown (Figure S3D, Protease mutation D25N).

Heme analogue blocks HO-2’s myristate-binding activity and enhances virion production

Superposition of the structure of heme-bound HO-2 with myristate-bound HO-2 revealed that the heme binding site is close to the opening of the hydrophobic pocket and suggests that heme binding could block the access of myristate to the pocket (Figure 4A). Indeed, a noncleavable heme analogue, tin protoporphyrin IX dichloride (SnPPIX), inhibited the interaction between HIV-1 MA and HO-2 in a dose-dependent manner (Figure 4B). SnPPIX had no effect on the interaction between HIV-1 MA and RRS, whose interaction with MA is independent of its N-myristoylation (Figure 4B). Addition of SnPP IX to cells transfected with viral DNA caused dramatic increases in virus yield, comparable to those seen after knockdown of HO-2 (Figure 3C and S4). These observations provide further evidence that the myristate-binding activity of HO-2 inhibits HIV-1 virion production.

Figure 4.

Figure 4

Open in a new tab

SnPP IX inhibits HO-2’s myristate-binding activity.

(A) Molecular surface of HO-2 catalytic active site and myristate binding site, colored by electrostatic potential. (heme in salmon and myristate in black).

(B) Heme analog SnPP IX inhibits HO-2’s binding to HIV-1 MA. Proteins from cells expressing MA-flag and treated with indicated concentrations of SnPP IX were immunoprecipitated using anti-flag antibody. Endogenous HO-2, RRS, and MA-flag were detected by immunoblot.

(C) SnPP IX treatment increases VLP production. Virus-producing 293A cells were treated with SnPP IX at indicated concentrations for 48 hours and VLPs were pelleted from culture supernatant. Gag in cell lysate and in the VLP were detected by HIV-1-specific antibody.

Also see Figure S4.

HO-2’s myristate-binding activity negatively regulates the membrane association of HIV-1 Gag

HIV-1 Gag is translated in the cytoplasm as a soluble protein, and then is rapidly transported to the plasma membrane to initiate virion assembly. Knockdown of HO-2 significantly increased the level of Gag in the membrane fraction (Figure 5A and 5B). Overexpression of wild type HO-2, but not mutant HO-2 deficient in myristate-binding activity (F53A), reduced the portion of membrane-associated Gag back to a level comparable to that in control cells (Figure 5B). These results indicate that the binding of HO-2 to the N-terminal myristate moiety of Gag inhibits the membrane association needed for virus production. To test whether HO-2 could dissociate Gag from membrane, we incubated recombinant HO-2 protein with membrane fractions from Gag-expressing 293A cells, and examined the levels of membrane-associated Gag by Western blot. We found that HO-2 could not dissociate Gag from membrane (Figure S5).

Figure 5.

Figure 5

Open in a new tab

HO-2 inhibits the membrane association of HIV-1 Gag

(A) Knockdown of HO-2 enhances membrane association of HIV-1 Gag. 293A cells were transfected twice with control non-targeting siRNA (NT) or the siRNA pool targeting HO-2 (siRNA HO-2) and then transfected with pNL4.3luc. 48 hours after transfection, membrane floatation assay was used to determine the distribution of Gag in the cytosol and membrane fraction. ATPA1 is a membrane fraction marker, while GAPDH is a cytosol fraction marker.

(B) HO-2’s myristate-binding activity inhibits HIV-1 Gag membrane association. Cells expressing WT HO-2, or myristate nonbinding mutant F53A HO-2, were treated to knock down endogenous HO-2 and transfected with pNL4.3luc. Lysates of cells were fractionated and the subcellular distribution of Gag was determined by immunoblot using HIV-1-specific antibody. ATPA1 is a membrane fraction marker.

Also see Figure S5.

To test for HO-2’s potential effects on Gag localization, we examined the subcellular localization of Gag in control and HO-2 KD 293A cells by confocal imaging after fixing and immunostaining for Gag. We did not observe dramatic changes in Gag subcellular localization (data not shown). In both cases Gag was localized in a punctate pattern in the plasma membrane and cytoplasm. We did not see any accumulation of Gag in the Golgi as often seen with some MA mutants (Ono et al., 2000). We conclude that the increased virus production by HO-2 KD is not associated with dramatic changes in the steady state distribution of Gag. The increased rate of flow of Gag through the cell is apparently occurring through the normal intracellular locations.

