Protein crystallography and examples of its applications in medicinal chemistry

The field of protein crystallography inspires and enthrals, whether it be for the beauty and symmetry of a perfectly formed protein crystal, the unlocked secrets of a novel protein fold, or the precise atomic-level detail yielded from a protein-ligand complex. Since 1958, when the first protein structure was solved, there have been tremendous advances in all aspects of protein crystallography, from protein preparation and crystallisation through to diffraction data measurement and structure refinement. These advances have significantly reduced the time required to solve protein crystal structures, while at the same time substantially improving the quality and resolution of the resulting structures. Moreover, the technological developments have induced researchers to tackle ever more complex systems, including ribosomes and intact membrane-bound proteins, with a reasonable expectation of success. In this review, the steps involved in determining a protein crystal structure are described and the impact of recent methodological advances identified. Protein crystal structures have proved to be extraordinarily useful in medicinal chemistry research, particularly with respect to inhibitor design. The precise interaction between a drug and its receptor can be visualised at the molecular level using protein crystal structures, and this information then used to improve the complementarity and thus increase the potency and selectivity of an inhibitor. The use of protein crystal structures in receptor-based drug design is highlighted by (i) HIV protease, (ii) influenza virus neuraminidase and (iii) prostaglandin H-2-synthetase. These represent, respectively, examples of protein crystal structures that (i) influenced the design of drugs currently approved for use in the treatment of HIV infection, (ii) led to the design of compounds currently in clinical trials for the treatment of influenza infection and (iii) could enable the design of highly specific non-steroidal anti-inflammatory drugs that lack the common side-effects of this drug class.

Super-CHO - A cell line capable of autocrine growth under fully defined protein-free conditions

Chinese Hamster Ovary (CHO) cells are widely used for the large scale production of recombinant biopharmaceuticals. Growth of the CHO-K1 cell line has been demonstrated in serum-free medium containing insulin, transferrin and selenium. In an attempt to get autocrine growth in protein-free medium, DNA coding for insulin and transferrin production was transfected into CHO-K1 cells. Transferrin was expressed well, with clones secreting approximately 1000 ng/10(6)cells/24h. Insulin was poorly expressed, with rates peaking at 5 ng/10(6)cells/24h. Characterisation of the secreted insulin indicated that the CHO cells were incompletely processing the insulin molecule. Site-directed mutagenesis was used to introduce a furin (prohormone converting enzyme) recognition sequence into the insulin molecule, allowing the production of active insulin. However, the levels were still too low to support autocrine growth. Further investigations revealed insulin degrading activity (presumably due to the presence of insulin degrading enzymes) in the cytoplasm of CHO cells. To overcome these problems insulin-like growth factor I (instead of insulin) was transfected into the cells. IGF-1 was completely processed and expressed at rates greater than 500 ng/10(6)cells/24h. In this paper we report autonomous growth of the transfected CHO-K1 cell line expressing transferrin and IGF-1 in protein-free medium without the addition of exogenous growth factors. Growth rates and final cell densities of these cells were identical to that of the parent cell line CHO-K1 growing in insulin, transferrin, and selenium supplemented serum-free media.

Identification and characterization of a G protein-coupled receptor homolog encoded by murine cytomegalovirus

This report describes the identification of a murine cytomegalovirus (MCMV) G protein-coupled receptor (GCR) homolog. This open reading frame (M33) is most closely related to, and collinear with, human cytomegalovirus UL33, and homologs are also present in human herpesvirus 6 and 7 (U12 for both viruses). Conserved counterparts in the sequenced alpha- or gammaherpesviruses have not been identified to date, suggesting that these genes encode proteins which are important for the biological characteristics of betaherpesviruses. We have detected transcripts for both UL33 and M33 as early as 3 or 4 h postinfection, and these reappear at late times. In addition, we have identified N-terminal splicing for both the UL33 and M33 RNA transcripts. For both open reading frames, splicing results in the introduction of amino acids which are highly conserved among known GCRs. To characterise the function of the M33 in the natural host, two independent MCMV recombinant viruses were prepared, each of which possesses an M33 open reading frame which has been disrupted with the beta-galactosidase gene. While the recombinant M33 null viruses showed no phenotypic differences in replication from wild-type MCMV in primary mouse embryo fibroblasts in vitro, they showed severely restricted growth in the salivary glands of infected mice. These data suggest that M33 plays an important role in vivo, in particular in the dissemination to or replication in the salivary gland, and provide the first evidence for the function of a viral GCR homolog in vivo.

