Characteristics and safety assessment of intractable proteins in genetically modified crops.

Genetically modified (GM) crops may contain newly expressed proteins that are described as "intractable". Safety assessment of these proteins may require some adaptations to the current assessment procedures. Intractable proteins are defined here as those proteins with properties that make it extremely difficult or impossible with current methods to express in heterologous systems; isolate, purify, or concentrate; quantify (due to low levels); demonstrate biological activity; or prove equivalency with plant proteins. Five classes of intractable proteins are discussed here: (1) membrane proteins, (2) signaling proteins, (3) transcription factors, (4) N-glycosylated proteins, and (5) resistance proteins (R-proteins, plant pathogen recognition proteins that activate innate immune responses). While the basic tiered weight-of-evidence approach for assessing the safety of GM crops proposed by the International Life Sciences Institute (ILSI) in 2008 is applicable to intractable proteins, new or modified methods may be required. For example, the first two steps in Tier I (hazard identification) analysis, gathering of applicable history of safe use (HOSU) information and bioinformatics analysis, do not require protein isolation. The extremely low level of expression of most intractable proteins should be taken into account while assessing safety of the intractable protein in GM crops. If Tier II (hazard characterization) analyses requiring animal feeding are judged to be necessary, alternatives to feeding high doses of pure protein may be needed. These alternatives are discussed here.

Improved protein quality in transgenic soybean expressing a de novo synthetic protein, MB-16.

To improve soybean [Glycine max (L.) Merrill] seed nutritional quality, a synthetic gene, MB-16 was introduced into the soybean genome to boost seed methionine content. MB-16, an 11 kDa de novo protein enriched in the essential amino acids (EAAs) methionine, threonine, lysine and leucine, was originally developed for expression in rumen bacteria. For efficient seed expression, constructs were designed using the soybean codon bias, with and without the KDEL ER retention sequence, and ß-conglycinin or cruciferin seed specific protein storage promoters. Homozygous lines, with single locus integrations, were identified for several transgenic events. Transgene transmission and MB-16 protein expression were confirmed to the T5 and T7 generations, respectively. Quantitative RT-PCR analysis of developing seed showed that the transcript peaked in growing seed, 5-6 mm long, remained at this peak level to the full-sized green seed and then was significantly reduced in maturing yellow seed. Transformed events carrying constructs with the rumen bacteria codon preference showed the same transcription pattern as those with the soybean codon preference, but the transcript levels were lower at each developmental stage. MB-16 protein levels, as determined by immunoblots, were highest in full-sized green seed but the protein virtually disappeared in mature seed. However, amino acid analysis of mature seed, in the best transgenic line, showed a significant increase of 16.2 and 65.9 % in methionine and cysteine, respectively, as compared to the parent. This indicates that MB-16 elevated the sulfur amino acids, improved the EAA seed profile and confirms that a de novo synthetic gene can enhance the nutritional quality of soybean.

Antibody light chain variable domains and their biophysically improved versions for human immunotherapy.

We set out to gain deeper insight into the potential of antibody light chain variable domains (VLs) as immunotherapeutics. To this end, we generated a naïve human VL phage display library and, by using a method previously shown to select for non-aggregating antibody heavy chain variable domains (VHs), we isolated a diversity of VL domains by panning the library against B cell super-antigen protein L. Eight domains representing different germline origins were shown to be non-aggregating at concentrations as high as 450 µM, indicating VL repertoires are a rich source of non-aggregating domains. In addition, the VLs demonstrated high expression yields in E. coli, protein L binding and high reversibility of thermal unfolding. A side-by-side comparison with a set of non-aggregating human VHs revealed that the VLs had similar overall profiles with respect to melting temperature (T(m)), reversibility of thermal unfolding and resistance to gastrointestinal proteases. Successful engineering of a non-canonical disulfide linkage in the core of VLs did not compromise the non-aggregation state or protein L binding properties. Furthermore, the introduced disulfide bond significantly increased their T(m)s, by 5.5-17.5 ° C, and pepsin resistance, although it somewhat reduced expression yields and subtly changed the structure of VLs. Human VLs and engineered versions may make suitable therapeutics due to their desirable biophysical features. The disulfide linkage-engineered VLs may be the preferred therapeutic format because of their higher stability, especially for oral therapy applications that necessitate high resistance to the stomach's acidic pH and pepsin.

