A role for the thiol-dependent reductase ERp57 in the assembly of MHC class I molecules.

An important mammalian defence strategy against intracellular pathogens is the presentation of cytoplasmically derived short peptides by major histocompatibility complex (MHC) class I molecules to cytotoxic T lymphocytes. MHC class I molecules assemble in the endoplasmic reticulum (ER) with chaperones, including calnexin and calreticulin, before binding to the transporter associated with antigen processing (TAP). We show here that the thiol-dependent reductase ERp57 (also known as ER60 protease) is involved in MHC class I assembly. ERp57 co-purified with the rat TAP complex (comprising TAP1 and TAP2), and associated with MHC class I molecules at an early stage in their biosynthesis. This association was sensitive to castanospermine, which inhibits the processing of glycoproteins. Human MHC class I molecules were also found to associate with ERp57. We conclude that ERp57 is a newly identified component of the MHC class I pathway, and that it appears to interact with MHC class I molecules before they associate with TAP.

Phosphorylated nitrate reductase from spinach leaves is inhibited by 14-3-3 proteins and activated by fusicoccin.

BACKGROUND: Nitrate reductase (NR) in leaves is rapidly inactivated in the dark by a two-step mechanism in which phosphorylation of NR on the serine at position 543 (Ser543) promotes binding to nitrate reductase inhibitor protein (NIP). The eukaryotic 14-3-3 proteins bind to many mammalian signalling components (Raf-1, Bcr, phosphoinositide 3-kinase, protein kinase C, polyomavirus middle-T antigen and Cdc25), and are implicated in the timing of mitosis, DNA-damage checkpoint control, exocytosis, and activation of the plant plasma-membrane H+-ATPase by fusicoccin. Their dimeric, saddle-shaped structures support the proposal that 14-3-3 proteins are 'adaptors' linking different signalling proteins, but their precise functions are still a mystery. RESULTS: We purified NIP to homogeneity and established by means of amino-acid sequencing that it is a mixture of several 14-3-3 isoforms. Mammalian and yeast 14-3-3 proteins were just as effective as NIP at inhibiting phosphorylated NR. The sequence around Ser543, the phosphorylation site in NR, is strikingly similar to the sequences around the phosphoserine residues (Ser259 and Ser621) of mammalian Raf-1 that interact with 14-3-3 proteins. We found that NIP activity was blocked by a synthetic phosphopeptide corresponding to residues 251-266 of Raf. Fusicoccin also blocked NIP activity, and plant plasma-membrane H+-ATPases were activated by either fusicoccin, the phosphoserine259-Raf-1 peptide, or protein phosphatase 2A. CONCLUSIONS: Our findings establish that the mechanism of inactivation of NR involves the phosphorylation of Ser 543 followed by interaction with one or more plant 14-3-3 proteins. These results support the idea of a common mechanism for binding of 14-3-3 to its targets in all eukaryotes, and suggest that the phosphoserine259-Raf-1 peptide and fusicoccin may be of general use for disrupting the interaction of 14-3-3 with its target proteins. We propose that the plant plasma-membrane H+-ATPase is regulated in an analogous manner to NR-NIP, and speculate that 14-3-3 proteins provide a link between 'sensing' the activity state of NR and signalling to other cellular processes in plants.

A novel role for Rab5-GDI in ligand sequestration into clathrin-coated pits.

BACKGROUND: Clathrin-coated pits are formed at the plasma membrane by the assembly of the coat components, namely clathrin and adaptors from the cytosol. Little is known about the regulation and mechanism of this assembly process. RESULTS: We have used an in vitro assay for clathrin-coated pit assembly to identify a novel component required for the invagination of newly formed coated pits. We have purified this cytosolic component and shown it to be a complex of Rab5 and GDI (guanine-nucleotide dissociation inhibitor), that was previously demonstrated to be involved in downstream processing of endocytic vesicles. Using a combination of quantitative electron microscopy and in vitro endocytosis assays, we have demonstrated that although coat proteins and ATP are sufficient to increase the number of new coated pits at the cell surface in permeabilised cells, the Rab5-GDI complex is required for ligand sequestration into clathrin-coated pits. CONCLUSIONS: We have identified Rab5 as a critical cytosolic component required for clathrin-coated pit function. Given the well-established role of Rab5 in the fusion of endocytic vesicles with endosomes, our results suggest that recruitment of essential components of the targeting and fusion machinery is coupled to the formation of functional transport vesicles.