HO-2’s myristate binding activity regulates the LPS-TLR4 signaling pathway via TRAM

The primary function of HO-2 is the metabolism of heme to carbon monoxide, iron, and biliverdin, subsequently reduced to bilirubin. By reducing levels of heme, a potent mediator of inflammation, and producing bilirubin, which protects from oxidative damage (Dore et al., 1999), HO-2 inhibits inflammatory responses (Seta et al., 2006). But HO-2 has many other physiological effects. HO-2 knockout mice display higher inflammatory cytokine levels and deficiency in wound healing (Bellner et al., 2009). Depletion of HO-2 enhances, while overexpression of HO-2 inhibits, the lipopolysaccharide (LPS)-induced inflammatory response in mouse cerebral vascular endothelial cells (Chen et al., 2014). These activities of HO-2 may not involve heme metabolism, and the finding that HO-2 binds myristate suggests that it regulates the function of specific myristoylated proteins. To search for such proteins, we constructed 293A cell lines stably expressing flag-tagged versions of either wild type HO-2 (HO-2 WT) or myristate nonbinding HO-2 mutants (HO-2 F53A or F57A), immunoprecipitated HO-2, and analyzed the bound proteins by mass spectrometry. We identified 29 proteins that were preferentially associated with wild-type HO-2 but not with F53A or F57A mutants (Figure S6A). Six of these contain an N-terminal glycine and are thus likely to be modified by myristate. Our attention was particularly drawn to Toll-like receptor adaptor molecule 2 (TRAM, aka TICAM2), an adaptor molecule involved in the innate immune signaling pathway downstream of the TLR4 cell surface receptor (Fitzgerald et al., 2003; Kagan et al., 2008; Sacre et al., 2007; Yamamoto et al., 2003). Myristoylation of TRAM is essential for its function in the LPS-TLR4 immune response (Rowe et al., 2006). We confirmed that HO-2 interacts with TRAM by coimmunoprecipitation, and we tested mutants to show that the interaction required myristoylation of TRAM and the hydrophobic pocket of HO-2, but not the heme oxygenase activity (Figure 6A).

Figure 6.

Figure 6

Open in a new tab

HO-2 inhibits the TRAM-dependent LPS-TLR4 pathway via its myristate-binding activity.

(A) Interaction between HO-2 and TRAM requires TRAM N-myristoylation and HO-2 myristate binding site. Cells expressing flag-tagged WT or mutant G2A TRAM were transfected with Myc-tagged versions of WT or myristate nonbinding mutant F53A, or catalytically inactive mutant H45A of HO-2. Proteins in lysates were immunoprecipitated using anti-flag antibody and probed by immunoblot.

(B) Knockout of HO-2 enhances TRAM signaling. Control or HO-2 KO cells were transfected with a RANTES-luciferase reporter and indicated amounts of TRAM expression DNA. Upper panel: Luciferase activities 24 h after transfection, relative to control cells without TRAM. Lower panel: immunoblot of HO-2.

(C) HO-2’s myristate-binding activity, but not its heme oxygenase activity is required for inhibition of TRAM signaling. HO-2 KO cells were engineered to stably express WT or myristate-nonbinding mutant (F53A) or catalytically inactive mutant (H45A) of HO-2, and then transfected with a luciferase reporter of RANTES gene expression, and indicated amounts of TRAM expression DNA. Upper panel: Luciferase activities 24 h after transfection, relative to control cells without TRAM. Lower panel: immunoblot of HO-2.

(D) Knockdown of HO-2 enhances the expression of RANTES induced by LPS. THP-1-MD2-CD14 cells stably expressing the indicated shRNAs were stimulated with indicated concentration of LPS for 24 h. Levels of RANTES in the supernatant were measured by ELISA.

(E) LPS induces the expression of HO-2. THP-1-MD2-CD14 cells were treated with 10 ng/ml LPS for indicated time. Levels of HO-2 mRNA were measured by qRT-PCR normalized to that of GAPDH and presented relative to LPS untreated cells.

Data (A-D) are representative of at least three independent experiments (mean and s.d.). Data (E) are mean and s.d. of triplicate determinations.

Also see Figure S6.

TRAM activates the IRF3- and NFκB-dependent immune and inflammatory response to induce the expression of the chemokine RANTES (C-C motif ligand 5, CCL5) (Fitzgerald et al., 2003; O'Neill et al., 2013; Yamamoto et al., 2003). The LPS-induced expression of RANTES is specifically dependent on TRAM (Fitzgerald et al., 2003). We used CRISPR technology to knock out HO-2 and then monitored the ability of ectopic expression of TRAM to activate a luciferase reporter dependent on the RANTES promoter (RANTES-luc). While TRAM induced RANTES-luc expression ~10 fold in control cells, TRAM increased RANTES-luc expression more than 25-fold in HO-2 KO lines (Figure 6B). Expression of wild-type HO-2 in these HO-2 KO cells dramatically reduced TRAM’s ability to induce RANTES-luc, while a mutant HO-2 deficient in myristate-binding activity (HO-2 F53A) did not (Figure 6C). Notably, a mutant HO-2 without heme oxygenase activity (HO-2 H45A) still inhibited TRAM’s function in activating RANTES-luc (Figure 6C). These results indicate that HO-2’s inhibitory effect on the function of TRAM is dependent on its myristate-binding activity, but not on its heme oxygenase activity.

