Search

The purpose of this discovery phase study was to identify candidate protein biomarkers for high-grade dysplastic cervical cells using mass spectrometry. Laser Capture Microdissection (LCM) was utilized to isolate high-grade dysplastic and normal cells from ThinPrep slides prepared from cervical cytological specimens. Following cell capture, samples were solubilized and proteins separated by gel electrophoresis in preparation for enzymatic digestion and liquid chromatography mass spectrometry analysis (LC-MS). Processed samples were subsequently analyzed using a linear ion trap coupled with a Fourier transform mass spectrometer (LTQ-FT MS). It was determined that both PreservCyt Solution and ThinPrep Pap Stain (Cytyc Corporation) were compatible with the sample processing and LC-MS analysis. In total, from 9 normal and 9 abnormal cervical cytological specimens, more than 1000 unique proteins were identified with high confidence, based on approximately 12,000 captured cells per specimen. Quantitative protein differences between HSIL (High-Grade Squamous Intraepithelial Lesion) and NILM (Negative for Intraepithelial Lesions or Malignancy) samples were determined by comparing the intensities of the representative (label-free) peptide ions. More than 200 proteins were found to exhibit a 3-fold difference in protein level. Interestingly, significant up-regulation of nuclear and mitochondrial proteins in HSIL specimens was noted. In several cases, the increased protein abundance observed in high-grade cells, as determined by quantitative LC-MS, was validated by immunocytochemical methods using ThinPrep cervical specimens. With the study of additional clinical specimens, the differential abundance of proteins in high-grade dysplastic cells versus morphologically normal cervical cells may lead to validated novel biomarkers for cervical disease.

Following on our recent work, on-line one-dimensional (1D) and two-dimensional (2D) porous layer open tubular/liquid chromatography-electrospray ionization-mass spectrometry (PLOT/LC-ESI-MS) platforms using 3.2 mx10 microm i.d. poly(styrene-divinylbenzene) (PS-DVB) PLOT columns have been developed to provide robust, high-performance, and ultrasensitive proteomic analysis. With the use of a PicoClear tee, the dead volume connection between a 50 microm i.d. PS-DVB monolithic micro-SPE column and the PLOT column was minimized. The micro-SPE/PLOT column assembly provided a separation performance similar to that obtained with direct injection onto the PLOT column at a mobile phase flow rate of 20 nL/min. The trace analysis potential of the platform was evaluated using an in-gel tryptic digest sample of a gel fraction (15-40 kDa) of a cervical cancer (SiHa) cell line. As an example of the sensitivity of the system, approximately 2.5 ng of protein in 2 microL of solution, an amount corresponding to 20 SiHa cells, was subjected to on-line micro-SPE-PLOT/LC-ESI-MS/MS analysis using a linear ion trap MS. A total of 237 peptides associated with 163 unique proteins were identified from a single analysis when using stringent criteria associated with a false positive rate of less than 1%. The number of identified peptides and proteins increased to 638 and 343, respectively, as the injection amount was raised to approximately 45 ng of protein, an amount corresponding to 350 SiHa cells. In comparison, only 338 peptides and 231 unique proteins were identified (false positive rate again less than 1%) from 750 ng of protein from the identical gel fraction, an amount corresponding to 6000 SiHa cells, using a typical 15 cmx75 microm i.d. packed capillary column. The greater sensitivity, higher recovery, and higher resolving power of the PLOT column resulted in the increased number of identifications from only approximately 5% of the injected sample amount. The resolving power of the micro-SPE/PLOT assembly was further extended by 2D chromatography via combination of the high-efficiency reversed-phase PLOT column with strong cation-exchange chromatography (SCX). As an example, 1071 peptides associated with 536 unique proteins were identified from 75 ng of protein from the same gel fraction, an amount corresponding to 600 cells, using five ion-exchange fractions in on-line 2D SCX-PLOT/LC-MS. The 2D system, implemented in an automated format, led to simple and robust operation for proteomic analysis. These promising results demonstrate the potential of the PLOT column for ultratrace analysis.

