Recycling CD1d1 molecules present endogenous antigens processed in an endocytic compartment to NKT cells

Mouse CD1d1 molecules present endogenous glycolipids to NKT cells. Although glycolipid presentation requires CD1d1 transport through the endocytic pathway, the processing requirements for such endogenous Ag presentation by CD1d1 molecules are undefined. We examined CD1d1 Ag presentation to NKT cells by disrupting endocytic trafficking and function in cells expressing normal and mutated CD1d1 expressed by recombinant vaccinia viruses. Consistent with previous studies, we found that preventing CD1d1 localization to endosomes by altering its cytoplasmic targeting sequences abrogated recognition by Valpha14Jalpha281(+) NKT cells without affecting recognition by Valpha14(-) NKT cells. Increasing the pH of acidic compartments by incubating cells with chloroquine or bafilomycin A1 blocked CD1d1 recognition by Valpha14(+) (but not Valpha14(-)) NKT cells without reducing levels of cell surface CD1d1. Similar results were obtained with primaquine, which interferes with the recycling of cell surface glycoproteins. These results suggest that the loading of a subset of glycolipid ligands onto CD1d1 molecules entails the delivery of cell surface CD1d1 molecules and an acidic environment in the endocytic pathway.     

Galactose-modified iNKT cell agonists stabilized by an induced fit of CD1d prevent tumour metastasis

Invariant natural killer T (iNKT) cells are known to have marked immunomodulatory capacity due to their ability to produce copious amounts of effector cytokines. Here, we report the structure and function of a novel class of aromatic α-galactosylceramide structurally related glycolipids with marked Th1 bias in both mice and men, leading to superior tumour protection in vivo. The strength of the Th1 response correlates well with enhanced lipid binding to CD1d as a result of an induced fit mechanism that binds the aromatic substitution as a third anchor, in addition to the two lipid chains. This induced fit is in contrast to another Th1-biasing glycolipid, α-C-GalCer, whose CD1d binding follows a conventional key-lock principle. These findings highlight the previously unexploited flexibility of CD1d in accommodating galactose-modified glycolipids and broaden the range of glycolipids that can stimulate iNKT cells. We speculate that glycolipids can be designed that induce a similar fit, thereby leading to superior and more sustained iNKT cell responses in vivo.     

Structure of the M2 transmembrane segment of GLIC, a prokaryotic Cys loop receptor homologue from Gloeobacter violaceus, probed by substituted cysteine accessibility

GLIC is a homopentameric proton-gated, prokaryotic homologue of the Cys loop receptor family of neurotransmitter-gated ion channels. Recently, crystal structures of GLIC hypothesized to represent an open channel state were published. To explore the channel structure in functional GLIC channels, we tested the ability of p-chloromercuribenzenesulfonate to react with 30 individual cysteine substitution mutants in and flanking the M2 channel-lining segment in the closed state (pH 7.5) and in a submaximally activated state (pH 5.0). Nine mutants did not tolerate cysteine substitution and were not functional. From positions 10' to 27', p-chloromercuribenzenesulfonate significantly modified the currents at pH 7.5 and 5.0 in all mutants except H234C (11'), I235C (12'), V241C (18'), T243C (20'), L245C (22'), and Y250C (27'), which were not functional, except for 12'. Currents for P246C (23') and K247C (24') were only significantly altered at pH 5.0. The reaction rates were all >1000 m(-1) s(-1). The reactive residues were more accessible in the activated than in the resting state. We infer that M2 is tightly associated with the adjacent transmembrane helices at the intracellular end but is more loosely packed from 10' to the extracellular end than the x-ray structures suggest. We infer that the charge selectivity filter is in the cytoplasmic half of the channel. We also show that below pH 5.0, GLIC desensitizes on a time scale of minutes and infer that the crystal structures may represent a desensitized state.     