To further confirm the involvement of HO-2 in the LPS-TLR4 pathway, we generated two clones of THP-1-MD2-CD14 cells deficient in HO-2 by shRNA knockdown, and examined the induction of RANTES by the ligand LPS. In the parental cells, the expression of RANTES was induced by LPS at a concentration of ~1 ng/ml, while in HO-2 knockdown cells, 1000X lower LPS sufficed (Figure 6D). HO-2 knockdown cells treated with LPS at 1 ng/ml released 3 to 5 fold more RANTES than control cells. (Figure 6D). These findings show that HO-2 is responsible for negatively regulating this important inflammatory pathway.

LPS treatment induces the expression of HO-2 in diaphragm and primary macrophages (Barreiro et al., 2002; Litvak et al., 2009), and we confirmed that this also occurs in a monocyte cell line (Figure 6E). These results indicate that HO-2 acts as a negative feedback regulator in the LPS-TLR4 pathway, consistent with the higher levels of inflammatory cytokines in an HO-2 KO mouse (Bellner et al., 2009). Our results provide a mechanistic explanation for this observation, suggesting that HO-2 neagtively regulates the LPS-TLR4 pathway by specifically targeting the TLR4 adaptor protein TRAM via myristate binding.

Discussion

We have here identified HO-2 as a myristate-binding protein that negatively regulates the functions of targeted myristoylated proteins. We note that HO-2 is expressed well in T cells (Figure S6B) and macrophages (Bellner et al., 2015), cells relevant to HIV-1 infection, and is induced by LPS but not increased by HIV-1 infection or interferon (Kleinman et al., 2014; Rusinova et al., 2013). HO-2, at its endogenous levels, is acting to inhibit or delay the association of its binding partners with membrane.

The involvement of HO-2 in the LPS-TLR4 pathway has been previously noted; overexpression of HO-2 inhibits, while knockdown of HO-2 enhances, the expression of IL-6 and TNFα induced by LPS in cerebral vascular endothelial cells (Chen et al., 2014). Our results provide a mechanistic explanation for these observations, suggesting that HO-2 regulates the LPS-TLR4 pathway by specifically targeting the TLR4 adaptor protein TRAM (Figure 6). LPS treatment has been shown to induce the expression of HO-2 in diaphragm and primary macrophages (Barreiro et al., 2002; Litvak et al., 2009). We also determined that LPS treatment induced the expression of HO-2 in THP-1-MD2-CD14 cells (a monocyte cell line expressing the two TLR4 accessory proteins MD2 and CD14) (Figure 6E). These results indicate that HO-2 acts as a negative feedback regulator in the LPS-TLR4 pathway, and are consistent with the observation that HO-2 KO mouse display higher levels of inflammatory cytokines (Bellner et al., 2009).

Many of the regulatory functions mediated by HO-2 may involve changes in the localization or trafficking of its binding partners. The myristoyl moiety of retroviral Gag proteins and TRAM protein plays a major role in their localization to the membrane (Ono and Freed, 1999; Rowe et al., 2006). Proteins that bind myristate thus have the potential to directly and profoundly affect the membrane localization of many cellular proteins. HO-2 may trap the myristate moiety of Gag and prevent it from inserting in its proper conformation into the membrane, thereby inhibiting Gag multimerization and HIV-1 virion production (Figure 7). Upon depletion of HO-2, myristoylated Gag is more efficiently delivered to the plasma membrane, resulting in higher yields of released virions.

Figure 7.

Figure 7

Open in a new tab

Working model for functions of HO-2.

(A) The binding of HO-2 to the N-terminal myristate of HIV-1 Gag traps the myristate moiety and prevents it from inserting into the membrane in its proper conformation, and thus inhibits HIV-1 virion production.

(B) HO-2 binds to the N-terminal myristate moiety and downregulates the function of TRAM. HO-2 is induced by LPS-TLR4 signaling and acts as a negative feedback regulator of the LPSTLR4 pathway.

Myristoylated TRAM is localized in membranes (Rowe et al., 2006) and the trafficking of TRAM from plasma membrane to the endogenous membrane is essential for its signaling function in the LPS-TLR4 pathway (Kagan et al., 2008). HO-2’s inhibitory effect on TRAM could be mediated either by blocking the proper membrane association of TRAM or by interfering with the proper trafficking of TRAM between different membrane compartments. UNC119, a myristate-binding protein mainly expressed in retinal cilium (Higashide and Inana, 1999; Swanson et al., 1998), has been shown to dissociate myristoylated target proteins from membrane and facilitate their trafficking through the cytosol between different membranes (Constantine et al., 2012; Wright et al., 2011; Zhang et al., 2011).