This paper describes approaches to optimize the chromatographic performance for our recently developed LC-MS platform, extended range proteomic analysis (ERPA), for comprehensive protein characterization at the ultratrace level. Large digested peptide fragments up to 10 kDa (e.g., from lysyl endopeptidase digestion) with or without modifications were well separated with high resolution using narrow bore (20 and 50 microm I.D.) poly(styrene-divinylbenzene) (PS-DVB) monolithic columns constructed by in situ solution polymerization. Importantly, the macroporous structure of the monolithic columns facilitated mass transport of large peptides with improved recovery relative to small pore size reversed-phase packings. High sequence coverage (95%), including identification of phosphorylated and glycosylated particles was achieved for beta-casein and epidermal growth factor receptor (EGFR) at the 4 and 20 fmol levels per injection, respectively, using the 20 microm I.D. PS-DVB monolithic column. For peptides with greater ionization efficiency, the detection limit could be lowered to approximately 400 zmol. Typically, the separation system produced a peak capacity of approximately 200 for a 10 cm column. This paper demonstrates that narrow-bore monolithic columns are suitable for high sensitivity and high-resolution separation of large peptide fragments by LC-MS analysis.

The CD3ϵγ and CD3ϵδ heterodimers along with the CD3ζζ homodimer are the signaling components of the T cell receptor (TCR). These invariant dimers are non-covalently associated on the T cell plasma membrane with a clone-specific (i.e. clonotypic) αβ heterodimer that binds its cognate ligand, a complex between a particular antigenic peptide, and an MHC molecule (pMHC). These four TCR dimers exist in a 1:1:1:1 stoichiometry. At the junction between the extracellular and transmembrane domains of each mammalian CD3ϵ, CD3γ, and CD3δ subunit is a highly conserved CXXC motif previously found to be important for thymocyte and T cell activation. The redox state of each CXXC motif is presently unknown. Here we show using LC-MS and a biotin switch assay that these CXXC segments are constitutively oxidized on resting and activated T cells, consistent with their measured reduction potential. NMR chemical shift perturbation experiments comparing a native oxidized CD3δ CXXC-containing segment with that of a mutant SXXS-containing CD3δ segment in LPPG (1-palmitoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (sodium salt)) micelles show extensive chemical shift differences in residues within the membrane-proximal motif as well as throughout the transmembrane and cytoplasmic domains as a result of the elimination of the native disulfide. Likewise, direct comparison of the native CD3δ segment in oxidizing and reducing conditions reveals numerous spectral differences. The oxidized CXXC maintains the structure within the membrane-proximal stalk region as well as that of its contiguous transmembrane and cytoplasmic domain, inclusive of the ITAM (immunoreceptor tyrosine-based activation motif) involved in signaling. These results suggest that preservation of the CD3 CXXC oxidized state may be essential for TCR mechanotransduction.

Glycation of interferon-beta-1b and human serum albumin was identified in a lyophilized glucose formulation by a sensitive LC-MS approach. The extent of glycation was measured by a label-free quantitation strategy. The glycation sites were determined by the accurate mass (FTICR MS) with MS/MS measurements on the corresponding tryptic peptides. The extent of glycation was measured by the ratio of the peak intensity between the glycated and the average value for three non-glycated peptides in the same run. Residues lysine 18 of interferon-beta-1b, and lysine 51, lysine 233, and lysine 545 of human serum albumin were more prone to be glycated than other sites in this lyophilized glucose formulation. Residues of lysine 51 and lysine 233 but not lysine 545 of human serum albumin are highly accessible to solvent as found in a solution storage study by Lapolla et al. The extent of glycation of both proteins and the number of glycation sites of human serum albumin were increased with the storage time at 25 degrees C. In total, two glycation sites of interferon beta-1b and 17 glycation sites of human serum albumin were identified in the lyophilized glucose formulation with a storage time at 25 degrees C of 35 days. Among the 17 glycation sites, only lysine 525 of human serum albumin has been found in vivo in diabetic patients by Shaklai et al. As expected, there was no glycation found on both interferon-beta-1b and human serum albumin in the control samples (similar lyophilized formulation but using mannitol instead of glucose).