3-Phosphoinositide-dependent protein kinase 1 (PDK1) phosphorylates and activates the p70 S6 kinase in vivo and in vitro.

BACKGROUND: The p70 S6 kinase, an enzyme critical for cell-cycle progression through the G1 phase, is activated in vivo by insulin and mitogens through coordinate phosphorylation at multiple sites, regulated by signaling pathways, some of which depend on and some of which are independent of phosphoinositide 3-kinase (Pl 3-kinase). It is not known which protein kinases phosphorylate and activate p70. RESULTS: Co-expression of p70 with 3-phosphoinositide-dependent protein kinase 1 (PDK1), a protein kinase that has previously been shown to phosphorylate and activate protein kinase B (PKB, also known as c-Akt), resulted in strong activation of the S6 kinase in vivo. In vitro, PDK1 directly phosphorylated Thr252 in the activation loop of the p70 catalytic domain, the phosphorylation of which is stimulated by PI 3-kinase in vivo and is indispensable for p70 activity. Whereas PDK1-catalyzed phosphorylation and activation of PKB in vitro was highly dependent on the presence of phosphatidylinositol 3,4,5-trisphosphate (Ptdlns (3,4,5)P3), PDK1 catalyzed rapid phosphorylation and activation of p70 in vitro, independent of the presence of Ptdlns(3,4,5)P3. The ability of PDK1 to phosphorylate p70 Thr252 was strongly dependent on the phosphorylation of the p70 noncatalytic carboxy-terminal tail (amino acids 422-525) and of amino acid Thr412. Moreover, once Thr252 was phosphorylated, its ability to cause activation of the p70 S6 kinase was also controlled by the p70 carboxy-terminal tail and by phosphorylation of p70 Ser394, and most importantly, Thr412. The overriding determinant of the absolute p70 activity was the strong positive cooperativity between Thr252 and Thr412 phosphorylation; both sites must be phosphorylated to achieve substantial p70 activation. CONCLUSIONS: PDK1 is one of the components of the signaling pathway recruited by Pl 3-kinase for the activation of p70 S6 kinase as well as of PKB, and serves as a multifunctional effector downstream of the Pl 3-kinase.

3-Phosphoinositide-dependent protein kinase-1 (PDK1): structural and functional homology with the Drosophila DSTPK61 kinase.

BACKGROUND: The activation of protein kinase B (PKB, also known as c-Akt) is stimulated by insulin or growth factors and results from its phosphorylation at Thr308 and Ser473. We recently identified a protein kinase, termed PDK1, that phosphorylates PKB at Thr308 only in the presence of lipid vesicles containing phosphatidylinositol 3,4,5-trisphosphate (Ptdlns(3,4,5)P3) or phosphatidylinositol 3,4-bisphosphate (Ptdlns(3,4)P2). RESULTS: We have cloned and sequenced human PDK1. The 556-residue monomeric enzyme comprises a catalytic domain that is most similar to the PKA, PKB and PKC subfamily of protein kinases and a carboxy-terminal pleckstrin homology (PH) domain. The PDK1 gene is located on human chromosome 16p13.3 and is expressed ubiquitously in human tissues. Human PDK1 is homologous to the Drosophila protein kinase DSTPK61, which has been implicated in the regulation of sex differentiation, oogenesis and spermatogenesis. Expressed PDK1 and DSTPK61 phosphorylated Thr308 of PKB alpha only in the presence of Ptdlns(3,4,5)P3 or Ptdlns(3,4)P2. Overexpression of PDK1 in 293 cells activated PKB alpha and potentiated the IGF1-induced phosphorylation of PKB alpha at Thr308. Experiments in which the PH domains of either PDK1 or PKB alpha were deleted indicated that the binding of Ptdlns(3,4,5)P3 or Ptdlns(3,4)P2 to PKB alpha is required for phosphorylation and activation by PDK1. IGF1 stimulation of 293 cells did not affect the activity or phosphorylation of PDK1. CONCLUSIONS: PDK1 is likely to mediate the activation of PKB by insulin or growth factors. DSTPK61 is a Drosophila homologue of PDK1. The effect of Ptdlns(3,4,5)P3/Ptdlns(3,4)P2 in the activation of PKB alpha is at least partly substrate directed.