HO-2 interacts with many different myristoylated proteins (Figure 1C), but the strength of the interaction varies widely. For example, based on the efficiency of coimmunoprecipitation, the interaction between HO-2 and HIV-1 MA is much stronger than the interaction between HO-2 and MLV MA or HIV-1 Gag (Figure 1C and S1C). HO-2 binds some but not all Src family members (Figure S1B). The myristoyl moiety of many myristoylated proteins may be buried and sequestered in hydrophobic pockets and thus not available for recognition by HO-2 (Hantschel et al., 2003; Patwardhan and Resh, 2010). The exposure of the myristate of HIV-1 Gag by the “myristyl switch” triggered upon PIP2 binding (Resh, 2004) may make it available for binding by HO-2.

It is possible that HO-2 binds and regulates molecules with hydrocarbon chains other than myristate. Two other proteins that bind myristate show some flexibility in the length of the acyl chains of the bound fatty acids: UNC119 can bind to laurate (12-carbon) or myristate (14-carbon) (Wright et al., 2011; Zhang et al., 2011), while NMT can bind to both myristate and palmitate (16-carbon). The position of myristic acid in complex with HO-2 (Figure 2A, 2B) suggests that HO-2 could bind and regulate proteins carrying acyl chains that are somewhat longer or shorter than myristate. We confirmed directly that HO-2 could indeed bind to both laurate and palmitate (Figure 2E and data not shown), and the set of proteins we identified as bound by HO-2 included several known or candidate palmitoylated proteins, including KIAA2013, MBLAC2, and SCAMP2 (Dowal et al., 2011). HO-2 also bound a number of lipid trafficking proteins, including extended synaptotagmin-2 (EXY2) and oxysterol-binding protein-related protein 8 (OSBPL8), and the interaction of HO-2 with these proteins is dependent on its hydrophobic pocket (Figure S6B). Further study may uncover a role for HO-2 in regulating the activity of other lipid modified proteins and possibly the trafficking of lipids in the cells.

In conclusion, we have here identified HO-2 as a myristate-binding protein, defined a hydrophobic pocket as the myristate binding site, and characterized its regulatory effects on both viral and host substrates (Figure 7). The findings that HO-2 regulates both retroviral assembly and the LPS-TLR4 pathway suggest that it has broad regulatory roles in diverse cellular processes involving selected myristoylated proteins.

Experimental Procedures

DNA Constructs and Cells Culture

See Supplemental Experimental Procedures.

Transfection, Virus packaging, and Infection

All the plasmid transfections in adherent cells were performed using lipofectamine 2000 (Invitrogen) following the manufacturer’s protocol, while DMRIE-C Transfection Reagent (Invitrogen) was used for transfections in Jurkat cells.

To generate NL4-3luc-based VSV-G pseudotyped viruses, viral vectors (pNL4-3luc) together with pVSV-G, were transfected into HEK293T cells. To package MLV-based VSV-G pseudotyped viruses, viral vectors together with pHIT60 (expressing MLV Gag and Gag-Pol) and pVSV-G were transfected into HEK293T cells. To package lentiviral vector based VSV-G pseudotyped viruses, viral vectors together with pCMVdeltaR8.2 (expressing HIV-1 Gag and Gag-Pol) and pVSV-G were transfected into HEK293T cells. 48 hours after transfection, virus in media was collected and filtered through a 45 μm membrane.

Unless otherwise indicated, viruses were 3-fold diluted with cell culture medium containing 20 mM HEPES (pH7.5) and 4 μg/ml polybrene. Adherent cells were infected by diluted viruses for 3 hours, while suspension cells were infected by diluted viruses overnight.

Virus-like particle (VLP) detection

The supernatant medium from cells (3 ml) was layered above 1 ml of 25% sucrose in TEN buffer [10mM Tris-Cl (pH 8.0), 0.1M NaCl, 1mM EDTA (pH 8.0)]. Samples were centrifuged at 100,000 × g (~ 28,000 rpm) for 2 h at 4°C (SW55 rotor, Beckman). The virus like particle pellets were resuspended in 100 μl of 1 × SDS loading buffer, resolved by SDS-PAGE, and analyzed by Western blot.

MLV Replication Assay

See Supplemental Experimental Procedures.

Immunoprecipitation and Western blot

See Supplemental Experimental Procedures.

Reagents

Reagents used included: myristic acid (Sigma, 70079); PMA (Sigma, P8139); Tin Protoporphyrin IX dichloride (Santa Cruz, sc-203452); LPS (Enzo Life Sciences, ALX-581-007-L001).

Protein expression, purification, and crystallization

See Supplemental Experimental Procedures.

Data collection, structure determination and refinement

See Supplemental Experimental Procedures.

siRNA Transfection

See Supplemental Experimental Procedures.

Luciferase Assay

Firefly luciferase activities were measured by Luciferase Assay System (Promega). Renilla and firefly luciferase activities were measured by the Dual-luciferase Reporter Assay System (Promega).