In vitro and in vivo anticancer activity of a synthetic glycolipid as Toll-like receptor 4 (TLR4) activator

Activation of Toll-like receptor 4 (TLR4) triggers the innate immune response and leads to the induction of adaptive immunity. TLR4 agonists are known to function as immunostimulants and exhibit promising therapeutic potential for cancer immunotherapy. We have previously developed a synthetic serine-based glycolipid (designated as CCL-34) that can activate TLR4-dependent signaling pathways. In this study, the anticancer immunity of CCL-34 was further demonstrated. CCL-34-activated macrophages induced cancer cell death via the apoptotic pathway, and this cytotoxicity was significantly inhibited by NG-monomethyl-L-arginine (an inducible NOS inhibitor). Notably, conditioned medium collected from CCL-34-treated splenocytes also induced cytotoxicity toward cancer cells. Furthermore, CCL-34 treatment suppressed tumor growth and increased the survival rate in TLR4-functional C3H/HeN mice but not in TLR4-defective C3H/HeJ mice. Increased apoptosis, the induction of cytokines (IFN-γ and IL-12) and chemokines (CXCL9 and CXCL10), and the elevation of leukocyte markers (CD11b, CD11c, CD4, and CD8) were detected at tumor sites in C3H/HeN mice but not in C3H/HeJ mice. Structure-and-activity relationship analysis of CCL-34 and its structural analogs revealed that a sugar moiety is essential for its activity. However, the substitution of the galactose in CCL-34 with glucose or fucose did not reduce its activity. Altogether, this study reveals the anticancer activity of a new synthetic TLR4 agonist and broadens the molecular basis of TLR4-activating glycolipids.     

Simplexide Induces CD1d-Dependent Cytokine and Chemokine Production from Human Monocytes

Monocytes are major effector cells of innate immunity and recognize several endogenous and exogenous molecules due to the expression of wide spectrum of receptors. Among them, the MHC class I-like molecule CD1d interacts with glycolipids and presents them to iNKT cells, mediating their activation. Simplexide belongs to a novel class of glycolipids isolated from marine sponges and is structurally distinct from other immunologically active glycolipids. In this study we have examined the effects of simplexide on cytokine and chemokine release from human monocytes. Simplexide induces a concentration- and time-dependent release of IL-6, CXCL8, TNF-α and IL-10 and increases the expression of IL6, CXCL8 and IL10 mRNA. Cytokine and chemokine release induced by simplexide from monocytes is dependent on CD1d since: i) a CD1d antagonist, 1,2-bis (diphenylphosphino) ethane [DPPE]- polyethylene glycolmonomethylether [PEG], specifically blocks simplexide-induced activation of monocytes; ii) CD1d knockdown inhibits monocyte activation by simplexide and iii) simplexide induces cytokine production from CD1d-transfected but not parental C1R cell line Finally, we have shown that simplexide also induces iNKT cell expansion in vitro. Our results demonstrate that simplexide, apart from activating iNKT cells, induces the production of cytokines and chemokines from human monocytes by direct interaction with CD1d.     

Hybrid characteristics of oligonucleotides consisting of isonucleoside 2',5'-anhydro-3'-deoxy-3'-(thymin-1-yl)-D-mannitol with different linkage modes

Oligonucleotides consisting of the isonucleoside repeating unit 2',5'-anhydro-3'-deoxy-3'-(thymin-1-yl)-D-mannitol (4) were synthesized with the monomeric unit 4 incorporated into oligonucleotides as 1'-->4' linkage 4a (oligomer I) or 6'-->4' linkage 4b (oligomer II). The hybrid properties of the two oligonucleotides I and II with their complementary strands were investigated by thermal denaturation and CD spectra. Oligonucleotide I (4a) formed a stable duplex with d(A)(14) with a slightly reduced T(m) value of 36.6 degrees C, relative to 38.2 degrees C for the control duplex d(T)(14)/d(A)(14), but oligomer II (4b) failed to hybridize with a DNA complementary single strand. The spectrum of the duplex oligomer I/d(A)(14) showed a positive CD band at 217 nm and a negative CD band at 248 nm attributable to a B-like conformation. Molecular modeling showed that in the case of oligomer I: the C6' hydroxy group of each unit could be located in the groove area when hybridized to the DNA single strand, which might contribute additional hydrogen bonding to the stability of duplex formation.     