In this publication, we will describe the combination of lectin affinity chromatography with nano high performance liquid chromatography (HPLC) coupled to a linear ion trap Fourier transform mass spectrometer (capillary LC-LTQ/FTMS) to characterize N-linked glycosylation structures in human plasma proteins. We used a well-characterized glycoprotein, tissue plasminogen activator (rt-PA), which is present at low levels in blood, as a standard to determine the dynamic range of this approach. N-linked glycopeptides derived from rt-PA could be characterized at a ratio of 1:200 in human plasma (rtPA: total plasma protein, w/w) by accurate mass measurement in the FTMS and fragmentation (MS n ) in the linear ion trap. We demonstrated that this platform has the potential to characterize the general N-linked glycosylation structures of abundant glycoproteins present in human plasma without the requirement for antibody-based purification, or additional carbohydrate analytical protocols. This conclusion was supported by the determination of carbohydrate structures for three glycoproteins, IgG, haptoglobin, and alpha-1-acid glycoprotein, at their natural levels in a human plasma sample, but only after the lectin enrichment step.

We have developed a new and sensitive LC-MS platform, Extended Range Proteomic Analysis (ERPA), which is able to achieve very high sequence coverage and comprehensive characterization of post-translational modifications in complex proteins. This new platform provides advantages of both the top-down and bottom-up proteomic approaches by combining (i) digestion of the protein with an enzyme, such as Lys-C, which cuts less frequently than trypsin, leading to on average a higher molecular weight peptide size, (ii) high-performance LC separation of the resulting fragments, (iii) a new data acquisition strategy using the LTQ-FTMS, a hybrid mass spectrometer that couples a linear ion trap with a Fourier transform ion cyclotron resonance (FTICR) cell, for analysis of peptides in the range of 0.5 to 10 kDa, and (iv) new data analysis methods for assigning large peptide structures and determining the site of attachment of post-translational modifications as well as structural features from the accurate precursor mass together with MS(2) and MS(3) fragmentations. The LC retention of the Lys-C fragments is increased, relative to a tryptic digest, due to the generally greater hydrophobicity of the larger peptides, a result that is particularly important for peptides containing hydrophilic modifications such as glycosylation and phosphorylation. Furthermore, additional positively charged arginine and lysine residues in the Lys-C fragments enhance the sensitivity of the post-translationally modified phospho- and glycopeptides by at least 10-fold relative to tryptic fragments. In typical operation, the FTICR cell provides a survey scan with the high mass resolution (100 000) and accurate mass (2 ppm) to characterize the higher charge-state precursor ions of the larger peptides. In parallel, the linear ion trap provides MS(2) and MS(3) fragmentation spectra, with a scan speed sufficiently fast for on-line LC-MS. Together, these data provide multiple means to determine or enhance the confidence of assignment of large or complicated peptide. Using ERPA, we demonstrate95% sequence coverage in the analysis of two heavily phosphorylated and glycosylated proteins, beta-casein at the 50 fmole level and the epidermal growth factor receptor (EGFR) at the 1 pmole level. In summary, the combination of digestion strategy, high-performance separation, and the hybrid LTQ-FTMS instrument enables comprehensive characterization of large proteins, including posttranslational modifications.

The outer membrane protein B (OmpB) of the typhus group rickettsiae is an immunodominant antigen and has been shown to provide protection against typhus in animal models. Consequently, OmpB is currently being considered as a potential rickettsiae vaccine candidate to be used in humans. The OmpB from virulent strains are heavily methylated while the attenuated strains are hypomethylated. Western blot analysis of partially digested OmpB revealed that one of the reactive fragments was located at the N-terminus (fragment A, aa 33-272). Recently, we have over expressed, purified, and chemically methylated the recombinant fragment A from Rickettsia prowazekii (Ap). The methylated Ap was thoroughly characterized by LC/MS/MS on the ProteomeX workstation. The protein sequence of Ap with and without methylation was 87.7% and 100% identified, respectively. This high sequence coverage enabled us to determine the sites and extent of methylation on the lysine residues in Ap. All the lysine residues except the C-terminus lysine were either mono-, di- or tri-methylated. In addition, carbamylation on the N-terminus glycine was identified using a combination of denovo sequencing (DeNovoX) and the pattern recognition (SALSA) program with accurate mass measurement. We demonstrated that the use of peptide identification (SEQUEST) in combination with SALSA and denovo sequencing provided a useful means to characterize the sequence and posttranslational modifications of given proteins.