Membrane floatation assay

Cells were washed and suspended in homogenization buffer and incubated on ice for 10 min. Cell suspensions were subjected to 30 strokes in a Dounce homogenizer and clarified by centrifugation. After homogenization, cell extracts were placed at the bottom of an ultracentrifuge tube. A discontinuous gradient was formed above the cell extracts. Samples were centrifuged at 100,000 × g for 18 h at 4°C. Fractions were collected from the top of the gradient. The total proteins of each fraction were precipitated with TCA, resolved by SDS-PAGE, and analyzed by Western blot. For detailed buffer compositions, please see Supplemental Experimental Procedures.

Cell fractionation

Cell fractionation was performed by using Plasma Membrane Protein Extraction Kit (Abcam, ab65400).

Elisa Assay

The levels of RANTES in the supernatant of culture cells were measured by Human CCL5 (RANTES) ELISA Kit (Biolegend, 440807) according to the manufacturer’s protocol.

Real-time PCR

See Supplemental Experimental Procedures.

Statistics

The mean values ± SD or mean values ± SE were calculated from at least three independent experiments unless otherwise indicated.

Supplementary Material

Highlights.

  • Heme oxygenase 2 (HO-2) binds myristate via a hydrophobic channel

  • HO-2 negatively regulates the functions of myristoylated proteins

  • HO-2 inhibits the production of HIV-1 and MLV virions

  • HO-2 is a negative feedback regulator of TLR4 signaling

In Brief.

Zhu et al. identify heme oxygenase 2 (HO-2) as a myristate-binding protein that interacts with an array of myristoylated proteins to negatively regulate their function. HO-2 binds HIV-1 Gag, inhibiting virion production, and TRAM, a key molecule in the LPS-TLR4 signaling pathway, to downregulate inflammatory responses.

Acknowledgements

We thank Heinrich G. Göttlinger, Leslie J. Parent, and Massimo Pizzato for generously providing reagents. Mass spectrometry was performed by Mary Ann Gawinowicz and Emily Chen in the Columbia University Medical Center protein core facility. This work was supported by the Howard Hughes Medical Institute and by National Institutes of Health (NIH) grants R01AI106629 (to S.P.G.) and R01CA030488 (to S.P.G.), and R35GM118093 (to L.T).

