Evidence that TNF-TNFR1-TRADD-TRAF2-RIP-TAK1-IKK pathway mediates constitutive NF-kappaB activation and proliferation in human head and neck squamous cell carcinoma.

Constitutively activated nuclear factor-kappaB (NF-kappaB) has been associated with a variety of aggressive tumor types, including head and neck squamous cell carcinoma (HNSCC); however, the mechanism of its activation is not fully understood. Therefore, we investigated the molecular pathway that mediates constitutive activation of NF-kappaB in a series of HNSCC cell lines. We confirmed that NF-kappaB was constitutively active in all HNSCC cell lines (FaDu, LICR-LON-HN5 and SCC4) examined as indicated by DNA binding, immunocytochemical localization of p65, by NF-kappaB-dependent reporter gene expression and its inhibition by dominant-negative (DN)-inhibitory subunit of NF-kappaB (IkappaBalpha), the natural inhibitor of NF-kappaB. Constitutive NF-kappaB activation in HNSCC was found to be due to constitutive activation of IkappaBalpha kinase (IKK); and this correlated with constitutive expression of phosphorylated forms of IkappaBalpha and p65 proteins. All HNSCC showed the expression of p50, p52, p100 and receptor-interacting protein; all linked with NF-kappaB activation. The expression of constitutively active NF-kappaB in HNSCC is mediated through the tumor necrosis factor (TNF) signaling pathway, as NF-kappaB reporter activity was inhibited by DN-TNF receptor-associated death domain (TRADD), DN-TNF receptor-associated factor (TRAF)2, DN-receptor-interacting protein (RIP), DN-transforming growth factor-beta-activated kinase 1 (TAK1), DN-kappa-Ras, DN-AKT and DN-IKK but not by DN-TRAF5 or DN-TRAF6. Constitutive NF-kappaB activation was also associated with the autocrine expression of TNF, TNF receptors and receptor-activator of NF-kappaB and its ligand in HNSCC cells but not interleukin (IL)-1beta. All HNSCC cell lines expressed IL-6, a NF-kappaB-regulated gene product. Furthermore, treatment of HNSCC cells with anti-TNF antibody downregulated constitutively active NF-kappaB, and this was associated with inhibition of IL-6 expression and cell proliferation. Our results clearly demonstrate that constitutive activation of NF-kappaB is mediated through the TRADD-TRAF2-RIP-TAK1-IKK pathway, making TNF a novel target in the treatment of head and neck cancer.

Activation of NF-kappaB by RANK requires tumor necrosis factor receptor-associated factor (TRAF) 6 and NF-kappaB-inducing kinase. Identification of a novel TRAF6 interaction motif.

Various members of the tumor necrosis factor (TNF) receptor superfamily activate nuclear factor kappaB (NF-kappaB) and the c-Jun N-terminal kinase (JNK) pathways through their interaction with TNF receptor-associated factors (TRAFs) and NF-kappaB-inducing kinase (NIK). We have previously shown that the cytoplasmic domain of receptor activator of NF-kappaB (RANK) interacts with TRAF2, TRAF5, and TRAF6 and that its overexpression activates NF-kappaB and JNK pathways. Through a detailed mutational analysis of the cytoplasmic domain of RANK, we demonstrate that TRAF2 and TRAF5 bind to consensus TRAF binding motifs located in the C terminus at positions 565-568 and 606-611, respectively. In contrast, TRAF6 interacts with a novel motif located between residues 340 and 358 of RANK. Furthermore, transfection experiments with RANK and its deletion mutants in human embryonic 293 cells revealed that the TRAF6-binding region (340-358), but not the TRAF2 or TRAF5-binding region, is necessary and sufficient for RANK-induced NF-kappaB activation. Moreover, a kinase mutant of NIK (NIK-KM) inhibited RANK-induced NF-kappaB activation. However, RANK-mediated JNK activation required a distal portion (427-603) of RANK containing the TRAF2-binding domain. Thus, our results indicate that RANK interacts with various TRAFs through distinct motifs and activates NF-kappaB via a novel TRAF6 interaction motif, which then activates NIK, thus leading to NF-kappaB activation, whereas RANK most likely activates JNK through a TRAF2-interacting region in RANK.