The mitogen-activated protein kinase signal-integrating kinase Mnk2 is a eukaryotic initiation factor 4E kinase with high levels of basal activity in mammalian cells.

The cap-binding translation initiation factor eukaryotic initiation factor 4E (eIF4E) is phosphorylated in vivo at Ser209 in response to a variety of stimuli. In this paper, we show that the mitogen-activated protein kinase (MAPK) signal-integrating kinase Mnk2 phosphorylates eIF4E at this residue. Mnk2 binds to the scaffolding protein eIF4G, and overexpression of Mnk2 results in increased phosphorylation of endogenous eIF4E, showing that it can act as an eIF4E kinase in vivo. We have identified eight phosphorylation sites in Mnk2, of which at least three potential MAPK sites are likely to be essential for Mnk2 activity. In contrast to that of Mnk1, the activity of overexpressed Mnk2 is high under control conditions and could only be reduced substantially by a combination of PD98059 and SB203580, while the activity of endogenous Mnk2 in Swiss 3T3 cells was hardly affected upon treatment with these inhibitors. These compounds did not abolish phosphorylation of eIF4E, implying that Mnk2 may mediate phosphorylation of eIF4E in Swiss 3T3 cells. In vitro phosphorylation studies show that Mnk2 is a significantly better substrate than Mnk1 for extracellular signal-regulated kinase 2 (ERK2), p38MAPKalpha, and p38MAPKbeta. Therefore, the high levels of activity of Mnk2 under several conditions may be explained by efficient activation of Mnk2 by low levels of activity of the upstream kinases. Interestingly, we found that the association of both Mnk1 and Mnk2 with eIF4G increased upon inhibition of the MAPK pathways while activation of ERK resulted in decreased binding to eIF4G. This might reflect a mechanism to ensure rapid, but transient, phosphorylation of eIF4E upon stimulation of the MAPK pathways.

Myeloid protein tyrosine phosphatase 1B (PTP1B) deficiency protects against atherosclerotic plaque formation in the ApoE-/- mouse model of atherosclerosis with alterations in IL10/AMPKα pathway.

OBJECTIVE: Cardiovascular disease (CVD) is the most prevalent cause of mortality among patients with Type 1 or Type 2 diabetes, due to accelerated atherosclerosis. Recent evidence suggests a strong link between atherosclerosis and insulin resistance due to impaired insulin receptor (IR) signaling. Moreover, inflammatory cells, in particular macrophages, play a key role in pathogenesis of atherosclerosis and insulin resistance in humans. We hypothesized that inhibiting the activity of protein tyrosine phosphatase 1B (PTP1B), the major negative regulator of the IR, specifically in macrophages, would have beneficial anti-inflammatory effects and lead to protection against atherosclerosis and CVD. METHODS: We generated novel macrophage-specific PTP1B knockout mice on atherogenic background (ApoE-/-/LysM-PTP1B). Mice were fed standard or pro-atherogenic diet, and body weight, adiposity (echoMRI), glucose homeostasis, atherosclerotic plaque development, and molecular, biochemical and targeted lipidomic eicosanoid analyses were performed. RESULTS: Myeloid-PTP1B knockout mice on atherogenic background (ApoE-/-/LysM-PTP1B) exhibited a striking improvement in glucose homeostasis, decreased circulating lipids and decreased atherosclerotic plaque lesions, in the absence of body weight/adiposity differences. This was associated with enhanced phosphorylation of aortic Akt, AMPKα and increased secretion of circulating anti-inflammatory cytokine interleukin-10 (IL-10) and prostaglandin E2 (PGE2), without measurable alterations in IR phosphorylation, suggesting a direct beneficial effect of myeloid-PTP1B targeting. CONCLUSIONS: Here we demonstrate that inhibiting the activity of PTP1B specifically in myeloid lineage cells protects against atherosclerotic plaque formation, under atherogenic conditions, in an ApoE-/- mouse model of atherosclerosis. Our findings suggest for the first time that macrophage PTP1B targeting could be a therapeutic target for atherosclerosis treatment and reduction of CVD risk.

The highly related DEAD box RNA helicases p68 and p72 exist as heterodimers in cells.

