Determination of interchain crosslinkages in insulin B-chain dimers by fast atom bombardment mass spectrometry

New methodology for identifying and locating crosslinkages in peptides is described. Pepsin is used to cleave insulin and B-chain dimers of insulin into fragments under conditions which retain the original peptide crosslinkages. After partial separation by HPLC, the peptides are analyzed by fast atom bombardment mass spectrometry (FABMS) to determine their molecular weights. The molecular weights of peptide fragments expected from the pepsin digests of human insulin and related model compounds are calculated from the amino acid sequence of the intact peptide. Digestion products are identified by matching their molecular weights, as determined by FABMS, with calculated molecular weights. Locations of interchain crosslinkages are deduced after the peptide fragments have been assigned to specific segments of the parent peptide. The validity of the method has been established by correctly identifying structurally important products in the pepsin digests of model compounds such as human, bovine, and porcine insulins. Procedures developed with the model compounds were used to identify crosslinkages in peptides of unknown structure isolated from an insulin A-chain/B-chain combination reaction mixture. Evidence is presented for formation of three different types of crosslinkages, disulfide, lanthionine, and sulfoxide.     

Characterization of the cysteinyl-containing peptides of the gamma subunit of coupling factor 1

The cysteinyl peptides of the gamma subunit of chloroplast coupling factor 1 (CF1) have been analyzed by high performance liquid chromatography. Analysis of the reduced enzyme alkylated with 4-vinylpyridine showed that the gamma subunit contains four cysteinyl residues. Two of these residues are involved in a disulfide linkage in CF1 either in solution or bound to washed thylakoid membranes. Two free sulfhydryls, one that is readily attacked by alkylating reagents and another that is less reactive, were also detected. Each of these four cysteinyl residues is present in a separate tryptic peptide derived from the gamma subunit. These results show that 4-vinylpyridine is an excellent reagent for the analysis of cysteinyl-containing peptides and support our analyses of the roles of cysteinyl residues in the gamma subunit in ATP synthesis and hydrolysis.     

Use of proline-specific endopeptidase in the isolation of all four "native" disulfides of hen egg white lysozyme

The disulfide peptides from the tryptic digestion of cyanogen bromide-treated hen egg white lysozyme (HEWL) were isolated by reverse phase high performance liquid chromatography (HPLC) and identified by amino acid analysis. Three peptides containing the I-VIII, II-VII, and III-V + IV-VI disulfide bonds were obtained. The two-disulfide peptide was further digested with proline-specific endopeptidase (PCE) (EC 3.4.21.26). Amino acid analysis of digest peptides separated by HPLC showed four peptides with the IV-VI disulfide bond as well as a peptide with the III-V disulfide bond. The IV-VI peptides were produced by hydrolysis of several alanine-X bonds as well as the prolyl-cystine bond. Our studies show that alanyl peptide bonds to lysyl, seryl, and leucyl residues are susceptible to hydrolysis by PCE preparations, thus substantially extending its known specificity range. The two-disulfide peptide was also digested sequentially with thermolysin and PCE; the resulting IV-VI and III-V peptides were identified by HPLC and amino acid analysis. PCE showed substantial activity at pH 5.3 as well as at pH 8.3. The lower pH is useful in studies of proteins or peptides where base-catalyzed reactions must be limited.     

Sequence of the cysteinyl-containing peptides of 4-aminobutyrate aminotransferase. Identification of sulfhydryl residues involved in intersubunit linkage

Three cysteine-containing tryptic peptides were isolated and sequenced from mitochondrial 4-aminobutyrate aminotransferase using DABIA (4-dimethylaminoazobenzene-4-iodoacetamide) as specific labeling reagent for sulfhydryl groups. The enzyme is a dimer made up of two identical subunits, but four out of the six cysteinyl residues/dimer form disulfide bonds when treated with iodosobenzoate to yield inactive enzyme species. To identify the cysteinyl residues undergoing reversible oxidation/reduction, the S-DABIA-labeling patterns of the fully reduced (active) and fully oxidized (inactive) forms of the enzyme were compared. Tryptic digests of the reduced enzyme contained three labeled peptides. If the enzyme was treated with iodosobenzoate prior to reaction with DABIA and tryptic digestion, only one labeled peptide was detected and identified (peptide I), indicating that the two missing cysteinyl-containing peptides (peptides II, III) have been oxidized. The sulfhydryl groups undergoing oxidation/reduction were found to be intersubunit, based on SDS/polyacrylamide gel electrophoresis results. The loss of catalytic activity of 4-aminobutyrate aminotransferase by oxidation of sulfhydryl residues is related to constraints imposed at the subunit interface by the insertion of disulfide bonds.     