Characterization of the intracellular domain of receptor activator of NF-kappaB (RANK). Interaction with tumor necrosis factor receptor-associated factors and activation of NF-kappab and c-Jun N-terminal kinase.

Various members of the tumor necrosis factor (TNF) receptor superfamily interact directly with signaling molecules of the TNF receptor-associated factor (TRAF) family to activate nuclear factor kappaB (NF-kappaB) and the c-Jun N-terminal kinase (JNK) pathway. The receptor activator of NF-kappaB (RANK), a recently described TNF receptor family member, and its ligand, RANKL, promote survival of dendritic cells and differentiation of osteoclasts. RANK contains 383 amino acids in its intracellular domain (residues 234-616), which contain three putative TRAF-binding domains (termed I, II, and III). In this study, we set out to identify the region of RANK needed for interaction with TRAF molecules and for stimulation of NF-kappaB and JNK activity. We constructed epitope-tagged RANK (F-RANK616) and three C-terminal truncations, F-RANK330, F-RANK427, and F-RANK530, lacking 85, 188, and 285 amino acids, respectively. From this deletion analysis, we determined that TRAF2, TRAF5, and TRAF6 interact with RANK at its C-terminal 85-amino acid tail; the binding affinity appeared to be in the order of TRAF2 > TRAF5 > TRAF6. Furthermore, overexpression of RANK stimulated JNK and NF-kappaB activation. When the C-terminal tail, which is necessary for TRAF binding, was deleted, the truncated RANK receptor was still capable of stimulating JNK activity but not NF-kappaB, suggesting that interaction with TRAFs is necessary for NF-kappaB activation but not necessary for activation of the JNK pathway.

Overexpression of the p80 TNF receptor leads to TNF-dependent apoptosis, nuclear factor-kappa B activation, and c-Jun kinase activation.

Because they have distinct intracellular domains, it has been proposed that the p60 and p80 forms of the TNF receptor mediate different signals. Several signaling proteins have been isolated that associate with either the p60 or the p80 receptor. By using TNF muteins specific to the p60 and p80 receptors, we have previously shown that cytotoxicity and nuclear factor-kappa B (NF-kappa B) activation are mediated through the p60 form of the endogenous receptor. What signals are mediated through the p80 receptor is less clear. This study was an effort to answer that question. HeLa cells, which express only p60 receptors, were transfected with p80 receptor cDNA and then examined for apoptosis, NF-kappa B activation, and c-Jun kinase activation induced by TNF and by p60 or p80 receptor-specific muteins. The p80 mutein, like TNF and the p60 mutein, induced apoptosis and activation of NF-kappa B and c-Jun kinase in cells overexpressing recombinant p80 receptor but had no effect on cells expressing a high level of endogenous p80 receptor. The apoptosis mediated through the p60 receptor was also potentiated after overexpression of the p80 receptor, suggesting a synergistic relationship between the two receptors. Interestingly, Abs to the p80 receptor blocked apoptosis induced by all ligands but by itself activated NF-kappa B in the p80-transfected cells. Overall, our results show that the p80 receptor, which lacks the death domain, mediated apoptosis, NF-kappa B activation, and c-Jun kinase activation, but only when it was overexpressed, whereas endogenous p60 receptor mediated similar signals without overexpression.

Sanguinarine (pseudochelerythrine) is a potent inhibitor of NF-kappaB activation, IkappaBalpha phosphorylation, and degradation.