Utility of standard pharmacopeial and nonpharmacopeial methods in distinguishing folded, unfolded, and process variant forms of interferon α-2.

Standard pharmacopeial test methods for biologics broadly focus on identity (active substance and impurities) and function (activity and toxicity). However, it is less clear which, if any, of the methods can identify a subtle change in protein therapeutics such as misfolding, unusual product-related impurities, or sequence or folding variants that may result from differences in manufacturing processes. In this study, we test the ability of standard pharmacopeia monograph methods and other common physicochemical methods (including circular dichroism spectropolarimetry, fluorescence spectroscopy, thermal denaturation, mass spectrometry, and capillary electrophoresis) to differentiate folding variants [purposely denatured interferon (IFN) α-2] and sequence variants (deliberately truncated, or truncated and chemically modified) from the IFN α-2 reference standards. The results show that the standard pharmacopeial methods are of limited utility in detecting alterations in protein structure, even when those alterations include changes in primary structure. None of the pharmacopeial methods were clear probes of higher order structure. The nonpharmacopeial methods were somewhat more successful but not a single method was able to distinguish all variants tested from the authentic standard. Taken together, the data underscore the requirement to use several different and complementary methods and stress conditions to assess primary and higher order structure when assessing the comparability in potential biosimilar protein products.

O-linked glycosylation leads to decreased thermal stability of interferon alpha 2b as measured by two orthogonal techniques.

PURPOSE: Thermal stability is considered an indication of protein fold and conformational stability. We investigate the influence of glycosylation on the thermal stability of interferon alpha 2b (IFN α-2b). METHODS: Far ultraviolet light circular dichroism spectroscopy (UV CD) and differential scanning calorimetry (DSC) were used to assess the thermal stability of the European Directorate for the Quality of Medicines IFN α-2b reference standards as well as an O-linked glycosylated IFN α-2b produced in human embryonic kidney cells. RESULTS: Assessment of thermal stability of IFN α-2b and glycosylated IFN α-2b by DSC revealed that non-glycosylated interferon (Tm=65.7 +/- 0.2°C, n=3) was more thermally stable than the glycosylated variant (Tm=63.8 C +/- 0.4°C, n=3). These observations were confirmed with far UV CD (Tm IFN α-2b=65.3 +/- 0.4°C, Tm glycosylated IFN α-2b=63.6 +/- 0.2°C, n=3). Enzymatic deglycosylation of IFN α-2b resulted in improved thermally stability when assessed with far UV CD and DSC. CONCLUSION: We demonstrate that O-linked glycosylation decreases the thermal stability of IFN α-2b compared to a non-glycosylated variant of the protein.

Excipient exchange in the comparison of preparations of the same biologic made by different manufacturing processes: an exploratory study with recombinant human growth hormone (rhGH).

Demonstrations of bio-similarity between subsequent entry (follow-on) biologics and innovator's formulated drug products may depend upon methods that either remove excipients completely or allow the exchange of excipients to give equivalent formulations. Excipient exchange through dialysis is perhaps the simplest of such methods but its use has been hotly debated. This debate, in the absence of published data, has relied largely on theoretical considerations. This study presents data that indicate that excipient exchange can allow comparisons of different formulations of the same therapeutic protein. The use of excipient exchange to and from one concentration of mannitol to another or to a mixture of glycine and mannitol was reproducibly demonstrated for recombinant human growth hormone (rhGH). We show that marketed rhGH products from several different manufacturers exhibit differences in conformational stability when compared directly. These differences, however, are shown to be the result of differences in formulation rather than in the drug substance itself and were removed through excipient exchange. The data presented, therefore, also indicate that failure to assure a common excipient background can lead to erroneous conclusions about the similarities and differences in the physico-chemical properties of two preparations of the same therapeutic protein made by different manufacturing processes.

Identification of N- and C-terminal residues involved in MAP kinase phosphatase 3 (MKP3) interdomain binding and auto-inhibition.