Reorganization of terminator DNA upon binding replication terminator protein: Implications for the functional replication fork arrest complex

Termination of DNA replication in Bacillus subtilis involves the polar arrest of replication forks by a specific complex formed between the replication terminator protein (RTP) and DNA terminator sites. While determination of the crystal structure of RTP has facilitated our understanding of how a single RTP dimer interacts with terminator DNA, additional information is required in order to understand the assembly of a functional fork arrest complex, which requires an interaction between two RTP dimers and the terminator site. In this study, we show that the conformation of the major B. subtilis DNA terminator, Terl, becomes considerably distorted upon binding RTP. Binding of the first dimer of RTP to the B site of Terl causes the DNA to become slightly unwound and bent by similar to 40 degrees. Binding of a second dimer of RTP to the A site causes the bend angle to increase to similar to 60 degrees. We have used this new data to construct two plausible models that might explain how the ternary terminator complex can block DNA replication in a polar manner, in the first model, polarity of action is a consequence of the two RTP-DNA half-sites having different conformations. These different conformations result from different RTP-DNA contacts at each half-site (due to the intrinsic asymmetry at the terminator DNA), as well as interactions (direct or indirect) between the RTP dimers on the DNA. In the second model, polar fork arrest activity is a consequence of the different affinities of RTP for the A and B sites of the terminator DNA, modulated significantly by direct or indirect interactions between the RTP dimers.

Chemical treatment of Escherichia coli .1. Extraction of intracellular protein from uninduced cells

Extraction of intracellular protein from Escherichia coli is traditionally achieved by mechanical disruption. A chemical treatment that destroys the integrity of the bacterial cell wall and could provide an alternative technique is examined in this study. Treatment with a combination of the chelating agent ethylenediaminetetraacetate (EDTA) (greater than 0.3 mM) and the chaotropic agent urea (6 M) is highly effective at releasing protein from uninduced E. coli. The 6 M urea in the presence of 3 mM EDTA can release cytoplasmic protein from both logarithmic-phase and stationary-phase E. coli cells at levels equivalent to mechanical disruption. The concentrations of the two chemical agents were the major variables affecting the maximum levels of protein release. Several minor variables and interactions were also identified. The kinetics of protein release is first order. For 2, 4, and 6 M urea with 3 mM EDTA, the time constant is approximately 2.5 min independent of urea concentration. Kinetics for 3 mM EDTA without urea is considerably slower, with a time constant of 12.3 min. (C) 1997 John Wiley & Sons, Inc.

Investigation of the relationship between protein, message and inducer concentrations in recombinant E-coli cells

Chloramphenicol acetyl transferase (CAT) protein and mRNA levels in E. coli were determined following induction of a tac::cat construct by isopropyl-beta-thiogalactopyranoside (IPTG). High cat mRNA levels did not directly reflect CAT protein levels, in either shakeflask experiments or fermentations. Furthermore, concentrations of IPTG resulting in the highest levels of expression of cat mRNA, were different to those resulting in highest levels of CAT protein. The data suggest that high transcriptional activities lead to limitations at the translational level.