Post-translational processing of preprosomatostatin-II examined using fast atom bombardment mass spectrometry

The products and an intermediate of preprosomatostatin-II processing in the anglerfish islet were purified and subjected to structural analysis. The peptides isolated identify the site of signal cleavage (between Ser-24 and Gln-25). The prohormone is further processed at Arg-97 and, to a lesser extent, at the two adjacent basic amino acid residues Lys-61 and Arg-62. A 28-residue somatostatin is also generated which can be hydroxylated at Lys-23. A proteolytic processing site which would form the 14-residue somatostatin does not appear to be used to a significant degree. Fast atom bombardment mass spectrometry (FABMS) was used to demonstrate that the amino-terminal residues of peptides 25-60, and 25-90 are pyroglutamic acid, a modification which precludes Edman degradation of these peptides. Analysis of the peptides and tryptic peptide maps by FABMS allowed confirmation of the sites of prohormone conversion and indicated that terminal basic residues were removed during processing. Three amino acid residues were also found to differ from the amino acid sequence deduced from the cDNA and were localized to specific regions by FABMS analysis. Residues found to differ from the cDNA (cDNA in parentheses) were: Asp-77 (Thr), Val-78 (Phe), and Gly-90 (Glu). Mass assignments were confirmed by running a single cycle of Edman degradation prior to FABMS. The peptides noted above were also examined by Edman sequence analysis. The sequence of a cDNA clone to preprosomatostatin-II was re-examined in light of the observed differences at the protein level. This study emphasizes the utility of FABMS in prohormone processing studies and in identification of post-translational processing events.     

Study of the primary structures of the peptide core of bovine estrus cervical mucin. Possible existence of small similar subunits

Two populations of tryptic peptides were isolated from bovine estrus cervical mucin (BCM). One contained all the carbohydrate, and was rich in threonine and serine. These glycopeptides had, like the whole mucin, alanine as their NH2-terminal residues. Their COOH-terminal residues were arginine. The second population of peptides was rich in carboxylic amino acids, contained two cysteinyl residues, and had, like the whole mucin, leucine as COOH-terminal residues. Their NH2-terminal residues were aspartic acid. The sum of the residues of one glycopeptide plus one cysteinyl-containing peptide corresponded to the number of residues constituting a putative subunit of BCM. The amino acid sequence of the major cysteinyl peptide was determined. A cluster of hydrophobic residues was found in the COOH-terminal region. The amino acid sequences of two of the glycopeptides were found identical up to the 22nd residue. The small number of tryptic peptides, as well as the large amount of NH2- and COOH-terminal amino acids found in BCM indicate that this glycoprotein is made up of similar subunits with a molecular weight of about 22,000, one of the glycopeptides representing the NH2-terminal part, and one of the cysteinyl peptides, the COOH-terminal part. However, the existence of these subunits was not confirmed by ultracentrifugation of BCM in dithiothreitol and sodium dodecyl sulfate. BCM was polydisperse and had a mean molecular weight of 507,000.     

Assignment of disulfide bonds in proteins by fast atom bombardment mass spectrometry

A rapid and sensitive method for assignment of disulfide bonds using fast atom bombardment mass spectrometry is described for hen egg white lysozyme and bovine ribonuclease A. The protein is initially digested to a mixture of peptides using chemical and enzymatic methods under conditions which minimize disulfide bond reduction and exchange. The digested sample is analyzed directly by fast atom bombardment mass spectrometry before and after chemical reduction of cystine residues. An important feature of the method is that it is not necessary to completely resolve the peptides in the digest chromatographically prior to analysis. The disulfide-containing peptides are also characterized directly by prolonged exposure of the sample to the high energy xenon atom beam which results in the reduction of cystine residues. Intra- as well as interchain disulfide bond assignments are made on the basis of the mass difference between the molecular ions (MH+) of the oxidized and reduced peptides. Confirmation of the mass assignments may be obtained from the mass spectra of the digests after one cycle of manual Edman degradation. Although the quantity of protein required to unambiguously assign all of the disulfide linkages will depend on the ease with which the appropriate peptide fragments can be formed, results from these studies indicate that approximately 1 nmol of protein is usually sufficient.     