The RNA helicases p68 and p72 are highly related members of the DEAD box family of proteins, sharing 90% identity across the conserved core, and have been shown to be involved in both transcription and mRNA processing. We previously showed that these proteins co-localise in the nucleus of interphase cells. In this study we show that p68 and p72 can interact with each other and self-associate in the yeast two-hybrid system. Co-immunoprecipitation experiments confirmed that p68 and p72 can interact in the cell and indicated that these proteins preferentially exist as hetero-dimers. In addition, we show that p68 can interact with NFAR-2, a protein that is also thought to function in mRNA processing. Moreover, gel filtration analysis suggests that p68 and p72 can exist in a variety of complexes in the cell (ranging from approximately 150 to approximately 400 kDa in size), with a subset of p68 molecules being in very large complexes (>2 MDa). The potential to exist in different complexes that may contain p68 and/or p72, together with a range of other factors, would provide the potential for these proteins to interact with different RNA substrates and would be consistent with recent reports implying a wide range of functions for p68/p72.

ATR is a caffeine-sensitive, DNA-activated protein kinase with a substrate specificity distinct from DNA-PK.

ATR is a large, > 300 kDa protein containing a carboxy-terminus kinase domain related to PI-3 kinase, and is homologous to the ATM gene product in human cells and the rad3/MEC1 proteins in yeast. These proteins, together with the DNA-PK, are part of a new family of PI-3 kinase related proteins. All members of this family play important roles in checkpoints which operate to permit cell survival following many forms of DNA damage. We have expressed ATR protein in HEK293 cells and purified the protein to near-homogeneity. We show that pure ATR is a protein kinase which is activated by circular single-stranded, double-stranded or linear DNA. Thus ATR is a new member of a sub-family of PIK related kinases, founded by the DNA-PK, which are activated in the presence of DNA. Unlike DNA-PK, ATR does not appear to require Ku proteins for its activation by DNA. We show directly that, like ATM and DNA-PK, ATR phosphorylates the genome surveillance protein p53 on serine 15, a site which is up-regulated in response to DNA damage. In addition, we find that ATR has a substrate specificity similar to, but unique from, the DNA-PK in vitro, suggesting that these proteins have overlapping but distinct functions in vivo. Finally, we find that the kinase activity of ATR in the presence and absence of DNA is suppressed by caffeine, a compound which is known to induce loss of checkpoint control. Our results are consistent with the notion that ATR plays a role in monitoring DNA structure and phosphorylation of proteins involved in the DNA damage response pathways.

Regulation of c-Myb through protein phosphorylation and leucine zipper interactions.

c-Myb function is modulated in part by a negative regulation domain which encompasses a leucine zipper (LZ). When E. coli-expressed c-Myb with wild type or mutated LZ proteins are assessed for DNA binding activity, the mutant form is substantially better at DNA binding than the wild type (WT) form. In contrast, the DNA binding activity of the WT protein is increased to an equivalent level of activity of the LZ-mutant when both are expressed in rabbit reticulocyte lysates (RRL) or insect cells. The possibility that phosphorylation overcomes the negative influence of the LZ was investigated. E. coli-expressed mutant, but not wild type c-Myb proteins, were shown to be substrates for Casein Kinase II (CKII) and cAMP-dependent Protein Kinase (PKA). The phosphorylation sites for CKII and PKA were serines 11 and 12, and 8 and 116, respectively. Serines 11 and 12 were found to be phosphorylated in recombinant wild type and mutant c-Myb expressed in insect cells and DNA binding was markedly reduced following phosphatase treatment. Substitution of serines 11 and 12 with glutamic acid and alanine in E. coli-expressed Myb demonstrated that these amino terminal residues influence the negative effect on DNA binding exerted by the LZ. Collectively, these observations support the notion that phosphorylation of serines 11 and 12 positively modulate DNA binding.

Conversion of SB 203580-insensitive MAP kinase family members to drug-sensitive forms by a single amino-acid substitution.