The nuclear factor NF-kappaB is a pleiotropic transcription factor whose activation results in inflammation, viral replication, and growth modulation. Due to its role in pathogenesis, NF-kappaB is considered a key target for drug development. In the present report we show that sanguinarine (a benzophenanthridine alkaloid), a known anti-inflammatory agent, is a potent inhibitor of NF-kappaB activation. Treatment of human myeloid ML-1a cells with tumor necrosis factor rapidly activated NF-kappaB, this activation was completely suppressed by sanguinarine in a dose- and time-dependent manner. Sanguinarine did not inhibit the binding of NF-kappaB protein to the DNA but rather inhibited the pathway leading to NF-kappaB activation. The reversal of inhibitory effects of sanguinarine by reducing agents suggests a critical sulfhydryl group is involved in NF-kappaB activation. Sanguinarine blocked the tumor necrosis factor-induced phosphorylation and degradation of IkappaBalpha, an inhibitory subunit of NF-kappaB, and inhibited translocation of p65 subunit to the nucleus. As sanguinarine also inhibited NF-kappaB activation induced by interleukin-1, phorbol ester, and okadaic acid but not that activated by hydrogen peroxide or ceramide, the pathway leading to NF-kappaB activation is likely different for different inducers. Overall, our results demonstrate that sanguinarine is a potent suppressor of NF-kappaB activation and it acts at a step prior to IkappaBalpha phosphorylation.

Inhibition of protein tyrosine phosphatases causes phosphorylation of tyrosine-331 in the p60 TNF receptor and inactivates the receptor-associated kinase.

Inhibition of protein tyrosine phosphatases blocks tumor necrosis factor (TNF)-induced growth modulation and NF-kappaB activation, both mediated primarily through the p60 TNF receptor. How inhibition of the phosphatases affects the p60 TNF receptor or the recently described receptor-associated serine/threonine kinase (p60TRAK) is not known. In this report, we show that this inhibition, when induced by pervanadate, caused the tyrosine phosphorylation of the cytoplasmic domain (CD) of the p60 receptor, as revealed by phosphoamino acid analysis. Furthermore, site-directed mutagenesis indicated that pervanadate specifically induced the phosphorylation of tyrosine-331, which is located in the death domain of the TNF receptor, a domain to which p60TRAK binds. This tyrosine residue was also phosphorylated by purified, recombinant pp60Src in vitro. Inhibition of protein tyrosine phosphatases by pervanadate also led to the inactivation of p60TRAK. In contrast, okadaic acid, a specific inhibitor of protein serine/threonine phosphatase, increased p60TRAK activity. Taken together, these results suggest that protein tyrosine phosphatases play an essential role in phosphorylation of the cytoplasmic domain of the TNF receptor and in regulation of the receptor-associated kinase, and this in turn may play a role in TNF-mediated growth modulation and NF-kappaB activation.

Early events in TNF signaling: a story of associations and dissociations.

At the cellular level, the multifunctional cytokine tumor necrosis factor (TNF) modulates growth and activates genes through various intermediates, including protein kinases, protein phosphatases, reactive oxygen intermediates, phospholipases, proteases, sphingomyelinases, and transcription factors. Unlike many cytokine receptors, however, the cytoplasmic domain (CD) of the TNF receptors lacks an intrinsic protein kinase activity and yet on interaction with ligand it phosphorylates various proteins. Although the kinetics of most of these activities differ, their interactions are coordinated through the selective interplay between the CD of the receptors and the associated proteins. A unique pathway has been identified by the ability of the TNF receptors to associate with a novel family of proteins. Two distinct families of proteins have emerged, the TNF receptor-associated factors (TRAFs) and the death domain homologues. The cloning of members of these gene families and the identification of the protein-interaction motifs found within their gene products has initiated the molecular identity of factors (TRADD, FADD/MORT, RIP, FLICE/MACH, and TRAFs) associated with both of the p60 and p80 forms of the TNF receptor and with other members of the TNF receptor superfamily. In this review, we summarize these and other TNF receptor-associated proteins and their potential roles in regulating the activation of nuclear factor-kappaB and apoptosis, two major responses activated by engagement of TNF receptors by the ligand.

The p80 TNF receptor-associated kinase (p80TRAK) associates with residues 354-397 of the p80 cytoplasmic domain: similarity to casein kinase.