Footnotes

Author Contributions

S.P.G., L.T., Y.Z. and S.L. planned the experiments and wrote the paper. The crystal structure of myristate-bound HO-2 was determined and analyzed by L.T. and S.L. Y.Z., S.L., Y.S., and C.W. conducted the experiments.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Accola MA, Strack B, Gottlinger HG. Efficient particle production by minimal Gag constructs which retain the carboxy-terminal domain of human immunodeficiency virus type 1 capsid-p2 and a late assembly domain. Journal of virology. 2000;74:5395–5402. doi: 10.1128/jvi.74.12.5395-5402.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aitken A, Cohen P, Santikarn S, Williams DH, Calder AG, Smith A, Klee CB. Identification of the NH2-terminal blocking group of calcineurin B as myristic acid. FEBS letters. 1982;150:314–318. doi: 10.1016/0014-5793(82)80759-x. [DOI] [PubMed] [Google Scholar]
  3. Barreiro E, Comtois AS, Mohammed S, Lands LC, Hussain SN. Role of heme oxygenases in sepsis-induced diaphragmatic contractile dysfunction and oxidative stress. American journal of physiology Lung cellular and molecular physiology. 2002;283:L476–484. doi: 10.1152/ajplung.00495.2001. [DOI] [PubMed] [Google Scholar]
  4. Bellner L, Marrazzo G, van Rooijen N, Dunn MW, Abraham NG, Schwartzman ML. Heme oxygenase-2 deletion impairs macrophage function: implication in wound healing. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2015;29:105–115. doi: 10.1096/fj.14-256503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bellner L, Martinelli L, Halilovic A, Patil K, Puri N, Dunn MW, Regan RF, Schwartzman ML. Heme oxygenase-2 deletion causes endothelial cell activation marked by oxidative stress, inflammation, and angiogenesis. The Journal of pharmacology and experimental therapeutics. 2009;331:925–932. doi: 10.1124/jpet.109.158352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bianchetti CM, Yi L, Ragsdale SW, Phillips GN., Jr. Comparison of apo- and heme-bound crystal structures of a truncated human heme oxygenase-2. The Journal of biological chemistry. 2007;282:37624–37631. doi: 10.1074/jbc.M707396200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bouamr F, Scarlata S, Carter C. Role of myristylation in HIV-1 Gag assembly. Biochemistry. 2003;42:6408–6417. doi: 10.1021/bi020692z. [DOI] [PubMed] [Google Scholar]
  8. Boutin JA. Myristoylation. Cellular signalling. 1997;9:15–35. doi: 10.1016/s0898-6568(96)00100-3. [DOI] [PubMed] [Google Scholar]
  9. Bryant M, Ratner L. Myristoylation-dependent replication and assembly of human immunodeficiency virus 1. Proceedings of the National Academy of Sciences of the United States of America. 1990;87:523–527. doi: 10.1073/pnas.87.2.523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen RJ, Yuan HH, Zhang TY, Wang ZZ, Hu AK, Wu LL, Yang ZP, Mao YJ, Ji DJ, Zhu XR. Heme oxygenase-2 suppress TNF-alpha and IL6 expression via TLR4/MyD88-dependent signaling pathway in mouse cerebral vascular endothelial cells. Molecular neurobiology. 2014;50:971–978. doi: 10.1007/s12035-014-8693-x. [DOI] [PubMed] [Google Scholar]
  11. Constantine R, Zhang H, Gerstner CD, Frederick JM, Baehr W. Uncoordinated (UNC)119: coordinating the trafficking of myristoylated proteins. Vision research. 2012;75:26–32. doi: 10.1016/j.visres.2012.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cross FR, Garber EA, Pellman D, Hanafusa H. A short sequence in the p60src N terminus is required for p60src myristylation and membrane association and for cell transformation. Molecular and cellular biology. 1984;4:1834–1842. doi: 10.1128/mcb.4.9.1834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dore S, Takahashi M, Ferris CD, Zakhary R, Hester LD, Guastella D, Snyder SH. Bilirubin, formed by activation of heme oxygenase-2, protects neurons against oxidative stress injury. Proceedings of the National Academy of Sciences of the United States of America. 1999;96:2445–2450. doi: 10.1073/pnas.96.5.2445. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dowal L, Yang W, Freeman MR, Steen H, Flaumenhaft R. Proteomic analysis of palmitoylated platelet proteins. Blood. 2011;118:e62–73. doi: 10.1182/blood-2011-05-353078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fitzgerald KA, Rowe DC, Barnes BJ, Caffrey DR, Visintin A, Latz E, Monks B, Pitha PM, Golenbock DT. LPS-TLR4 signaling to IRF-3/7 and NF-kappaB involves the toll adapters TRAM and TRIF. The Journal of experimental medicine. 2003;198:1043–1055. doi: 10.1084/jem.20031023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Giang DK, Cravatt BF. A second mammalian N-myristoyltransferase. The Journal of biological chemistry. 1998;273:6595–6598. doi: 10.1074/jbc.273.12.6595. [DOI] [PubMed] [Google Scholar]
  17. Gottlinger HG, Sodroski JG, Haseltine WA. Role of capsid precursor processing and myristoylation in morphogenesis and infectivity of human immunodeficiency virus type 1. Proceedings of the National Academy of Sciences of the United States of America. 1989;86:5781–5785. doi: 10.1073/pnas.86.15.5781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hantschel O, Nagar B, Guettler S, Kretzschmar J, Dorey K, Kuriyan J, Superti-Furga G. A myristoyl/phosphotyrosine switch regulates c-Abl. Cell. 2003;112:845–857. doi: 10.1016/s0092-8674(03)00191-0. [DOI] [PubMed] [Google Scholar]