BACKGROUND: Specific inhibitors of protein kinases have great therapeutic potential, but the molecular basis underlying their specificity is only poorly understood. We have investigated the drug SB 203580 which belongs to a class of pyridinyl imidazoles that inhibits the stress-activated protein (SAP) kinases SAPK2a/p38 and SAPK2b/p38 beta 2 but not other mitogen-activated protein kinase family members. Like inhibitors of other protein kinases, SB 203580 binds in the ATP-binding pocket of SAPK2a/p38. RESULTS: The SAP kinases SAPK1 gamma/JNK1, SAPK3 and SAPK4 are not inhibited by SB 203580, because they have methionine in the position equivalent to Thr106 in the ATP-binding region of SAPK2a/p38 and SAPK2b/p38 beta 2. Using site-directed mutagenesis of five SAP kinases and the type I and type II TGF beta receptors, we have established that for a protein kinase to be inhibited by SB 203580, the sidechain of this residue must be no larger than that of threonine. Sensitivity to inhibition by SB 203580 is greatly enhanced when the sidechain is even smaller, as in serine, alanine or glycine. Thus, the type I TGF beta receptor, which has serine at the position equivalent to Thr106 of SAPK2a/p38 and SAPK2b/p38 beta 2, is inhibited by SB 203580. CONCLUSIONS: These findings explain how drugs that target the ATP-binding site can inhibit protein kinases specifically, and show that the presence of threonine or a smaller amino acid at the position equivalent to Thr106 of SAPK2a/p38 and SAPK2b/p38 beta 2 is diagnostic of whether a protein kinase is sensitive to the pyridinyl imidazole class of inhibitor.

Identification of a regulatory phosphorylation site in the hinge 1 region of nitrate reductase from spinach (Spinacea oleracea) leaves.

Purified nitrate reductase (NR) from spinach leaves was phosphorylated in vitro by NR-inactivating kinase on Ser-543 which is located in the hinge 1 region between the molybdenum-cofactor and haem-binding domains. Phosphorylation of Ser-543 allowed NR to be inhibited by the inhibitor, NIP. Degraded NR preparations in which a proportion of the subunits had lost 45 amino acids from the N-terminus during purification could be phosphorylated by NR kinase on Ser-543, but could not subsequently be fully inhibited by NIP, suggesting a role for the N-terminal tail of NR in NIP binding.

Microcystin affinity purification of plant protein phosphatases: PP1C, PP5 and a regulatory A-subunit of PP2A.

Proteins of approximately 35, 55 and 65kDa were purified from cauliflower extracts by microcystin-Sepharose chromatography and identified by amino acid sequencing as plant forms of protein (serine/threonine) phosphatase 1 (PP1) catalytic subunit, PP5 and a regulatory A-subunit of PP2A, respectively. Peptides that corresponded both to the tetratricopeptide (TPR) repeat and catalytic domains of PP5 were identified. Similar to mammalian PP5,the casein phosphatase activity of plant PP5 was activated >10-fold by arachidonic acid, with half-maximal stimulation occurring at approximately 100 microM lipid.

Differential effects of fatty acid and phospholipid activators on the catalytic activities of a structurally novel protein kinase from rat liver.

The lipid responsiveness of the structurally unique protein kinase, referred to as PAK-1, recently isolated from rat liver [(1994) J. Biol. Chem. 269, in press], is characterised by the high sensitivity (low micromolar) of its ribosomal S6(229-239) peptide kinase activity to both cardiolipin and the cis-unsaturated fatty acids and insensitivity to phosphatidylserine. Autophosphorylation of PAK-1 exhibited even greater sensitivity (submicromolar) to cardiolipin, but was relatively less affected by phosphatidylserine. Oleate, the most potent activator of PAK-1's peptide kinase activity was relatively ineffectual with autophosphorylation. These and other unusual characteristics, including high levels of basal catalytic activities, suggest a novel mechanism of regulation distinct from that of the protein kinase Cs.

Purification of the hepatic glycogen-associated form of protein phosphatase-1 by microcystin-Sepharose affinity chromatography.

The form of protein phosphatase-1 associated with hepatic glycogen (PP1G) was purified to near homogeneity from rat liver by affinity chromatography on microcystin-Sepharose and gel-filtration. The enzyme is a heterodimer consisting of the catalytic subunit of PP1 (the alpha and beta isoforms) complexed to a 33 kDa glycogen-binding (GL) subunit. The GL subunit binds phosphorylase a with high affinity, and is responsible for the enhanced dephosphorylation of glycogen synthase by PP1G and its allosteric inhibition by phosphorylase a.

The major myosin phosphatase in skeletal muscle is a complex between the beta-isoform of protein phosphatase 1 and the MYPT2 gene product.