The cytoplasmic domain of the p80 TNF receptor associates with a protein kinase, termed p80TRAK, that phosphorylates both the p60 and p80 TNF receptors. To determine the region of the cytoplasmic domain that is necessary for binding of p80TRAK and the region that it phosphorylates, a series of deletions of the p80 cytoplasmic domain were constructed and expressed as glutathione-S-transferase fusion proteins. These fusions were then used to examine the binding of p80TRAK derived from cellular extracts. We found that out of 174 residues (266-439) in the cytoplasmic domain of p80 receptor, 44 residues (354-397) were sufficient for binding of p80TRAK. Interestingly, this was also the region that contained the phosphorylation site for p80TRAK. Phosphoamino acid analysis of this region revealed phosphorylation primarily on serine residues. Furthermore, we found that, like p80TRAK, purified casein kinase 1 (CK1) also binds to residues 354-397 of the p80 TNF receptor and causes its phosphorylation. Additionally, the activity of p80TRAK was inhibited by CK1-7, the CK1-specific inhibitor. Thus, our results indicate that p80TRAK associates with a short stretch of approximately 44 residues located in the cytoplasmic domain of the p80 TNF receptor and that this kinase is similar to CK1.

Site-specific tyrosine phosphorylation of IkappaBalpha negatively regulates its inducible phosphorylation and degradation.

The transcription factor NF-kappaB is retained in the cytoplasm by its interaction with the inhibitory subunit known as IkappaB. Signal-induced serine phosphorylation and subsequent ubiquitination of IkappaBalpha target it for degradation by the 26 S proteasome. Recently, pervanadate, a protein-tyrosine phosphatase inhibitor, was shown to block the degradation of IkappaBalpha, thus inhibiting NF-kappaB activation. We investigated the mechanism by which pervanadate inhibits the degradation of IkappaBalpha. Western blot analysis of IkappaBalpha from tumor necrosis factor-treated cells revealed a slower migrating IkappaBalpha species that was subsequently degraded. However, pervanadate-treated cells also revealed a slower migrating species of IkappaBalpha that appeared in a time- and dose-dependent manner and was not degraded by tumor necrosis factor. The slower migrating species of IkappaBalpha from pervanadate-treated cells was tyrosine-phosphorylated as revealed by cross-reactivity with anti-phosphotyrosine antibodies, by the ability of the specific tyrosine phosphatase PTP1B to dephosphorylate it, and by phosphoamino acid analysis of IkappaBalpha immunoprecipitated from 32P-labeled cells. By site-specific mutagenesis and deletion analysis, we identified Tyr-42 on IkappaBalpha as the phosphoacceptor site. Furthermore, in an in vitro reconstitution system, tyrosine-phosphorylated IkappaBalpha was protected from degradation. Our results demonstrate that inducible phosphorylation and degradation of IkappaBalpha are negatively regulated by phosphorylation at Tyr-42, thus preventing NF-kappaB activation.

Characterization of the M(r) 65,000 lymphokine-activated killer proteins phosphorylated after tumor target binding: evidence that pp65a and pp65b are phosphorylated forms of L-plastin.

Contact between lymphokine-activated killer (LAK) cells and natural killer-resistant tumor targets SK-Mel-1 (human melanoma) or Raji (human lymphoma) stimulates phosphorylation of two M(r) 65,000 LAK proteins (pp65a and pp65b) with nearly identical isoelectric points. Phosphoamino acid analysis of pp65a and pp65b detected phosphorylation exclusively on serine residues. Phosphotyrosine could not be detected on either substrate after immunoblotting with an antiphosphotyrosine antibody, and herbimycin A treatment failed to inhibit p65 phosphorylation induced by target contact. However, phorbol myristate acetate treatment alone induced LAK pp65a and pp65b phosphorylation, suggesting phosphorylation may be mediated by protein kinase C or a protein kinase C-regulated kinase. The molecular weight and isoelectric points of pp65a and pp65b are similar to that reported for the human actin-bundling protein, L-plastin (L-fimbrin). On two-dimensional SDS-PAGE gel immunoblots, a peptide specific anti-L-plastin antiserum bound to pp65a and pp65b, suggesting that the phosphoproteins are similar or identical to L-plastin. In addition, two adjacent M(r) 65,000 LAK proteins were also detected by the antiserum and may correspond to unphosphorylated forms of L-plastin. On the basis of known properties of phosphorylated L-plastin, it is hypothesized that p65 phosphorylation in LAKs may regulate adhesion to tumor targets.