Interdomain binding has been shown to play an important role in the regulation of MAP kinase phosphatase 3 (MKP3), a phosphatase involved in control of ERK signalling pathways. In this study the residues in N- and C-terminal domains responsible for MKP3 interdomain binding are identified. Peptides from the N-terminal substrate-binding domain of MKP3 were assessed for their ability to bind the C-terminal catalytic domain using surface plasmon resonance. The data indicate that the residues 77-97 (the Post-KIM peptide) in the MKP3 N-terminal domain are responsible for its binding to the C-terminal catalytic domain. Residues in the C-terminal domain that might be important to interdomain binding were identified using data in the existing literature. Variants in which these residues had been altered were examined by circular dichroism and enzymatic assays to ensure retention of their structure and catalytic properties before being assessed for their ability to bind the Post-KIM peptide. The data show that glutamic acid 248, asparagine 267 and, to a lesser extent, arginine 299 are important for the interaction between the MKP3 C-terminal and the N-terminal domains. The identified residues map to a region on the surface of the C-terminal domain that appears complementary to the N-terminal domain surface defined by the Post-KIM peptide. This interdomain binding site is distinct from the substrate interaction sites.

Approach to the profiling and characterization of influenza vaccine constituents by the combined use of size-exclusion chromatography, gel electrophoresis and mass spectrometry.

A combination of separation and identification techniques was used to rapidly and reproducibly analyze influenza vaccine constituents. Size-exclusion HPLC analysis reduced significantly the complexity by providing a constituents profile according to size. Significantly, no sample treatment was required prior to analysis thus eliminating a potential source of artifacts and degradation. Distinct profiles were associated with influenza strains as well as with vaccines from different manufacturers. Samples analyzed over several years allowed evaluation of method performance and provided stability-indicating data relating to the structural integrity of separated components. Collected chromatographic peaks were identified by gel electrophoresis and MALDI/MS of tryptic digests from excised gel bands. The challenge in obtaining high quality analytical data from complex mixtures clearly demonstrated the value of separation steps prior to MS identification. The method presented here is not intended to replace existing methodology; it is intended to provide a product specific profile to be used as a rapid screen for manufacturer, year (for annual influenza vaccines), stability or counterfeit product. It is a new screening method that provides a rapid and robust indication of products which require further investigation as a result of a deviation in their characteristic profile. Until now this tool did not exist.

Influence of bovine serum albumin on the secondary structure of interferon alpha 2b as determined by far UV circular dichroism spectropolarimetry.

Many therapeutic biologics are formulated with excipients, including the protein excipient human serum albumin (HSA), to increase stability and prevent protein aggregation and adsorption onto glass vials. One biologic formulated with albumin is interferon alpha-2b (IFN alpha-2b). As is the case with other therapeutic biologics, the increased structural complexity of IFN alpha-2b compared to a small molecule drug requires that both the correct chemical structure (amino acid sequence) and also the correct secondary and tertiary structures (3 dimensional fold) be verified to assure safety and efficacy. Although numerous techniques are available to assess a biologic's primary, secondary and tertiary structures, difficulties arise when assessing higher order structure in the presence of protein excipients. In these studies far UV circular dichroism spectropolarimetry (far UV-CD) was used to determine the secondary structure of IFN alpha-2b in the presence of a protein excipient (bovine serum albumin, BSA). We demonstrated that the secondary structure of IFN alpha-2b remains mostly unchanged at a variety of BSA to IFN alpha-2b protein ratios. A significant difference in alpha helix and beta sheet content was noted when the BSA to IFN alpha-2b ratio was 5:1 (w/w), suggesting a potential conformational change in IFN alpha-2b secondary structure when BSA is in molar excess.

SARS coronavirus: unusual lability of the nucleocapsid protein.