Myosin is dephosphorylated by distinct forms of protein phosphatase 1 (PP1) in smooth muscle and skeletal muscle that are composed of PP1 complexed to different regulatory subunits. The smooth muscle myosin phosphatase (smPP1M) has been characterised previously and is composed of PP1beta complexed to M110 and M21 subunits that enhance the dephosphorylation of smooth muscle myosin, but not skeletal muscle myosin. In contrast, the regulatory subunit(s) of skeletal muscle myosin phosphatase (skPP1M) greatly enhance(s) the dephosphorylation of skeletal muscle myosin. Here we identify a regulatory subunit of skPP1M as the product of the MYPT2 gene, a protein whose sequence is 61% identical to the M110 subunit of smPP1M. Surprisingly, the M21 subunit of smPP1M appears to be produced from the same gene that encodes MYPT2.

Purification of type 1 protein (serine/threonine) phosphatases by microcystin-Sepharose affinity chromatography.

A microcystin (MC)-Sepharose column was prepared by addition of 2-aminoethanethiol to the alpha, beta-unsaturated carbonyl of the N-methyldehydroalanine residue of MC-LR, followed by reaction of the introduced amino group with N-hydroxysuccinimide-activated CH-Sepharose. The MC-Sepharose bound protein phosphatase-1 (PP1) with high capacity and purified human PP1 gamma in one step from E. coli extracts. It was also used to purify forms of PP1 bound to myofibrils from skeletal muscle. Two of these comprised PP1 complexed to N-terminal fragments of the M-subunit which enhance its myosin phosphatase activity, while the third comprised PP1 and an N-terminal fragment of the glycogen-binding (G)-subunit.

Primary structure of protein B from Pseudomonas putida, member of a new class of 2Fe-2S ferredoxins.

The primary structure of the 2Fe-2S ferredoxin (protein B) from the benzene dioxygenase system of Pseudomonas putida strain NCIB 12190 was determined by gas-phase sequencing of the protein and its fragments. Fast atom bombardment mass spectrometry indicated a molecular mass of 11,860 Da. The sequence contained five cysteine residues, four of which would be required to coordinate the iron-sulphur cluster. The amino acid sequence determined in the present study is compared to that of a protein deduced from the DNA sequence from another strain of Pseudomonas putida. Little sequence homology was observed when protein B was compared to 2Fe-2S ferredoxins from plant and cyanobacterial sources. The novel sequence determined here suggests a new class of ferredoxin, which is consistent with the observed mid-point redox potential being significantly less negative (-155 mV) than those of the 2Fe-2S ferredoxins involved in photosynthesis (-310 to -455 mV).

Amino acid sequence and expression of the hepatic glycogen-binding (GL)-subunit of protein phosphatase-1.

A full-length cDNA encoding the putative hepatic glycogen-binding (GL) subunit of protein phosphatase-1 (PP1) was isolated from a rat liver library. The deduced amino acid sequence (284 residues, 32.6 kDa) was 23% identical (39% similar) to the N-terminal region of the glycogen-binding (GM) subunit of PP1 from striated muscle. The similarities between GM and GL were most striking between residues 63-86, 144-166 and 186-227 of human GM (approximately 40% identity), nearly all the identities with the putative yeast homologue GAC1 being located between 144-166 and 186-227. The cDNA was expressed in E. coli, and the expressed protein transformed the properties of PP1 to those characteristic of the hepatic glycogen-associated enzyme. These experiments establish that the cloned protein is GL.

Purification and characterisation of p99, a nuclear modulator of protein phosphatase 1 activity.

We have purified a form of protein phosphatase 1 (PP1) from HeLa cell nuclei, in which the phosphatase is complexed to a regulatory subunit termed p99. We report here the cloning and characterisation of the p99 component. p99 mRNA is widely expressed in human tissues and immunofluorescence analysis with anti-p99 antibodies showed a punctate nucleoplasmic staining with additional accumulations within the nucleolus. The C-terminus of p99 contains seven RGG RNA-binding motifs, followed by eleven decapeptide repeats containing six or more of the following conserved residues (GHRPHEGPGG), and finally a putative zinc finger domain. Recombinant p99 suppresses the phosphorylase phosphatase activity of PP1 by > 90% and the canonical PP1-binding motif on p99 (residues 396-401) is unusual in that the phenylalanine residue is replaced by tryptophan.