The p60 tumor necrosis factor (TNF) receptor-associated kinase (TRAK) binds residues 344-397 within the cytoplasmic domain involved in TNF signaling.

The p60 form of the tumor necrosis factor (TNF) receptor lacks motifs characteristic of tyrosine or serine/threonine protein kinases. Our recent observations have indicated that a p60 TNF receptor-associated kinase (p60-TRAK) from U-937 cells physically interacts with and causes the phosphorylation of the cytoplasmic domain of the TNF receptor. To define which region of the cytoplasmic domain is necessary for physical interaction with p60-TRAK, we constructed a series of deletions (grouped into three sets delta 1-delta 5, delta 6-delta 12, and delta 13-delta 16) of the p60 cytoplasmic domain, expressed them as glutathione S-transferase (GST) fusion proteins, and used them in affinity precipitations, followed by in vitro kinase assays. Our detailed analysis indicated that a serine-, threonine-, and proline-rich region (residues 243-274, delta 2) and the N-terminal half of the cytoplasmic domain (residues 243-323, delta 3) neither associated with p60-TRAK nor underwent phosphorylation. We found that out of 222 residues (205-426) in the cytoplasmic domain, only 54 (344-397, delta 12) were sufficient for binding p60-TRAK and for phosphorylation of the cytoplasmic domain. A region of approximately 30 residues (397-426) at the C-terminal end was found to interfere with optimal binding of the p60-TRAK activity. Thus, our results indicate that the minimal region of the cytoplasmic domain necessary for interacting with p60-TRAK and for phosphorylation resides within the domain previously reported to be needed for signaling the cytotoxic effect of TNF.

Reconstitution of nuclear factor kappa B activation induced by tumor necrosis factor requires membrane-associated components. Comparison with pathway activated by ceramide.

Tumor necrosis factor (TNF) is known to induce the activation of a nuclear transcription factor, nuclear factor kappa B (NF-kappa B), in a wide variety of cell types. The post-receptor binding events that culminate in TNF-dependent NF-kappa B activation are not understood. To dissect this pathway, we developed a reconstitution system consisting of membrane, cytosolic, and post-nuclear fractions. Our results indicate that when incubated with the post-nuclear fraction derived from TNF-untreated cells, the membrane fraction from TNF-treated cells causes the activation of NF-kappa B with kinetics similar to that observed in intact cells. Under these conditions, the cytosolic fraction has no effect. This activation is tyrosine kinase-dependent since erbstatin completely abolished the effect. Furthermore, as revealed by immunoblotting, no degradation of the inhibitory subunit of NF-kappa B was observed. In this reconstitution system, we can also demonstrate the activation of NF-kappa B by ceramide, but this activation is not tyrosine kinase-dependent. Overall, our results indicate that intermediates required for NF-kappa B activation by TNF or ceramide are membrane-bound, but the mechanism of activation by TNF is most likely different from that of ceramide.

TNF and its receptor antibody agonist differ in mediation of cellular responses.

TNF binds to two distinct receptors designated p60 and p80. Because Abs to the p60 receptor (anti-p60) can mimic TNF, we therefore compared the cellular signaling of TNF with that of anti-p60. We demonstrate both qualitative and quantitative differences between TNF and anti-p60. HepG2 cells, which express the p60 receptor, were found to be completely resistant to TNF but highly sensitive to the antiproliferative effects of anti-p60. In contrast, normal fibroblasts were found to be several fold more sensitive to TNF than to anti-p60. Several other epithelial cell lines that also express primarily the p60 receptor showed quantitative differences in mediation of cellular responses by TNF and anti-p60. The blocking of the p60 receptor by TNF had no effect on the response of HepG2 cells to anti-p60, suggesting a difference in their binding sites. Anti-p60, however, inhibited the effect of TNF on fibroblasts. Ab against the p80 receptor had no effect by itself or on the effect of TNF and anti-p60. The difference in the response to TNF and anti-p60 could not be correlated to the differences in the level of expression of p60 receptor on these cells. Furthermore, cycloheximide potentiated the TNF-mediated effect but not that mediated through anti-p60, thus also indicating a difference in the mechanism of action of these two agents. Overall, these results demonstrate that TNF and anti-p60, although both working through the p60 receptor, differ in their cellular signaling.