Structural characterization of mycobacterial phosphatidylinositol mannoside binding to mouse CD1d

Mycobacterial phosphatidylinositol tetramannosides (PIM4) are agonists for a distinct population of invariant human (Valpha24) and mouse (Valpha14) NKT cells, when presented by CD1d. We determined the crystal structure at 2.6-A resolution of mouse CD1d bound to a synthetic dipalmitoyl-PIM2. Natural PIM2, which differs in its fatty acid composition is a biosynthetic precursor of PIM4, PIM6, lipomannan, and lipoarabinomannan. The PIM2 headgroup (inositol-dimannoside) is the most complex to date among all the crystallized CD1d ligands and is remarkably ordered in the CD1d binding groove. A specific hydrogen-bonding network between PIM2 and CD1d orients the headgroup in the center of the binding groove and above the A' pocket. A central cluster of hydrophilic CD1d residues (Asp(153), Thr(156), Ser(76), Arg(79)) interacts with the phosphate, inositol, and alpha1-alpha6-linked mannose of the headgroup, whereas additional specificity for the alpha1- and alpha2-linked mannose is conferred by Thr(159). The additional two mannoses in PIM4, relative to PIM2, are located at the distal 6' carbon of the alpha1-alpha6-linked mannose and would project away from the CD1d binding groove for interaction with the TCR. Compared with other CD1d-sphingolipid structures, PIM2 has an increased number of polar interactions between its headgroup and CD1, but reduced specificity for the diacylglycerol backbone. Thus, novel NKT cell agonists can be designed that focus on substitutions of the headgroup rather than on reducing lipid chain length, as in OCH and PBS-25, two potent variants of the highly stimulatory invariant NKT cell agonist alpha-galactosylceramide.     

Intersection of group I CD1 molecules and mycobacteria in different intracellular compartments of dendritic cells

Human CD1a, CD1b, and CD1c molecules can present mycobacterial glycolipids to T cells. Because phagosomes containing viable mycobacteria represent early endosomal compartments, we studied where mycobacterial glycolipids intersect with CD1 molecules in infected APC. CD1b and CD1c, but not CD1a, localized to late endosomes/lysosomes. CD1a and CD1c were predominantly expressed on the cell surface and in mycobacterial phagosomes of the early endosomal stage. In contrast, CD1b was present in a subset of mycobacterial phagosomes representing mature phagolysosomes. Released mycobacterial glycolipids including lipoarabinomannan and phosphatidylinositol mannosides were transported from the phagosome into late endosomes/lysosomes and to uninfected bystander cells. The macrophage mannose receptor, which has been implicated in glycolipid uptake by APC for CD1b-mediated presentation, was absent from mycobacterial phagosomes and may therefore not be involved in trafficking of glycolipids between phagosomes and late endosomes/lysosomes. In conclusion, all three CD1 molecules have access to mycobacteria and glycolipids thereof, but at different intracellular sites. This allows sampling by CD1a, CD1b, and CD1c of mycobacterial glycolipids from different intracellular sites of the infected cell, which has important implications for processing and presentation of such Ags during mycobacterial infections.     

Nanomapping of CD1d–glycolipid complexes on THP1 cells by using simultaneous topography and recognition imaging

CD1d molecule, a monomorphic major histocompatibility complex class I‐like molecule, presents different types of glycolipids to invariant natural killer T (iNKT) cells that play an important role in immunity to infection and tumors, as well as in regulating autoimmunity. Here, we present simultaneous topography and recognition imaging (TREC) analysis to detect density, distribution and localization of single CD1d molecules on THP1 cells that were loaded with different glycolipids. TREC was conducted using magnetically coated atomic force microscopy tips functionalized with a biotinylated iNKT cell receptor (TCR). The recognition map revealed binding sites visible as dark spots, resulting from oscillation amplitude reduction during specific binding between iNKT TCR and the CD1d–glycolipid complex. THP1 cells were pulsed with three different glycolipids (α‐GalCer, C20 and OCH12) for 4 and 16 hr. Whereas CD1d–α‐GalCer and CD1d–C20:2 complexes on cellular membrane formed smaller microdomains up to ~10 000 nm² (dimension area), OCH12 loaded CD1d complexes presented larger clusters with a dimension up to ~30 000 nm². Moreover, the smallest size of recognition spots was about 25 nm, corresponding to a single CD1d binding site. TREC successfully revealed the distribution and localization of CD1d–glycolipid complexes on THP1 cell with single molecule resolution under physiological conditions. Copyright © 2013 John Wiley & Sons, Ltd.     