The severe acute respiratory syndrome (SARS) is a contagious disease that killed hundreds and sickened thousands of people worldwide between November 2002 and July 2003. The nucleocapsid (N) protein of the coronavirus responsible for this disease plays a critical role in viral assembly and maturation and is of particular interest because of its potential as an antiviral target or vaccine candidate. Refolding of SARS N-protein during production and purification showed the presence of two additional protein bands by SDS-PAGE. Mass spectroscopy (MALDI, SELDI, and LC/MS) confirmed that the bands are proteolytic products of N-protein and the cleavage sites are four SR motifs in the serine-arginine-rich region-sites not suggestive of any known protease. Furthermore, results of subsequent testing for contaminating protease(s) were negative: cleavage appears to be due to inherent instability and/or autolysis. The importance of N-protein proteolysis to viral life cycle and thus to possible treatment directions are discussed.

Inhibition of mitogen-activated protein kinase phosphatase 3 activity by interdomain binding.

Mitogen-activated protein (MAP) kinase phosphatase 3 (MKP3) is a cytoplasmic dual specificity phosphatase that functions to attenuate signaling via dephosphorylation and subsequent deactivation of its substrate and allosteric regulator, extracellular signal-regulated protein kinase 2 (ERK2). Expression of MKP3 has been shown to be under the control of ERK2, thus providing an elegant feedback mechanism for regulating the rate and duration of proliferative signals. Previously published studies suggest that MKP3 might serve as a tumor suppressor; however, significantly elevated, rather than reduced, levels of this protein have been reported in early lesions. Because overexpression of this phosphatase is counterintuitive to a proposed tumor suppressor function, the observed cellular tolerance suggested a self-inactivation mechanism. Using surface plasmon resonance, we have provided direct evidence of physical interaction between the N- and C-terminal domains. Kinetic analysis using dimethyl sulfoxide to activate the C-terminal fragment in the absence of ERK2 showed that the isolated C-terminal domain had higher catalytic efficiency than the similarly activated full-length protein. Furthermore, when the isolated N-terminal domain was added to the activated C-terminal domain, a dose-dependant inhibition of catalytic activity was observed. The similarity between the K(I) and K(D) values obtained indicate that interdomain binding stabilizes the inactive conformation of the catalytic site and implies that the N-terminal domain functions as an allosteric inhibitor of phosphatase activity. Finally, we have provided evidence for oligomerization of MKP3 in pancreatic cancer cells expressing elevated levels of this phosphatase.

Glycation improves the thermostability of trypsin and chymotrypsin.

A novel glycation procedure, in vacuo glycation, was used to attach glucose covalently to the lysine residues of trypsin and chymotrypsin. Glycated trypsin and glycated chymotrypsin have greatly increased thermostability compared to the native enzymes. For example, glycated bovine trypsin, incubated at 50 degrees C and pH 8.0 for 3 h, retained more than 50% of its original activity whereas the native enzyme was inactivated under the same conditions. Similarly, after incubation at 50 degrees C and pH 8.0, glycated bovine chymotrypsin retained 45% of its original activity and the native enzyme was inactivated. Glycated porcine trypsin is exceptionally thermostable and could be used to digest native ribonuclease at 70 degrees C without the need for prior denaturation. The apparent increase in the thermal stability of the glycated proteins observed in activity measurements is also reflected by an increase in the T(m) values determined with differential scanning calorimetry (DSC) and circular dichroism (CD). The glycation does not alter the activity or specificity of these enzymes.

Cytotoxic doses of ketoconazole affect expression of a subset of hepatic genes.

Ketoconazole is a widely prescribed antifungal drug, which has also been investigated as an anticancer therapy in both clinical and pre-clinical settings. However, severe hepatic injuries were reported to be associated with the use of ketoconazole, even in patients routinely monitored for their liver functions. Several questions concerning ketoconazole-induced hepatic injury remain unanswered, including (1) does ketoconazole alter cytochrome P450 expression at the transcriptional level?, (2) what types of gene products responsible for cytotoxicity are induced by ketoconazole?, and (3) what role do the major metabolites of ketoconazole play in this pathophysiologic process? A mouse model was employed to investigate hepatic gene expression following hepatotoxic doses of ketoconazole. Hepatic gene expression was analyzed using a toxicogenomic microarray platform, which is comprised of cDNA probes generated from livers exposed to various hepatoxicants. These hepatoxicants fall into five well-studied toxicological categories: peroxisome proliferators, aryl hydrocarbon receptor agonists, noncoplanar polychlorinated biphenyls, inflammatory agents, and hypoxia-inducing agents. Nine genes encoding enzymes involved in Phase I metabolism and one Phase II enzyme (glutathione S-transferase) were found to be upregulated. Serum amyloid A (SAA1/2) and hepcidin were the only genes that were downregulated among the 2364 genes assessed. In vitro cytotoxicity and transcription analyses revealed that SAA and hepcidin are associated with the general toxicity of ketoconazole, and might be usefully explored as generalized surrogate markers of xenobiotic-induced hepatic injury. Finally, it was shown that the primary metabolite of ketoconazole (de-N-acetyl ketoconazole) is largely responsible for the hepatoxicity and the downregulation of SAA and hepcidin.