Structure of TcpG, the DsbA protein folding catalyst from Vibrio cholerae

The efficient and correct folding of bacterial disulfide bonded proteins in vivo is dependent upon a class of periplasmic oxidoreductase proteins called DsbA, after the Escherichia coli enzyme. In the pathogenic bacterium Vibrio cholerae, the DsbA homolog (TcpG) is responsible for the folding, maturation and secretion of virulence factors. Mutants in which the tcpg gene has been inactivated are avirulent; they no longer produce functional colonisation pill and they no longer secrete cholera toxin. TcpG is thus a suitable target for inhibitors that could counteract the virulence of this organism, thereby preventing the symptoms of cholera. The crystal structure of oxidized TcpG (refined at a resolution of 2.1 Angstrom) serves as a starting point for the rational design of such inhibitors. As expected, TcpG has the same fold as E. coli DsbA, with which it shares similar to 40% sequence identity. Ln addition, the characteristic surface features of DsbA are present in TcpG, supporting the notion that these features play a functional role. While the overall architecture of TcpG and DsbA is similar and the surface features are retained in TcpG, there are significant differences. For example, the kinked active site helix results from a three-residue loop in DsbA, but is caused by a proline in TcpG (making TcpG more similar to thioredoxin in this respect). Furthermore, the proposed peptide binding groove of TcpG is substantially shortened compared with that of DsbA due to a six-residue deletion. Also, the hydrophobic pocket of TcpG is more shallow and the acidic patch is much less extensive than that of E. coli DsbA. The identification of the structural and surface features that are retained or are divergent in TcpG provides a useful assessment of their functional importance in these protein folding catalysts and is an important prerequisite for the design of TcpG inhibitors. (C) 1997 Academic Press Limited.

Myelin proteolipid protein expressed in COS-1 cells is targeted to actin-associated surfaces

The compact myelin sheath represents one of the largest expanses of membrane-membrane contact in the body and, in the central nervous system, requires the myelin proteolipid protein (PLP) for assembly, To determine whether the molecular properties of PLP promote membrane adhesion and direct its subcellular localization in the absence of oligodendrocyte-specific targeting mechanisms, PLP was expressed in COS-I fibroblasts, Immunofluorescence staining indicated that PUP was translated effectively, transited the rough endoplasmic reticulum and Golgi apparatus, was delivered to the cell surface, and was endocytosed, In the plasma membrane, the PLP distribution was patchy and only sporadically coincided with sites of membrane-membrane contact between PLP-expressing cells, PLP was not randomly distributed, however, but correlated closely with microfilament locations in leading edge membranes and microvilli, as demonstrated by phalloidin double labeling, Our results indicate that even in non-myelinating cells, PLP can be concentrated in membranes associated with movement and growth, and suggest possible roles for the actin cytoskeleton in PLP localization, As PLP, DM20, and the DM20-like M6 protein all associate with actin-enriched membranes, this may be a common feature of PLP/DM20 gene family members. (C) 1997 Wiley-Liss, Inc.

The myosin-I-binding protein Acan125 binds the SH3 domain and belongs to the superfamily of leucine-rich repeat proteins

The SH3 domains of src and other nonreceptor tyrosine kinases have been shown to associate with the motif PXXP, where P and X stand for proline and an unspecified amino acid, but a motif that binds to the SH3 domain of myosin has thus far not been characterized. We previously showed that the SH3 domain of Acanthamoeba myosin-IC interacts with the protein Acan125. We now report that the Acan125 protein sequence contains two tandem consensus PXXP motifs near the C terminus. To test for binding, we expressed a polypeptide, AD3p, which includes 344 residues of native C-terminal sequence and a mutant polypeptide, AD3 Delta 977-994p, which lacks the sequence RPKPVPPPRGAKPAPPPR containing both PXXP motifs. The SH3 domain of Acanthamoeba myosin-IC bound AD3p and not AD3 Delta 977-994p, showing that the PXXP motifs are required for SH3 binding. The sequence of Acan125 is related overall to a protein of unknown function coded by Caenorhabditis elegans gene K07G5.1. The K07G5.1 gene product contains a proline-rich segment similar to the SH3 binding motif found in Acan125. The aligned sequences show considerable conservation of leucines and other hydrophobic residues, including the spacing of these residues, which matches a motif for leucine-rich repeats (LRRs). LRR domains have been demonstrated to be sites for ligand binding. Having an LRR domain and an SH3-binding domain, Acan125 and the C. elegans homologue define a novel family of bifunctional binding proteins.