The paradox of immune molecular recognition of alpha-galactosylceramide: low affinity,low specificity for CD1d,high affinity for alpha beta TCRs

CD1 resembles both class I and class II MHC but differs by the important aspect of presenting lipid/glycolipids, instead of peptides, to T cells. Biophysical studies of lipid/CD1 interactions have been limited, and kinetics of binding are in contradiction with functional studies. We have revisited this issue by designing new assays to examine the loading of CD1 with lipids. As expected for hydrophobic interactions, binding affinity was not high and had limited specificity. Lipid critical micelle concentration set the limitation to these studies. Once loaded onto CD1d, the recognition of glycolipids by alphabeta T cell receptor was studied by surface plasmon resonance using soluble Valpha14-Vbeta8.2 T cell receptors. The Valpha14 Jalpha18 chain could be paired with NK1.1 cell-derived Vbeta chain, or any Vbeta8 chain, to achieve high affinity recognition of alpha-galactosylceramide. Biophysical analysis indicated little effect of temperature or ionic strength on the binding interaction, in contrast to what has been seen in peptide/MHC-TCR studies. This suggests that there is less accommodation made by this TCR in recognizing alpha-galactosylceramide, and it can be assumed that the most rigid part of the Ag, the sugar moiety, is critical in the interaction.     

Modulation of the RNA binding and protein processing activities of poliovirus polypeptide 3CD by the viral RNA polymerase domain

To study the role of the RNA polymerase domain (3D) in the proteinase substrate recognition and RNA binding properties of poliovirus polypeptide 3CD, we generated recombinant 3C and 3CD polypeptides and purified them to near homogeneity. By using these purified proteins in in vitro cleavage assays with structural and non-structural viral polyprotein substrates, we found that 3CD processes the poliovirus structural polyprotein precursor (P1) 100 to 1000 times more efficiently than 3C processes P1. We also found that trans-cleavage of other 3CD molecules and sites within the non-structural P3 precursor is more efficiently mediated by 3CD than 3C. However, 3C and 3CD appear to be equally efficient in the processing of a non-structural polyprotein precursor, 2C3AB. Four mutated 3CD polyproteins with site-directed lesions in the 3D domain of the proteinase were analyzed for their ability to process viral polyprotein precursors and to form a ternary complex with RNA sequences encoded in the 5' terminus of the viral genome. Analysis of mutated 3CD polypeptides revealed that specific mutations within the 3D amino acid sequences of 3CD confer differential effects on 3CD activity. All four mutated 3CD proteins tested were able to process the P1 structural precursor with wild type or near wild type efficiency. However, three of the mutated enzymes demonstrated an impaired ability to process some sites within the P3 non-structural precursor, relative to wild type 3CD. One of the mutant 3CD polypeptides, 3CD-3DK127A, also displayed a defect in its ability to form a ternary ribonucleoprotein complex with poliovirus 5' RNA sequences.     

Antibody responses to glycolipid-borne carbohydrates require CD4+ T cells but not CD1 or NKT cells

Naturally occurring anti-carbohydrate antibodies play a major role in both the innate and adaptive immune responses. To elicit an anti-carbohydrate immune response, glycoproteins can be processed to glycopeptides and presented by the classical antigen-presenting molecules, major histocompatibility complex (MHC) Class I and II. In contrast, much less is known about the mechanism(s) for anti-carbohydrate responses to glycolipids, although it is generally considered that the CD1 family of cell surface proteins presents glycolipids to T cells or natural killer T (NKT) cells. Using model carbohydrate systems (isogloboside 3 and B blood group antigen), we examined the anti-carbohydrate response on glycolipids using both antibody neutralisation and knockout mouse-based experiments. These studies showed that CD4(+) T cells were required to generate antibodies to the carbohydrates expressed on glycolipids, and unexpectedly, these antibody responses were CD1d and NKT cell independent. They also did not require peptide help. These data provide new insight into glycolipid antigen recognition by the immune system and indicate the existence of a previously unrecognised population of glycolipid antigen-specific, CD1-independent, CD4(+) T cells.     

Sequence effects of single base loops in intramolecular quadruplex DNA

We have examined the properties of intramolecular G-quadruplexes in which the G3 tracts are separated by single base loops. The most stable complex contained 1',2'-dideoxyribose in all three loops, while loops containing T and C were slightly less stable (by about 2 degrees C). Quadruplexes containing loops with single A residues were less stable by 8 degrees C for each T to A substitution. These folded sequences display similar CD spectra, which are consistent with the formation of parallel stranded complexes with double-chain reversal loops. These results demonstrate that loop sequence, and not just length, affects quadruplex stability.     

Reversion of a human immunodeficiency virus type 1 integrase mutant at a second site restores enzyme function and virus infectivity.