This paper describes the profiling of human growth hormone (hGH) in human plasma in order to assess the dynamic range of the ion-trap mass spectrometer for proteomic studies of complex biological samples. Human growth hormone is an example of a low-level plasma protein in vivo, present at subfemtomole levels. This study was performed on a plasma sample in which hGH has been spiked at 10-fold above the natural level, that is approximately 16 pg/microL of plasma. Initially, the measurement was carried out without any sample enrichment and consisted of the following steps: the full set of plasma proteins were reduced, alkylated, and digested with trypsin, and the resulting peptides were separated on a capillary C-18 column and then detected by ion-trap mass spectrometry (1D LC/MS). In addition, this study provided a global view of the serum proteome with over 200 plasma proteins being preliminarily identified. In the MS/MS analysis, hGH was detected by characterization of the first tryptic peptide (T1). The initial identification was confirmed by alternative approaches, which also allowed the evaluation of different sample purification protocols. First, the plasma sample containing hGH was fractionated on a reversed-phase HPLC column and digested, and hGH could now be identified by MS/MS measurements of two tryptic peptides (T1 and T4) by the same 1D LC/MS protocol. In addition, the assignment of peptide identity was made with higher certainty (as measured by an algorithm score). The plasma sample was also fractionated by 1D and 2D gel electrophoresis, the selected bands were digested and analyzed again by the 1D LC/MS protocol. In both cases using the gel prepurifications, hGH was identified with additional peptides. Finally, the plasma sample was analyzed by 2D chromatography (ion exchange and reversed phase) on a new instrumental platform (ProteomeX), and hGH was identified by the observation of five tryptic peptides. In conclusion, these experiments were able to detect growth hormone in the low femtomole level with a dynamic range of 1 in 40 000 by several independent approaches. The amount of growth hormone, while 10-fold above normal in vivo levels, represents concentrations that may be present in disease states (such as acromegaly) and also in doping control measurements. These studies have demonstrated that shotgun sequencing approaches (LC/MS/MS) not only can profile high-abundance proteins in complex biological fluids but also have the potential to identify and quantitate low-level proteins present in such complex mixtures without extensive prepurification protocols. A key to such studies, however, is to use targeted approaches that reduce the complexity of the solute mixture that is presented to the mass spectrometer at a given time point. The various sample preparation protocols described here all improved the quality of the hGH measurement, although in this study the 2D chromatographic approach gave the greatest sequence coverage.

This study uses multiple enzyme digests to increase the sequence coverage of proteins identified by the shotgun sequencing approach to proteomic analysis. The enzymes used were trypsin, Lys-C, and Asp-N, which cleave at arginine and lysine residues, lysine, and aspartic acid residues, respectively. This approach was evaluated with the glycoprotein, tissue plasminogen activator, t-PA and gave enhanced sequence coverage, compared with a single enzymatic digest. The approach was then evaluated with a complex proteomic sample, namely plasma. It was found that trypsin and Lys-C were able to detect overlapping but distinct sets of proteins and a digital recombination of the data gave a significant increase in both the number of protein identifications as well as an increase in the number of peptides identified per protein (which improves the certainty of the assignment).