Over-expression and refolding of MAP kinase phosphatase 3.

MAP kinase phosphatase 3 (MKP3, also known as DUSP6 and PYST1) is involved in extracellular signal receptor kinase (ERK) regulation and functions as a specific phosphatase to the activated (phosphorylated) forms of ERK1 and ERK2. MKP3 displays allosteric activation, which aids in tightly regulating its function to ERK substrates, but not other related MAPKs. Due to MKP3's specificity for the ERK signaling pathway, the development of specific activators or inhibitors to the enzyme have been suggested in order to expressly influence the ERK1 and ERK2 pathways. To produce the high yields of MKP3 protein necessary for physico-chemical characterization of MKP3 and for high throughput screening of its small-molecule activators and inhibitors, we have cloned, purified and, and identified refolding conditions suitable for producing full-length, human MKP3 from Escherichia coli inclusion bodies. Furthermore, we demonstrate the use of a 96-well plate format refolding assay in which the ERK-induced activity of MKP3 is simulated by 33% DMSO. The assay allowed for rapid detection of MKP3's function following a refolding screen in the absence of ERK and thus provides quick and inexpensive testing of MKP3 activity. Following screening, the refolded product was confirmed to be correctly folded by steady-state kinetic analysis and by the CD spectroscopy-determined secondary structure content. CD data were consistent with 36% helix and 14% sheet, which compared to an expected 32.9% helix and 12.4% sheet. These data indicated that MKP3 was properly folded, making it a suitable protein for use in functional studies.

A novel cross-linked RNase A dimer with enhanced enzymatic properties.

A new cross-linked ribonuclease A (RNase A) dimer composed of monomeric units covalently linked by a single amide bond between the side-chains of Lys(66) and Glu(9) is described. The dimer was prepared in the absence of water by incubating a lyophilized preparation of RNase, sealed under vacuum, in an oven at 85 degrees C. It was determined that the in vacuo procedure does not induce any significant conformational changes to the overall structure of RNase A, yet the amide cross-link has an increased acid lability, indicating that it is exposed and conformationally strained. Examination of X-ray crystallographic structures indicates that Lys(66) and Glu(9) are not close enough for the in vacuo dimer to adopt any of the known domain-swapped conformations. Therefore, the in vacuo RNase A dimer appears to be a novel dimeric structure. The in vacuo RNase A dimer also exhibits a twofold increase in activity over monomeric RNase A on a per monomer basis. This doubling of enzymatic activity was shown using dsRNA and ssRNA as substrates. In addition to this enhanced ability to degrade RNA, the dimer is not inhibited by the cellular ribonuclease inhibitor protein (cRI).

Novel cross-linked enzyme-antibody conjugates for Western blot and ELISA.

Covalent cross-linking of enzymes to antibodies to produce immunoconjugates for Western blot analysis and ELISAs was achieved using in vacuo cross-linking methodology [Simons, B.L., King, M.C., Cyr, T., Hefford, M.A., Kaplan, H., 2002. Covalent cross-linking of proteins without chemical reagents. Protein Sci. 11, 1558.]. The advantageous feature of this methodology for producing enzyme-antibody conjugates is that the cross-linking is accomplished without the use of chemical modifying or activating reagents. This reduces the potential activity loss due to chemical modification and allows easy recovery of any free antibody or native enzyme. In vacuo cross-linking of horseradish peroxidase (HRP) to anti-rabbit immunoglobulin G (IgG) produced an enzyme-linked antibody with an improved sensitivity for antigen detection compared to a commercial conjugate prepared by conventional chemical cross-linking methods. A soluble multi-enzyme-based immunoconjugate was prepared by the in vacuo cross-linking of HRP to a high molecular weight polyglutamic acid polymer followed by the in vacuo cross-linking of a limiting amount of antibody to yield an antibody-(HRP)n-polyglutamate complex. This complex had a detection signal 100-fold greater than that of the 1:1 enzyme-antibody conjugates prepared by chemical cross-linking.