  19. Henderson LE, Krutzsch HC, Oroszlan S. Myristyl amino-terminal acylation of murine retrovirus proteins: an unusual post-translational proteins modification. Proceedings of the National Academy of Sciences of the United States of America. 1983;80:339–343. doi: 10.1073/pnas.80.2.339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Higashide T, Inana G. Characterization of the gene for HRG4 (UNC119), a novel photoreceptor synaptic protein homologous to unc-119. Genomics. 1999;57:446–450. doi: 10.1006/geno.1999.5791. [DOI] [PubMed] [Google Scholar]
  21. Hill CP, Worthylake D, Bancroft DP, Christensen AM, Sundquist WI. Crystal structures of the trimeric human immunodeficiency virus type 1 matrix protein: implications for membrane association and assembly. Proceedings of the National Academy of Sciences of the United States of America. 1996;93:3099–3104. doi: 10.1073/pnas.93.7.3099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kagan JC, Su T, Horng T, Chow A, Akira S, Medzhitov R. TRAM couples endocytosis of Toll-like receptor 4 to the induction of interferon-beta. Nature immunology. 2008;9:361–368. doi: 10.1038/ni1569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kleinman CL, Doria M, Orecchini E, Giuliani E, Galardi S, De Jay N, Michienzi A. HIV-1 infection causes a down-regulation of genes involved in ribosome biogenesis. PloS one. 2014;9:e113908. doi: 10.1371/journal.pone.0113908. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lad L, Schuller DJ, Shimizu H, Friedman J, Li H, Ortiz de Montellano PR, Poulos TL. Comparison of the heme-free and -bound crystal structures of human heme oxygenase-1. The Journal of biological chemistry. 2003a;278:7834–7843. doi: 10.1074/jbc.M211450200. [DOI] [PubMed] [Google Scholar]
  25. Lad L, Wang J, Li H, Friedman J, Bhaskar B, Ortiz de Montellano PR, Poulos TL. Crystal structures of the ferric, ferrous, and ferrous-NO forms of the Asp140Ala mutant of human heme oxygenase-1: catalytic implications. Journal of molecular biology. 2003b;330:527–538. doi: 10.1016/s0022-2836(03)00578-3. [DOI] [PubMed] [Google Scholar]
  26. Litvak V, Ramsey SA, Rust AG, Zak DE, Kennedy KA, Lampano AE, Nykter M, Shmulevich I, Aderem A. Function of C/EBPdelta in a regulatory circuit that discriminates between transient and persistent TLR4-induced signals. Nature immunology. 2009;10:437–443. doi: 10.1038/ni.1721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Maines MD. Heme oxygenase: function, multiplicity, regulatory mechanisms, and clinical applications. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 1988;2:2557–2568. [PubMed] [Google Scholar]
  28. O'Neill LA, Golenbock D, Bowie AG. The history of Toll-like receptors - redefining innate immunity. Nature reviews Immunology. 2013;13:453–460. doi: 10.1038/nri3446. [DOI] [PubMed] [Google Scholar]
  29. Ono A, Freed EO. Binding of human immunodeficiency virus type 1 Gag to membrane: role of the matrix amino terminus. Journal of virology. 1999;73:4136–4144. doi: 10.1128/jvi.73.5.4136-4144.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ono A, Orenstein JM, Freed EO. Role of the Gag matrix domain in targeting human immunodeficiency virus type 1 assembly. Journal of virology. 2000;74:2855–2866. doi: 10.1128/jvi.74.6.2855-2866.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Pal R, Reitz MS, Jr., Tschachler E, Gallo RC, Sarngadharan MG, Veronese FD. Myristoylation of gag proteins of HIV-1 plays an important role in virus assembly. AIDS research and human retroviruses. 1990;6:721–730. doi: 10.1089/aid.1990.6.721. [DOI] [PubMed] [Google Scholar]
  32. Palmiter RD, Gagnon J, Vogt VM, Ripley S, Eisenman RN. The NH2-terminal sequence of the avian oncovirus gag precursor polyprotein (Pr76gag) Virology. 1978;91:423–433. doi: 10.1016/0042-6822(78)90388-4. [DOI] [PubMed] [Google Scholar]
  33. Patwardhan P, Resh MD. Myristoylation and membrane binding regulate c-Src stability and kinase activity. Molecular and cellular biology. 2010;30:4094–4107. doi: 10.1128/MCB.00246-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pepinsky RB, Papayannopoulos IA, Campbell S, Vogt VM. Analysis of Rous sarcoma virus Gag protein by mass spectrometry indicates trimming by host exopeptidase. Journal of virology. 1996;70:3313–3318. doi: 10.1128/jvi.70.5.3313-3318.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Prawan A, Kundu JK, Surh YJ. Molecular basis of heme oxygenase-1 induction: implications for chemoprevention and chemoprotection. Antioxidants & redox signaling. 2005;7:1688–1703. doi: 10.1089/ars.2005.7.1688. [DOI] [PubMed] [Google Scholar]
  36. Rahman MN, Vlahakis JZ, Szarek WA, Nakatsu K, Jia Z. X-ray crystal structure of human heme oxygenase-1 in complex with 1-(adamantan-1-yl)-2-(1H-imidazol-1-yl)ethanone: a common binding mode for imidazole-based heme oxygenase-1 inhibitors. Journal of medicinal chemistry. 2008;51:5943–5952. doi: 10.1021/jm800505m. [DOI] [PubMed] [Google Scholar]