This paper will review the challenges of developing a proteomics strategy. A key issue is the integration of the two-dimensional (2-D) gel platform with mass spectrometry measurements. The use of both matrix-assisted laser/desorption ionization (on off-line coupling) and electrospray (on-line) ionization are complementary. While the use of one-dimensional and 2-D gels are essential to many aspects of proteomics research (sample preparation, preliminary fractionation and quantitation, storage of protein components), the emergence of shotgun sequencing based on high performance liquid chromatography and tandem mass spectrometry offers a powerful new approach. The latter has particular utility in the characterization of low level samples and complex post-translational modifications. The development of capillary columns, such as 75 to 150 micron, that can be packed in a reproducible manner has been a key step in the development of high sensitivity liquid chromatography/mass spectrometry analysis.

Insulin-like growth factor binding protein-4 (IGFBP-4) is, like the other five IGFBPs, a critical regulator of the activity of insulin-like growth factor (IGF)-I and IGF-II. Whereas IGFBP-1 and IGFBP-2 are not glycosylated, IGFBP-3 and IGFBP-4 are N-glycosylated and IGFBP-5 and IGFBP-6 are O-glycosylated. In this study we identified the glycosylation of IGFBP-4 using a nanoflow LC/MS/MS techniques. Although N-linked oligosaccharides are structurally diverse, their variants are well reported in the literature. Based on the molecular weight of the possible oligosaccharide moieties, we identified five different glycosylation isoforms of the protein. Identified glycans were biantennary and differ in the number of sialic acid terminal residues and/or core modification with fucose.

Alternative splicing happens in all types of cells including benign or normal cells. However, the expression levels of the non-canonical variants in the normal cells are low that there is no notable effect by them. Studies have shown that altered cellular mechanisms in tumor cells can affect the transcription and splicing processes. If a variant is able to fold into a stable structure similar to that of the canonical variant, it may mimic the structural features and interact with interaction partners without processing them further. Thus, alternative splicing could provide a mechanism for turning an activator into an effective inhibitor. Moreover, the ratio of the expression of the splice isoforms plays a significant role in the normal functioning of a biological system. We hypothesize that due to the altered mechanisms in tumor environment, some non-canonical variants are expressed at higher levels that they have distinct functions compared to their corresponding canonical proteins or may even be drivers of certain pathological processes. In this study, we propose to analyze the alternative splice variants that are expressed in ERBB2+ (ERBB2+, ER-, PR-) subtypes of breast cancer. ERBB2+ subtypes are known to be aggressive with poor prognosis and account for 15-20% of breast cancers in US. Breast cancer cell lines have been used widely to investigate breast cancer mechanisms and to develop new therapeutic approaches. SKBR3, which is ER –, PR - with ERBB2/HER2 amplification has been used successfully as a preclinical model to screen for therapeutic agents targeting ERBB2 and to delineate mechanisms of resistance to ERBB2-based therapies. SUM190 and SUM149 serve as models for inflammatory breast cancer, the most lethal form of breast cancer. Both these cell lines are ER –, PR – and clinically very aggressive, but ERBB2 is amplified in SUM190 and expressed at low levels in SUM149. The contrasting ERBB2 expression observed in the similar clinical IBC phenotype represented by SUM149 and SUM190 and over-expression of ERBB2 in SKBR3, representing a non IBC epithelial adenocarcinoma tumor type, makes this group of cell lines ideal for the comparisons that we set out to produce. By integrating the proteomics and RNA-Seq data from SKBR3, SUM190 and SUM149, we studied the splice variants that are expressed in the Her2+ breast cancer subtypes. We identified more than one splice variant for 1904 genes expressed in at least one of the three cancer cell lines. Multiple variants of genes that are key players in the downstream ERBB signaling pathways along with variants specific to one cancer cell line compared to the other two cancer cell lines and normal mammary cells were found. The top-ranking Gene Ontology and BioCarta pathways for these specific variants pointed to distinct mechanisms; aminoglycan metabolism for SKBR3, lipid biosynthesis and metabolism in SUM190 and cell adhesion in SUM149. The transcript profiles for mRNA splicing mechanism show similarities between the inflammatory models SUM190 and SUM149, which was further supported by the identification of a novel peptide from the intronic region of 17-beta dehydrogenase 4 from both of these cell lines. The overall transcript profile and some key ERBB downstream pathways show a more robust clustering between SKBR3 and SUM190. Detailed experimental studies on the distinct variants identified from each of these three breast cancer cell lines may unveil their roles in the underlying cancer progression and metastasis. Citation Format: Rajasree Menon, Hogune Im, Michael Snyder, Emma (Yue) Zhang, Rui Chen, Shiaw-Lin Wu, William S. Hancock, Gilbert S. Omenn. Splice variants in aggressive human breast cancer subtypes. [abstract]. In: Proceedings of the Third AACR International Conference on Frontiers in Basic Cancer Research; Sep 18-22, 2013; National Harbor, MD. Philadelphia (PA): AACR; Cancer Res 2013;73(19 Suppl):Abstract nr B38.