The amino acid sequence of ribonuclease U1, a guanine-specific ribonuclease from the fungus Ustilago sphaerogena

The complete amino acid sequence of ribonuclease U1 (RNase U1), a guanine-specific ribonuclease from a fungus, Ustilago sphaerogena, was determined by conventional protein sequencing, using peptide fragments obtained by several enzymatic cleavages of the performic acid-oxidized protein. The oxidized protein was first cleaved by trypsin and the resulting peptides were purified and their amino acid sequences were determined. These tryptic peptides were aligned with the aid of overlapping peptides isolated from a chymotryptic digest of the oxidized protein. The amino acid sequence thus deduced was further confirmed by isolation and analysis of peptides obtained by digestion of the oxidized protein with lysyl endopeptidase. The location of the disulfide bonds was deduced by isolation and analysis of cystine-containing peptides from a chymotryptic digest of heat-denatured RNase U1. These results showed that the protein is composed of a single polypeptide chain of 105 amino acid residues cross-linked by two disulfide bonds, having a molecular weight of 11,235, and that the NH2-terminus is blocked by a pyroglutamate residue. It has an overall homology with other guanine-specific or related ribonucleases, and shows 48% identity with RNase T1 and 38% identity with RNase U2.     

The folding pathway of reduced lysozyme.

Studies on the mechanism of the glutathione regeneration (Saxena, V.P., and Wetlaufer, D.B. (1970) Biochemistry 9, 5015-5023) of hen egg lysozyme have been carried out. The first two stoichiometric disulfides in lysozyme are formed about 8 times more rapidly than the second two. Almost no enzymic activity is regained until the first two disulfides are formed, thus ruling out an all-or-none mechanism. The disulfide peptides formed early in the regeneration have been isolated and identified. The results show a limited search of folding intermediates, and outline a folding pathway. The early disulfides involve cysteinyl residues III, IV, V, and VI. At the same time cysteinyl residues I, II, VII, and VIII are still reduced, as demonstrated by their isolation as S-alkylated derivatives. At slightly later times a peptide is found which contains the (native) disulfide between cysteinyl residues II and VII. It is likely, but as yet unproven, that formation of disulfide I-VIII completes the cross-linking of lysozyme.     

Characterization of disulfide bond position in proteins and sequence analysis of cystine-bridged peptides by tandem mass spectrometry.

Tandem mass spectrometry employing high-energy, collisionally activated dissociation (CAD) is shown to be a useful method for sequencing through the cystine bridge of intermolecularly disulfide-bonded peptides. A characteristic triplet of intense fragment ions is observed corresponding to cleavage through and to either side of the disulfide bridge. These fragments define the masses of the linked peptides. Fragments due to peptide chain cleavage are also observed at lower abundance in the product-ion spectra and can be sufficient to sequence both of the disulfide-linked peptides without any prior knowledge of the peptide or protein sequence. Even in cases where the peptide sequence-related product-ion yields are poor, the intensities of the disulfide cleavage ions are usually sufficient to determine the molecular weights of the component cystine-bridged peptides. In this paper we demonstrate that the high-energy CAD tandem MS approach may be used to characterize disulfide-bonded peptides directly in complex enzymatic or chemical digests of native proteins. This obviates the need for individual purification of intermolecularly disulfide-linked peptides prior to analysis. The techniques are illustrated here for synthetic, inter- and intramolecularly disulfide-linked peptides and for human transforming growth factor-alpha (des-Val-Val-TGF-alpha), a compact protein containing 48 residues and three disulfides.     

Isolation and characterization of a resistant core peptide of recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF); confirmation of the GM-CSF amino acid sequence by mass spectrometry.

A trypsin-resistant core peptide of recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) was isolated and analyzed by high-energy Cs+ liquid secondary-ion (LSI) mass spectrometric analysis. This analysis provided successful detection of the high-mass disulfide-linked core peptide as well as information confirming the existence of disulfide pairing. Similarly, LSI mass spectrometric analysis of the peptide fragments isolated chromatographically from a Staphylococcus aureus V8 protease digest of rhGM-CSF provided rapid confirmation of the cDNA-derived sequence and determination of the existing disulfide bonds between cysteine residues 54-96 and 88-121. Electrospray ionization mass spectrometry was employed to measure the molecular weight of the intact protein and to determine the number of the disulfide bonds in the protein molecule by comparative analysis of the protein before and after reduction with beta-mercaptoethanol.     