The integration of a DNA copy of the retroviral RNA genome into the host cell genome is essential for viral replication. The virion-associated integrase protein, encoded by the 3' end of the viral pol gene, is required for integration. Stable virus-producing T-cell lines were established for replication-defective human immunodeficiency virus type 1 carrying single amino acid substitutions at conserved residues in the catalytic domain of integrase. Phenotypically reverted virus was detected 12 weeks after transfection with the integrase mutant carrying the P-109-->S mutation (P109S). Unlike the defective P109S virus, the revertant virus (designated P109SR) grew in CD4+ SupT1 cells. In addition to the Ser substitution at Pro-109, P109SR had a second substitution of Ala for Thr at position 125 in integrase. Site-directed mutagenesis was used to show that the P109S T125A genotype was responsible for the P109SR replication phenotype. The T125A substitution also rescued the in vitro enzyme activities of recombinant P109S integrase protein. P109S integrase did not display detectable 3' processing or DNA strand transfer activity, although 5 to 10% of wild-type disintegration activity was detected. P109S T125A integrase displayed nearly wild-type levels of 3' processing, DNA strand transfer, and disintegration activities, confirming that T125A is a second-site intragenic suppressor of P109S. P109S integrase ran as a large aggregate on a size exclusion column, whereas wild-type integrase ran as a monomer and P109S T125A integrase ran as a mixed population. Pro-109 and Thr-125 are not immediately adjacent in the crystal structure of the integrase catalytic domain. We suggest that the T125A substitution restores integrase function by stabilizing a structural alteration(s) induced by the P109S mutation.     

Importance of the B ring and its substitution on the alpha-glucosidase inhibitory activity of baicalein, 5,6,7-trihydroxyflavone

Hydroxychromones and B-ring-substituted 5,6,7-trihydroxyflavones were prepared to evaluate the contribution of the B ring of baicalein (5,6,7-trihydroxyflavone, 1) to its potent alpha-glucosidase inhibitory activity. Hydroxychromones, which lack 6-hydroxyl substitution, did not show any inhibitory activity, while 5,6,7-trihydroxy-2-methylchromone (5) showed high activity. Among the tested B-ring-substituted 5,6,7-trihydroxyflavones, the 4'-hydroxy-, 3',4'-dihydroxy-, and 3',4',5'-trihydroxy-substituted derivatives were found to give more activity than that of 1. The methoxy-substituted derivatives, however, showed less activity than 1. The results suggest that the B ring of 1 was not essential, although advantageous to the activity; hydroxyl substitution on the B ring of 5,6,7-trihydroxyflavones was favorable to the activity, whereas methoxyl substitution was unfavorable; at least 4'-hydroxyl substitution of 5,6,7-trihydroxyflavones was required for enhanced activity, in which the number of hydroxyl groups did not take part.     

Crystal Structures of Bovine CD1d Reveal Altered αGalCer Presentation and a Restricted A’ Pocket Unable to Bind Long-Chain Glycolipids

NKT cells play important roles in immune surveillance. They rapidly respond to pathogens by detecting microbial glycolipids when presented by the non-classical MHC I homolog CD1d. Previously, ruminants were considered to lack NKT cells due to the lack of a functional CD1D gene. However, recent data suggest that cattle express CD1d with unknown function. In an attempt to characterize the function of bovine CD1d, we assessed the lipid binding properties of recombinant Bos taurus CD1d (boCD1d) in vitro. BoCD1d is able to bind glycosphingolipids (GSLs) with fatty acid chain lengths of C₁₈, while GSLs with fatty acids of C₂₄ do not bind. Crystal structures of boCD1d bound to a short-chain C₁₂-di-sulfatide antigen, as well as short-chain C₁₆-αGalCer revealed that the Á pocket of boCD1d is restricted in size compared to that of both mouse and human CD1d, explaining the inability of long chain GSL’s to bind to boCD1d. Moreover, while di-sulfatide is presented similarly compared to the presentation of sulfatide by mouse CD1d, αGalCer is presented differently at the cell surface, due to an amino acid Asp151Asn substitution that results in loss of intimate contacts between the αGalCer headgroup and CD1d. The altered αGalCer presentation by boCD1d also explains its lack of cross-activation of mouse iNKT cells and raises the interesting question of the nature and function of bovine lipid-reactive T cells.     