  37. Reil H, Bukovsky AA, Gelderblom HR, Gottlinger HG. Efficient HIV-1 replication can occur in the absence of the viral matrix protein. The EMBO journal. 1998;17:2699–2708. doi: 10.1093/emboj/17.9.2699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rein A, McClure MR, Rice NR, Luftig RB, Schultz AM. Myristylation site in Pr65gag is essential for virus particle formation by Moloney murine leukemia virus. Proceedings of the National Academy of Sciences of the United States of America. 1986;83:7246–7250. doi: 10.1073/pnas.83.19.7246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Resh MD. A myristoyl switch regulates membrane binding of HIV-1 Gag. Proceedings of the National Academy of Sciences of the United States of America. 2004;101:417–418. doi: 10.1073/pnas.0308043101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Resh MD. Intracellular trafficking of HIV-1 Gag: how Gag interacts with cell membranes and makes viral particles. AIDS reviews. 2005;7:84–91. [PubMed] [Google Scholar]
  41. Resh MD, Ling HP. Identification of a 32K plasma membrane protein that binds to the myristylated amino-terminal sequence of p60v-src. Nature. 1990;346:84–86. doi: 10.1038/346084a0. [DOI] [PubMed] [Google Scholar]
  42. Rowe DC, McGettrick AF, Latz E, Monks BG, Gay NJ, Yamamoto M, Akira S, O'Neill LA, Fitzgerald KA, Golenbock DT. The myristoylation of TRIF-related adaptor molecule is essential for Toll-like receptor 4 signal transduction. Proceedings of the National Academy of Sciences of the United States of America. 2006;103:6299–6304. doi: 10.1073/pnas.0510041103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rusinova I, Forster S, Yu S, Kannan A, Masse M, Cumming H, Chapman R, Hertzog PJ. Interferome v2.0: an updated database of annotated interferon-regulated genes. Nucleic acids research. 2013;41:D1040–1046. doi: 10.1093/nar/gks1215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Saad JS, Loeliger E, Luncsford P, Liriano M, Tai J, Kim A, Miller J, Joshi A, Freed EO, Summers MF. Point mutations in the HIV-1 matrix protein turn off the myristyl switch. Journal of molecular biology. 2007;366:574–585. doi: 10.1016/j.jmb.2006.11.068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sacre SM, Lundberg AM, Andreakos E, Taylor C, Feldmann M, Foxwell BM. Selective use of TRAM in lipopolysaccharide (LPS) and lipoteichoic acid (LTA) induced NF-kappaB activation and cytokine production in primary human cells: TRAM is an adaptor for LPS and LTA signaling. Journal of immunology. 2007;178:2148–2154. doi: 10.4049/jimmunol.178.4.2148. [DOI] [PubMed] [Google Scholar]
  46. Schuller DJ, Wilks A, Ortiz de Montellano PR, Poulos TL. Crystal structure of human heme oxygenase-1. Nature structural biology. 1999;6:860–867. doi: 10.1038/12319. [DOI] [PubMed] [Google Scholar]
  47. Seta F, Bellner L, Rezzani R, Regan RF, Dunn MW, Abraham NG, Gronert K, Laniado-Schwartzman M. Heme oxygenase-2 is a critical determinant for execution of an acute inflammatory and reparative response. The American journal of pathology. 2006;169:1612–1623. doi: 10.2353/ajpath.2006.060555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sigal CT, Resh MD. The ADP/ATP carrier is the 32-kilodalton receptor for an NH2-terminally myristylated src peptide but not for pp60src polypeptide. Molecular and cellular biology. 1993;13:3084–3092. doi: 10.1128/mcb.13.5.3084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Swanson DA, Chang JT, Campochiaro PA, Zack DJ, Valle D. Mammalian orthologs of C. elegans unc-119 highly expressed in photoreceptors. Investigative ophthalmology & visual science. 1998;39:2085–2094. [PubMed] [Google Scholar]
  50. Thinon E, Serwa RA, Broncel M, Brannigan JA, Brassat U, Wright MH, Heal WP, Wilkinson AJ, Mann DJ, Tate EW. Global profiling of co- and post-translationally N-myristoylated proteomes in human cells. Nature communications. 2014;5:4919. doi: 10.1038/ncomms5919. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wright KJ, Baye LM, Olivier-Mason A, Mukhopadhyay S, Sang L, Kwong M, Wang W, Pretorius PR, Sheffield VC, Sengupta P, et al. An ARL3-UNC119-RP2 GTPase cycle targets myristoylated NPHP3 to the primary cilium. Genes & development. 2011;25:2347–2360. doi: 10.1101/gad.173443.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wright MH, Heal WP, Mann DJ, Tate EW. Protein myristoylation in health and disease. Journal of chemical biology. 2010;3:19–35. doi: 10.1007/s12154-009-0032-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Yamamoto M, Sato S, Hemmi H, Uematsu S, Hoshino K, Kaisho T, Takeuchi O, Takeda K, Akira S. TRAM is specifically involved in the Toll-like receptor 4-mediated MyD88-independent signaling pathway. Nature immunology. 2003;4:1144–1150. doi: 10.1038/ni986. [DOI] [PubMed] [Google Scholar]
  54. Zha J, Weiler S, Oh KJ, Wei MC, Korsmeyer SJ. Posttranslational N-myristoylation of BID as a molecular switch for targeting mitochondria and apoptosis. Science. 2000;290:1761–1765. doi: 10.1126/science.290.5497.1761. [DOI] [PubMed] [Google Scholar]
  55. Zhang H, Constantine R, Vorobiev S, Chen Y, Seetharaman J, Huang YJ, Xiao R, Montelione GT, Gerstner CD, Davis MW, et al. UNC119 is required for G protein trafficking in sensory neurons. Nature neuroscience. 2011;14:874–880. doi: 10.1038/nn.2835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhou W, Resh MD. Differential membrane binding of the human immunodeficiency virus type 1 matrix protein. Journal of virology. 1996;70:8540–8548. doi: 10.1128/jvi.70.12.8540-8548.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

NIHMS841903-supplement.pdf (2.4MB, pdf)

RESOURCES