An antibody-free enrichment approach enabled by reductive glutaraldehydation for monomethyllysine proteome analysis

Protein lysine monomethylation is an important post-translational modification participated in regulating many biological processes. There is growing interest in identifying these methylation events. However, the introduction of one methyl group on lysine residues has negligible effect on changing the physical and chemical properties of proteins or peptides, making enriching and identifying monomethylated lysine (Kme1) proteins or peptides extraordinarily challenging. In this study, we proposed an antibody-free chemical proteomics approach to capture Kme1 peptides from complex protein digest. By exploiting reductive glutaraldehydation, 5-aldehyde-pentanyl modified Kme1 residues and piperidine modified primary amines were generated at the same time. The peptides with aldehyde modified Kme1 residues were then enriched by solid-phase hydrazide chemistry. This chemical proteomics approach was validated by using several synthetic peptides. It was demonstrated that it can enrich and detect Kme1 peptide from peptide mixture containing 5000-fold more bovine serum albumin tryptic digest. Besides, we extended our approach to profile Kme1 using heavy methyl stable isotope labeling by amino acids in cell culture (hmSILAC) labeled Jurkat T cells and Hela cells. Totally, 29 Kme1 sites on 25 proteins were identified with high confidence and 11 Kme1 sites were identified in both two types cells. This is the first antibody-free chemical proteomics approach to enrich Kme1 peptides from complex protein digest, and it provides a potential avenue for the analysis of methylome.     

Primary structure of a human IgA1 immunoglobulin. III. Isolation, composition, and amino acid sequence of the thermolysin peptides.

As part of the strategy for determination of the complete covalent structure of a human IgA immunoglobulin, 66 peptides were isolated from a thermolysin digest of reduced and carboxymethylated IgA alpha1 chain Bur and were purified. They range in length from 2 to 24 residues. Some of the peptides have been characterized and sequenced in order to supply needed information that was not obtained from the chymotryptic and tryptic peptides. These thermolysin peptides provide much necessary data to produce a rigorous proof for the primary structure of the human alpha1 chain. The remaining peptides from the thermolysin digest whose amino acid composition and NH2-terminal residues were sufficient to identify them unequivocally have also been assigned in the structure. They supply additional information that helps remove ambiguity in the structure, and they provide useful data about the profile of the peptide bonds that are susceptible to thermolysin digestion.     

Assignment of disulfide bonds in proteins by partial acid hydrolysis and mass spectrometry

A new method is described for locating disulfide bonds in proteins which cannot be cleaved between half-cystinyl residues by enzymic methods, as is often the case for tightly coiled proteins, or for proteins in which half-cystinyl residues are not separated by residues required for enzymic cleavage. Partial acid hydrolysis of a model protein, hen egg-white lysozyme, produces a mixture of disulfide-containing peptides from which the disulfide connections may be deduced. The usefulness of a combination of HPLC, fast atom bombardment mass spectrometry, and computer-assisted analysis to identify disulfide-containing peptides present in the partial acid hydrolysate of the model protein is demonstrated. Chromatographic fractions of the hydrolysate were analyzed by mass spectrometry before and after chemical reduction of the disulfide bonds to determine the molecular weights of disulfide-containing peptides. Computer-assisted analysis was then used to relate the molecular weights of these peptides to specific segments of the protein from which the disulfide connectivities could be determined. Partial acid hydrolysis of proteins, which is attractive because it proceeds relatively independent of the amino acid sequence and structure, and because disulfide interchange is unlikely to occur in dilute acid, has become practical because disulfide-containing peptides present in complex mixtures can be identified rapidly and definitively by this method.     