Cardiolipin binds to CD1d and stimulates CD1d-restricted γδ T cells in the normal murine repertoire

Cardiolipin (CL), a major phospholipid in bacterial cell walls, is sequestered from the immune system in mammalian mitochondria and is, therefore, a potential danger signal. Based on growing evidence that phospholipids constitute natural ligands for CD1 and that CD1d-restricted T cells recognize phospholipids, we hypothesized that CD1d binds and presents CL and that T cells in the normal immune repertoire respond to CL in a CD1d-restricted manner. We determined the murine CD1d-CL crystal structure at 2.3 ? resolution and established through additional lipid loading experiments that CL, a tetra-acylated phospholipid, binds to murine CD1d with two alkyl chains buried inside the CD1d binding groove and the remaining two exposed into the solvent. We furthermore demonstrate the functional stimulatory activity of CL, showing that splenic and hepatic γδ T cells from healthy mice proliferate in vitro in response to mammalian or bacterial CL in a dose-dependent and CD1d-restricted manner, rapidly secreting the cytokines IFN-γ and RANTES. Finally, we show that hepatic γδ T cells are activated in vivo by CD1d-bearing dendritic cells that have been pulsed with CL, but not phosphatidylcholine. Together, these findings demonstrate that CD1d is able to bind and present CL to a subset of CL-responsive γδ T cells that exist in the spleen and liver of healthy mice and suggest that these cells could play a role in host responses to bacterial lipids and, potentially, self-CL. We propose that CL-responsive γδ T cells play a role in immune surveillance during infection and tissue injury.     

Analysis of the conformation and stability of human bronchial lysozyme by circular dichroism.

1. Human bronchial lysozyme was isolated from nonpurulent secretions and studied by circular dichroism (CD) spectroscopy for its conformational properties. 2. The two negative bands at 208 and 222 nm indicated that the peptide chain adopted an alpha-helical structure in physiological conditions. 3. The molecule was stable at pH 1.0 but not at pH 12.0. 4. Increasing ionic strength by adding NaCl up to 1 M did not change the CD spectra. 5. Complete unfolding of the molecule by guanidinium chloride was obtained only at the concentration of 6 M. 6. Bronchial lysozyme was also denatured by sodium dodecyl sulphate. 7. The molecule was stable when mild reduction was performed at 37 degrees C for 30 min but was completely unfolded after heating at 100 degrees C for 3 min.     

Specificity of the hepatitis C virus NS3 serine protease: effects of substitutions at the 3/4A, 4A/4B, 4B/5A, and 5A/5B cleavage sites on polyprotein processing.

Cleavage at four sites (3/4A, 4A/4B, 4B/5A, and 5A/5B) in the hepatitis C virus polyprotein requires a viral serine protease activity residing in the N-terminal one-third of the NS3 protein. Sequence comparison of the residues flanking these cleavage sites reveals conserved features including an acidic residue (Asp or Glu) at the P6 position, a Cys or Thr residue at the P1 position, and a Ser or Ala residue at the P1' position. In this study, we used site-directed mutagenesis to assess the importance of these and other residues for NS3 protease-dependent cleavages. Substitutions at the P7 to P2' positions of the 4A/4B site had varied effects on cleavage efficiency. Only Arg at the P1 position or Pro at P1' substantially blocked processing at this site. Leu was tolerated at the P1 position, whereas five other substitutions allowed various degrees of cleavage. Substitutions with positively charged or other hydrophilic residues at the P7, P3, P2, and P2' positions did not reduce cleavage efficiency. Five substitutions examined at the P6 position allowed complete cleavage, demonstrating that an acidic residue at this position is not essential. Parallel results were obtained with substrates containing an active NS3 protease domain in cis or when the protease domain was supplied in trans. Selected substitutions blocking or inhibiting cleavage at the 4A/4B site were also examined at the 3/4A, 4B/5A, and 5A/5B sites. For a given substitution, a site-dependent gradient in the degree of inhibition was observed, with a 3/4A site being least sensitive to mutagenesis, followed by the 4A/4B, 4B/5A, and 5A/5B sites. In most cases, mutations abolishing cleavage at one site did not affect processing at the other serine protease-dependent sites. However, mutations at the 3/4A site which inhibited cleavage also interfered with processing at the 4B/5A site. Finally, during the course of these studies an additional NS3 protease-dependent cleavage site has been identified in the NS4B region.