Physical and functional association of a serine-threonine protein kinase to the cytoplasmic domain of the p80 form of the human tumor necrosis factor receptor in human histiocytic lymphoma U-937 cells.

Tumor necrosis factor (TNF) binds two distinct cell surface receptors designated p60 and p80. Our previous studies indicate that a protein kinase from U-937 cells binds to and phosphorylates the p60 receptor. While the p80 receptor is phosphorylated in vivo, no association of a protein kinase has been described. We employed a fusion protein comprising of glutathione S-transferase and the cytoplasmic domain of the p80 receptor (GST-p80CD) to identify cellular proteins that might associate with this receptor. From 35S- and 32P-labeled cells, a protein of 59 kDa bound specifically to GST-p80CD. In vitro kinase reactions indicated that serine/threonine protein kinase activity associated with GST-p80CD and causes its phosphorylation. Additionally, a 59-kDa phosphoprotein was also identified after kinase reactions of proteins bound to GST-p80CD. This kinase activity required either Mg2+ or Mn2+ for optimal activity, and it phosphorylated myelin basic protein, histone H2B, and also the cytoplasmic domain of the p60 receptor. Treatment of cells with TNF increased the p80 receptor-associated kinase activity by 200%. In summary, our results provide evidence of a novel ligand-activated serine/threonine protein kinase that associates with the cytoplasmic domain of the p80 receptor and causes the phosphorylation of both forms of the TNF receptor. This p80 TNF receptor-associated protein and the associated kinase described here are referred to as p80-TRAP and p80-TRAK, respectively.

Identification of a protein kinase associated with the cytoplasmic domain of the p60 tumor necrosis factor receptor.

Tumor necrosis factor (TNF) has been shown to bind two distinct receptors, designated p60 and p80, with high affinity, resulting, within minutes, in phosphorylation of several proteins. The receptors themselves do not exhibit protein kinase activity nor have any associated proteins been identified. We employed the glutathione-S-transferase (GST) fusion protein system consisting of the cytoplasmic domain of p60 (GST-p60CD delta 1) as a probe to help us identify receptor-associated proteins from human histiocytic lymphoma U-937 cells. We found that a protein of approximately 52 kDa (pp52) bound to GST-p60CD delta 1 from [35S]methionine- and 32P-labeled cells. The associated protein was phosphorylated on serine and threonine residues. Furthermore, we identified serine/threonine kinase activity associated with p60CD delta 1 that required either Mn2+ or Mg2+ for optimal activity. The preferred substrates for this kinase, in addition to p60CD delta 1, included casein and histone H1, but not histone H2B, myelin basic protein, enolase, or the cytoplasmic domain of p80. As was the case in U-937 cells, p60CD delta 1-associated kinase activity was also identified in human breast adenocarcinoma MCF-7 cells and human foreskin fibroblasts. TNF stimulation of MCF-7 and foreskin fibroblasts for 5-15 min induced approximately 50 and 240% increases in phosphorylation of p60CD delta 1, respectively. Thus, our results provide the first evidence for protein kinase activity that is specifically associated with the cytoplasmic domain of the p60 form of the TNF receptor and causes its phosphorylation. This p60 TNF receptor-associated protein and the associated kinase described here are referred to as p60-TRAP and p60-TRAK, respectively.

Modulation of Syrian hamster 3-hydroxy-3-methylglutaryl-CoA reductase activity by phosphorylation. Role of serine 871.