Atrial natriuretic peptide (ANP) is a peptide with 25 amino acid residues (hANP 4-28) and one intra-chain disulfide bond. We used the size-exclusion chromatography (SEC), sodium dodecyl sulfate in polyacrylamide gel electrophoresis (SDS-PAGE), and reversed-phase liquid chromatography with electrospray mass spectrometry (LC-MS) methods to examine the freeze-dried products of ANP to determine the types, sizes, and amounts of the multimer formation in different stability protocols (by varying the conditions with lyophilization cycles, excipients, storage temperatures, and times). Under the non-annealing lyophilization cycle or lyophilization with high concentration of bulking agent (mannitol), multimer formation increased with increasing storage times. Two kinds of multimers were observed; the major portion is reducible and the minor portion is non-reducible. The reducible multimers are disulfide-linked multimers as determined by LC-MS. The non-reducible multimer was mainly a dimer, possibly linked by a covalent bond between the side chain of tyrosine in one molecule and the dehydroalanine intermediate in another molecule, based on the evidences of the mass of the non-reducible dimer along with the elution position in SEC, and the change of the UV spectrum in the aromatic region. The analysis of degradants suggests that the mechanism start from an beta-elimination of disulfide linkage to form a free thiolate ion (HS-) and a dehydroalanine-type peptide intermediate. The HS- then catalyzed ANP to form the disulfide-linked multimers. The dehydroalanine-type ANP intermediate then reacted with another ANP molecule to form a non-disulfide-linked dimer through reaction with the side chain of tyrosine. These results suggest that the source of multimer formation be initiated by phase transition (from amorphous to crystalline phase) either in the freeze-dried process or during storage. That phase change may induce a drastic change in pH and moisture to damage the peptide. The detailed mechanism and the kinetics of ANP multimerization are discussed. The formation of the multimers was diminished by using the thermal treatment (the annealing step) with a proper ratio of mannitol to ANP peptide in the lyophilization, and/or increase of the acetate buffer concentration in the formulation.

SDS-PAGE is one of the most widely used techniques for the analysis of proteins, owing to the importance of determining molecular weight. The capillary electrophoresis sieving technique described here can also provide molecular weight data and with greater ease, speed, and accuracy. In addition, CE through a polymer network shows promise as a high-resolution separation tool for determining protein purity. Impurities present at extremely low levels can be quantitated with relative ease compared to SDS-PAGE. The technology is currently available in the form of kits sold by Applied Biosystems, Beckman, and Bio-Rad and should provide an attractive alternative to SDS-PAGE, especially in applications where high resolution and quantitation are required.

Publisher Summary Hydrophobic interaction chromatography (HIC) is a useful tool for purifying proteins, with maintenance of the biological activity. In this method, hydrophobic ligands (e.g., n-alkyl or phenyl groups) are chemically bound to matrices (polymer or silica gels) and protein components, interact with these ligands, through the application of a high concentration of an anti-chaotropic salt [e.g., (NH4)2S04)]. This interaction is diminished, by decreasing salt concentration, leading to protein elution. In this chapter, general guidelines, concerning the separation conditions in HIC, are discussed in this chapter, with a variety of examples (e.g., Escherichia coli-expressed proteins, glycosylated proteins, membrane proteins, and antibodies). Protein structure and conformation play a central role in determining chromatographic retention and behavior. Conformational effects (e.g., choices of column temperature and type vs protein structure and hydrophobicity) are important tools, with which to manipulate the protein structure, for variant separation. The introduction of new matrix supports (e.g., perfusive and nonporous particle) has enhanced the speed of separation dramatically (from hours to a few minutes). To enlarge the separation window and increase selectivity, mobile-phase additives (e.g., detergents, metal ions, urea, guanidine hydrochloride, ethylene glycol, glycerol, and organic solvents) are particularly important and are widely applied.