E7 oncoprotein of human papillomavirus type 16 expressed constitutively in the epidermis has no effect on E7-specific B- or Th-repertoires or on the immune response induced or sustained after immunization with E7 protein

A line of FVB (H-2(q)) mice transgenic for the E6/E7 open reading frames of Human Papillomavirus type 16 driven from the alpha-A crystallin promoter expresses E7 mRNA in lens and skin epithelium. E7 protein is detectable in adult skin, coinciding with the development or inflammatory skin disease, which progresses to papillomata and squamous carcinomata in some mice. By examining the outcome of parenteral immunization with E7 protein, we sought to determine whether endogenous expression of E7 in skin had induced a preexisting immune outcome, i.e., specific immunity or tolerance, or whether the mice remain naive (''ignorant'') to E7. Our data show that the antibody response to defined E7 B-epitopes, the proliferative response to Th epitopes, and the delayed-type hypersensitivity (DTH) response to whole E7 did not differ between groups or young and old E6/E7 transgenic mice (likely having different degrees of lifetime exposure to E7 protein) or between E6/E7-transgenic and nontransgenic parental strain control mice. Although an E7-specific CTL response could not be induced in the H-2(q) background of these mice, incorporation of a D-b allele into the genome allowed comparison of D-b-restricted CTL responses in E6/E7 transgenic and nontransgenic mice. Experiments indicated that the E7-immunization-induced CTL response did not differ significantly between E6/E7 transgenic and nontransgenic mice. We interpret these results to indicate that in spite of expression of E7 protein in adult skin, E6/E7 transgenic mice remain immunologically naive (ignorant) of E7 epitopes presented by immunization. (C) 1997 Academic Press.

Growth hormone-binding protein in normal and pathologic gestation: Correlations with maternal diabetes and fetal growth

To date, measurements of GH-binding protein (GHBP) during human pregnancy have been carried out using;assays susceptible to interference by the elevated levels of human placental GH typical of late gestation. We recruited a large cohort of pregnant women (n = 140) for serial measurements of GHBP and used the ligand immunofunctional assay for GHBP. For normal gravidas, GHBP levels fell throughout gestation. Mean levels were 1.07 nmol/L (SE = 0.18) in the first trimester, 0.90 nmol/L (SE = 0.08) at 18-20 weeks, 0.73 nmol/L (SE = 0.05) at 28-30 weeks, and 0.62 nmol/L (SE = 0.06) at 36-38 weeks. GHBP levels in the first trimester correlated significantly with maternal body mass index (r = 0.58; P < 0.01). GHBP levels in pregnancies complicated by noninsulin-dependent diabetes mellitus (NIDDM) were substantially elevated at all gestational ages. The mean value in the first quarter (2.29 nmol/L) was more than double the normal mean (P < 0.01). In contrast, patients with insulin-dependent diabetes mellitus (IDDM) showed reduced GHBP concentrations at 36-38 weeks. The correlation between body mass index and GHBP is consistent with a metabolic role for GHBP during pregnancy, as is the dramatic elevation in GHBP observed in cases of NIDDM. At 36 weeks gestation, GHBP was significantly elevated (P < 0.01) in those women whose neonates had low birth weight (

Purification by reflux electrophoresis of whey proteins and of a recombinant protein expressed in Dictyostelium discoideum

Protein purification that combines the use of molecular mass exclusion membranes with electrophoresis is particularly powerful as it uses properties inherent to both techniques. The use of membranes allows efficient processing and is easily scaled up, while electrophoresis permits high resolution separation under mild conditions. The Gradiflow apparatus combines these two technologies as it uses polyacrylamide membranes to influence electrokinetic separations. The reflux electrophoresis process consists of a series of cycles incorporating a forward phase and a reverse phase. The forward phase involves collection of a target protein that passes through a separation membrane before trailing proteins in the same solution. The forward phase is repeated following clearance of the membrane in the reverse phase by reversing the current. We have devised a strategy to establish optimal reflux separation parameters, where membranes are chosen for a particular operating range and protein transfer is monitored at different pH values. In addition, forward and reverse phase times are determined during this process. Two examples of the reflux method are described. In the first case, we describe the purification strategy for proteins from a complex mixture which contains proteins of higher electrophoretic mobility than the target protein. This is a two-step procedure, where first proteins of higher mobility than the target protein are removed from the solution by a series of reflux cycles, so that the target protein remains as the leading fraction. In the second step the target protein is collected, as it has become the leading fraction of the remaining proteins. In the second example we report the development of a reflux strategy which allowed a rapid one-step preparative purification of a recombinant protein, expressed in Dictyostelium discoideum. These strategies demonstrate that the Gradiflow is amenable to a wide range of applications, as the protein of interest is not necessarily required to be the leading fraction in solution. (C) 1997 Elsevier Science B.V.