Attenuation of Syrian hamster 3-hydroxy-3-methyl-glutaryl coenzyme A reductase (HMG-CoA reductase, EC 1.1.1.34) activity by in vitro phosphorylation was studied using AMP-activated protein kinase and wild-type and mutant forms of HMG-CoA reductase. The only residue of the wild-type enzyme phosphorylated was Ser871. Substrates protected against kinase-mediated attenuation of activity, consistent with substrate-induced conformational changes at the C-terminal region. Although close to the catalytic histidine His865, Ser871 appears to play no direct role in catalysis or substrate recognition. Mutant enzymes S871A, S871H, S871N, and S871Q exhibited from 62-106% of wild-type activity and had wild-type Km values for HMG-CoA and NADPH. Replacement of Ser871 by aspartate or glutamate, but not by glutamine, asparagine, histidine, or tyrosine, severely attenuated activity. Attenuation of catalytic activity that accompanies phosphorylation thus appears to result primarily from the introduction of negative charge, not merely steric hindrance. Other than the wild-type enzyme, only mutant enzyme S871T was phosphorylated, and phosphorylation was accompanied by attenuation of activity. The AMP-activated kinase thus can also phosphorylate threonyl residues.

Syrian hamster 3-hydroxy-3-methylglutaryl-coenzyme A reductase expressed in Escherichia coli: production of homogeneous protein.

When overexpressed in Escherichia coli, the catalytic domain of Syrian hamster 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMG-CoA reductase, EC 1.1.1.34) is catalytically active, but exhibits major heterogeneity. This heterogeneity reflects deletion of about 60 aminoacyl residues from the C-terminus, presumably a result of proteolytic cleavage or premature termination of translation. With the intent of separating the intact and truncated proteins via immunoaffinity chromatography, we constructed the expression phagemid pKFT7-21. This construct encodes the catalytic domain of Syrian hamster HMG-CoA reductase with the C-terminal extension Glu-Glu-Phe, an epitope recognized by a specific antibody. Following overexpression, the modified catalytic domain RcatEEF had high catalytic activity and exhibited no heterogeneity. It therefore was possible to purify RcatEEF to over 95% homogeneity without resorting to immunoaffinity chromatography. The yield of homogeneous protein averaged 20-25 mg per liter of cells with a final specific activity of up to 40 mumol NADPH oxidized per minute per milligram. The EEF modification thus should prove useful for the purification of the catalytic domains of other eukaryotic HMG-CoA reductases which exhibit heterogeneity.

His865 is the catalytically important histidyl residue of Syrian hamster 3-hydroxy-3-methylglutaryl-coenzyme A reductase.

Involvement in catalysis of a histidyl residue of Syrian hamster 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase was suggested by the ability of diethyl pyrocarbonate to abolish catalytic activity, accompanying spectral changes, and reactivation by hydroxylamine. The 7 histidines present in the catalytic domain of the hamster enzyme were changed to glutamine (His474, His487, His634, His751, His860, and His865), lysine (His865), or tyrosine (His868). Overexpression in Escherichia coli yielded six soluble mutant proteins, one insoluble protein (H634Q), and one which was degraded in vivo (H487Q). Following purification to homogeneity, mutant enzymes H474Q, H751Q, H860Q, and H868Y had essentially wild-type catalytic activity, while mutant enzymes H865K and H865Q had less than 0.6% wild-type activity. The low activity of mutant enzymes H865K and H865Q is unlikely to reflect altered structural integrity since both chromatographed on affinity supports like wild-type enzyme and had Km values for (S)-HMG-CoA (31 and 16 microM) and for NADPH (60 and 24 microM) close to those for wild-type enzyme (31 and 52 microM for (S)-HMG-CoA and NADPH, respectively). His865 of hamster HMG-CoA reductase, the histidine of the consensus Leu-Val-Xaa-Ser-His-Met-Xaa-Xaa-Asn-Arg-Ser motif and the only histidine conserved among the catalytic domains of all HMG-CoA reductases, thus appears to be a general acid/base functional in catalysis.