Publisher Summary 5-Hydroxylysine (Hyl) is a common modification of collagens and collagen-like domains of the Clq subcomponent of complement, acetyl cholinesterase, pulmonary surfactant apoproteins, mannose binding proteins, types I and II macrophage receptors, and bovine conglutinin. This chapter describes a modified amino acid analysis program capable of detecting hydroxylysine at levels down to 0.05 residues/mol to determine the distribution of Hyl in rtPA, as well as in some other proteins derived from mammalian cells. In recent study, a known post-translational modification, hydroxylysine in unexpected places is identified. The modification appears to be cell line-independent, as it was found in tPA from Bowes melanoma cells, as well as rtPA from transfected CHO and 293 cells. The Hyl modifications are overlooked in earlier characterization efforts because of the co-elution of Hyl with numerous His residues during amino acid analysis, and also because the Hyl residues are found in minor tryptic peptides that doesn't result from cleavage at the expected sites. These findings suggest that surface accessibility could play a role in the extent of Lys hydroxylation; however, partial hydroxylation of the Lys residues will make it difficult to determine the biological significance of this modification in non-collagenous proteins.

This paper examines the retention behavior of recombinant DNA-derived human growth hormone (rhGH) in reversed-phase chromatography and its separation from the closely related N-methionyl variant (Met-hGH). It is first shown that retention for rhGH decreases with increasing column temperature when 1-propanol (1-PrOH) is used as organic modifier. On the other hand, retention increases with temperature when acetonitrile (CH3CN) is employed. The differences in behavior for the two organic modifiers could be related to conformational changes in the protein as determined by solution and adsorption intrinsic fluorescence spectroscopy. Specifically, desorption and elution of rhGH using 1-PrOH could be correlated with a solvent-induced conformational change, with retention decreasing with increasing temperature due to the increasing ease of structural alteration. On the other hand for CH3CN the increase in retention correlated with temperature rise was related to a partial structural change yielding a more hydrophobic species. In this case, a surface-driven process is suggested. The work then turned to the separation of rhGH and Met-hGH where it was found for both organic modifiers optimum separation occurred at 45 degrees C and pH 6.5. Separate studies revealed that during the conformational change Met-hGH appeared more hydrophobic than rhGH since protein-protein aggregation was observed at a lower 1-PrOH concentration. It is suggested that this hydrophobic difference, which was optimized under the conditions cited above, resulted in the separation. The study demonstrates the importance of conformational changes in retention behavior and separation of protein samples.

The separation of charge variants of recombinant DNA-derived (rDNA) proteins by high-performance capillary electrophoresis (HPCE) has been explored with the following examples: human growth hormone (rhGH), a soluble form of a T4 receptor protein (rCD4) and tissue plasminogen activator (rt-PA). The separation of rhGH and deamidated variants were examined over the range of low pH (2.5) to high pH (8.0) on a coated silica capillary. No resolution was observed at pH 2.5 while at pH values of 6.5 or greater the deamidated species were separated. At pH 3.5 the variant was partially resolved but no detector signal was observed at pH 4.5 and 5.0. HPCE was also used to monitor a glycoprotein (rt-PA) with charge heterogeneity presumably due to variable sialic acid content. At a pH value of 4.5, the charge heterogeneity was only observed as peak broadening. For rCD4 multiple peaks were observed at pH 5.5 but no signal was observed at pH 6.5 or above (the pI of rCD4 is 8). These results suggest that HPCE will prove to be a valuable technique for the analysis of charged variants present in rDNA products either as a consequence of natural microheterogeneity or due to degradative processes such as deamidation.