Enhancement of essential amino acid contents in crops by genetic engineering and protein design.

The importance and urgency of providing humans and animals with quality proteins are reflected in the growing scientific and industrial interest in augmenting the nutritive value of the world's protein sources. Such nutritive value is determined by the protein content in 'essential amino acids', those that cannot be synthesized de novo and that must be supplied from the diet. It is the object of this review to discuss recent advances in the genetic modification of crops that aim to provide enhanced quantities of essential amino acids.

Covalent protein immobilization on glass surfaces: application to alkaline phosphatase.

Lyophilized alkaline phosphatase (ALPase) was immobilized on aminated glass surfaces using the in vacuo cross-linking process [Simons, B.L., King, M.C., Cyr, T., Hefford, M.A., Kaplan, H., 2002. Zero-length cross-linking of lyophilized proteins. Protein Sci. 11, 1558-1564]. In this procedure, amide bonds were formed between carboxyl groups on the protein and amino groups on the glass surface. After the non-covalently attached enzyme was removed the immobilized ALPase not only retained its activity but could also be used, washed and reused at least six times without significant loss of activity. An average of 1.4+/-0.6 mg of reusable ALPase per gram of glass fibre was immobilized based on the activity of the soluble equivalent.

Thermal stability: a means to assure tertiary structure in therapeutic proteins.

To be both safe and effective, a therapeutic product must have the correct chemical structure and be free of harmful contaminants. Structure in protein therapeutic products, however, implies not only the correct sequence of amino acids (primary structure) but also the proper folding of that amino acid chain in three-dimensional space (tertiary structure). This work is part of a general strategy to develop a battery of physico-chemical methods that could give assurances of structure (and hence function) in formulated therapeutic proteins in the absence of in vivo data. It focuses on recombinant human growth hormone (rhGH), a well-characterized therapeutic protein, and examines the utility of thermodynamic parameters in assessing its tertiary structure. Resistance of solutions of formulated rhGH to thermal denaturation was followed using Fourier Transform Infrared Spectroscopy (FTIR) by observing decreases in total helicity and increases in intermolecular beta-sheet formation. Under conditions known to induce changes in the intra-molecular ionic and H-bonding patterns stabilizing the tertiary structure but not affecting the protein's secondary structure or global fold, we have observed upwards of a 12 degrees C shift in the melting temperature of the protein. Furthermore, our results indicated that the T(m) of unfolding of rhGH was sensitive to much more subtle changes in the protein structure. Thus, resistance to thermal denaturation may well be a useful means to measure structure in formulations of well-characterized therapeutic proteins.

Covalent cross-linking of proteins without chemical reagents.

A facile method for the formation of zero-length covalent cross-links between protein molecules in the lyophilized state without the use of chemical reagents has been developed. The cross-linking process is performed by simply sealing lyophilized protein under vacuum in a glass vessel and heating at 85 degrees C for 24 h. Under these conditions, approximately one-third of the total protein present becomes cross-linked, and dimer is the major product. Chemical and mass spectroscopic evidence obtained shows that zero-length cross-links are formed as a result of the condensation of interacting ammonium and carboxylate groups to form amide bonds between adjacent molecules. For the protein examined in the most detail, RNase A, the cross-linked dimer has only one amide cross-link and retains the enzymatic activity of the monomer. The in vacuo cross-linking procedure appears to be general in its applicability because five different proteins tested gave substantial cross-linking, and co-lyophilization of lysozyme and RNase A also gave a heterogeneous covalently cross-linked dimer.