Production of proteins by recombinant DNA technology for use as pharmaceuticals requires the use of the most powerful tools of analytical protein chemistry in order to confirm purity and identity of the product and reliability of the process. Capillary electrophoresis is an emerging technology that shows high sensitivity and selectivity and may have promise in this application. The technique combines the instrumental control and quantification features of high-performance liquid chromatography with the separating power of electrophoresis, and thereby has attracted broad interest. In this report, human growth hormone expressed in bacteria has been analyzed by both free zone electrophoresis and isoelectric focusing in a coated capillary to demonstrate the separation of the native molecule from its deamidated variant. A capillary zone electrophoretic tryptic map has also been develope and characterized. This map complements the widely employed reversed-phase high- performance liquid chromatography tryptic mapping systems that are important in protein characterization. Certain drawbacks to capillary zone electrophoresis compared to other analytical methods are noted, including relatively poor reproducibility and low sample tolerance. For applications as demostrated here, however, the speed, separating power and sensitivity of the technique compensate for these shortcomings.

The thermal behavior of a series of standard proteins in hydrophobic interaction chromatography on a previously developed weakly hydrophobic ether-bonded phase column has been studied. Depending on the temperature and protein, conformational changes can occur in the chromatographic system. Methods for recognizing these conformational effects are presented, including retention and peak width variations with temperature, and Z values (the slope of the plot of log k' vs. log %B solvent. The Z value is shown to be a general index characterizing protein retention as a function of salt concentration. In addition, on-line UV spectroscopic analysis, (absorbance ratios and second derivative spectroscopy) with a photodiode array detector, is shown to corroborate chromatographic trends. Lysozyme maintains a stable conformation over the temperature range 10-40 degrees C, whereas beta-lactoglobulin A has a conformational transition between 25 degrees C and 40 degrees C. Calcium-depleted alpha-lactalbumin, a rather labile species, maintains a stable conformation up to ca. 20 degrees C, and then undergoes structural changes. Finally, cytochrome c appears to be relatively stable up to ca. 65 degrees C. Since conformational changes for this protein occur at ca. 35 degrees C on more hydrophobic phases, the extent of hydrophobicity of the stationary phase is important for maintenance of the native state. Based on this work, hydrophobic-interaction chromatography at sub-ambient temperatures appears promising.

We have studied the conformational behavior of α-lactalbumin (α-LACT) in hydrophobic interaction chromatography (HIC). Retention characterization in terms of Z (Slope of plot 1n k ′ vs . 1n o B , where k ′ is the capacity factor and O B is the volume fraction of mobile phase B) has been explored, and the relationship of Z to other slopes, such as S (slope of the plot of 1n k ′ vs . o B f) has been derived. The reason for the sensitivity of Z to conformational change are discussed. The enhanced broadening of α-LACT in a temparature transition region of conformational change has been studied by spectral analysis using on-line photodiode array detection. The influence of Ca 2+ and Mg 2+ addition to the mobile phase is further explored. Since α-LACT is a calcium binding protein, addition of this metal leads to stabilization, i.e higher column temperatures are required for conformational change. On the other hand, addition of Mg 2+ appears to destabilize the protein. We have explored the use of a more hydrophobic support, C 2 -(ethyl) ether phase, for the elution of α-LACT. In this case, two widely separated peaks ae observed. By spectral analysis the first peak is shown to be native and the later eluted, broad second peak to ber an unfolded mixture of species. As previously observed in reversed-phase liquid chromatography, the second peak grows at the expense of the first, as the column temperature is raised. The second peak also grows as the contact time of the protein with the surface increases. This behavior can be ascribed to the conformational change of α-LACT in the column, the late eluted species under the second peak binding significantly more strongly to the phase than the native peak. Reinjection of the late eluted fraction reveals the reformation of the native species takes place in solution within 30 min. As before, addition of Ca 2+ reduces the extent of unfolding under any specific condition. These results add further to our understand and ability to control conformational changes in high-performance liquid chromatography.