rhIL-12二硫键、N-糖基化位点及C端氨基酸序列分析

重组人白细胞介素12（rhIL-12）是一种已经用于治疗肿瘤,寄生虫、病毒性感染及造血障碍等疾病研究的异二聚体糖蛋白。结构确证是质量控制的重要内容,此研究对CHO细胞表达的rhIL-12二硫键配对方式、N-糖基化位点以及C端氨基酸序列进行了分析,使用Trypsin、Chymotrypsin和Glu-C三种酶分别对rhIL-12进行非还原酶解,尽可能地在其所有半胱氨酸残基之间断裂而形成二硫键相连的肽段,然后使用LC-MS/MS对酶解后的肽段样品进行分析,确定了rhIL-12样品中存在和理论配对方式相符的7对二硫键。将rhIL-12二硫键还原后并烷基化修饰保护,分别采用Trypsin,Chymotrypsin和GluC进行酶解,并用LC-MS/MS对酶解后肽段进行了质谱肽图及C端氨基酸序列分析,确定了rhIL-12 p35亚基C端氨基酸序列的8个氨基酸、p40亚基C端氨基酸序列的15个氨基酸。对rhIL-12样品还原及烷基化后用Trypsin变性酶解,所得肽段在H₂O及H₂¹⁸O水中分别用PNGase F糖苷酶处理酶切产物。并通过二级质谱分析脱糖后糖肽段分子量变化,从而确定了rhIL-12的3个N糖基化修饰位点,分别为p35亚基的71位和85位以及p40亚基的200位。通过建立酶解结合二级质谱鉴定的方法,证明了新药rhIL-12的二硫键位点、C端氨基酸序列和糖基化位点与理论一致。     

The use of cysteinyl peptides to effect portage transport of sulfhydryl-containing compounds in Escherichia coli

We describe a method by which sulfhydryl compounds may be transported into Escherichia coli as the mixed disulfides with a cysteine residue of a di- or tripeptide. Transport occurs through the di- or oligopeptide transport systems, and it is suggested that subsequent release of the sulfhydryl compound occurs as a result of a disulfide exchange reaction with components of the sulfhydryl-rich cytoplasm. The free sulfhydryl compounds used here (2-mercaptopyridine and 4-[N-(2-mercaptoethyl)]aminopyridine-2,6-dicarboxylic acid) show weak growth-inhibitory properties in their own right, but disulfide linkage to a cysteinyl peptide results in a considerable enhancement (up to 2 orders of magnitude). This is the first example of the use of the peptide transport systems of E. coli to effect portage transport of a poorly permeant molecule by using attachment to the side chain of one of the amino acid residues of a peptide; all previous examples have involved the incorporation of amino acid analogues into the peptide backbone. The synthesis of cysteinyl peptides containing disulfide-linked 2-mercaptopyridine is described. Displacement of the 2-mercaptopyridine by sulfhydryl compounds of interest proceeds rapidly and quantitatively in aqueous alkaline solution to provide the required peptide disulfides.     

Accurate mass comparison coupled with two endopeptidases enables identification of protein termini

Protein termini play important roles in biological processes, but there have been few methods for comprehensive terminal proteomics. We have developed a new method that can identify both the amino and the carboxyl termini of proteins. The method independently uses two proteases, (lysyl endopeptidase) Lys-C and peptidyl-Lys metalloendopeptidase (Lys-N), to digest proteins, followed by LC-MS/MS analysis of the two digests. Terminal peptides can be identified by comparing the peptide masses in the two digests as follows: (i) the amino terminal peptide of a protein in Lys-C digest is one lysine residue mass heavier than that in Lys-N digest; (ii) the carboxyl terminal peptide in Lys-N digest is one lysine residue mass heavier than that in Lys-C digest; and (iii) all internal peptides give exactly the same molecular masses in both the Lys-C and the Lys-N digest, although amino acid sequences of Lys-C and Lys-N peptides are different (Lys-C peptides end with lysine, whereas Lys-N peptides begin with lysine). The identification of terminal peptides was further verified by examining their MS/MS spectra to avoid misidentifying pairs as termini. In this study, we investigated the usefulness of this method using several protein and peptide mixtures. Known protein termini were successfully identified. Acetylation on N-terminus and protein isoforms, which have different termini, was also determined. These results demonstrate that our new method can confidently identify terminal peptides in protein mixtures.     

Pyruvoyl-dependent histidine decarboxylases. Comparative sequences of cysteinyl peptides of the enzymes from Lactobacillus 30a, Lactobacillus buchneri, and Clostridium perfringens

The two cysteinyl residues present in histidine decarboxylase from Lactobacillus 30a differ greatly in reactivity. One (class 1) reacts readily in the native state with dithiobis-(2-nitrobenzoate) with complete loss of enzyme activity; the other (class 2) reacts only after denaturation of the enzyme (Lane, R. S., and Snell, E. E. (1976) Biochemistry 15, 4175-4179). These differences in reactivity permitted use of covalent (disulfide) chromatography to isolate separate peptides that contain these two residues. Sequence analysis showed that the class 1 cysteinyl residue is at position 147 in a hydrophilic portion of the alpha chain (Huynh, Q. K., Recsei, P. A., Vaaler, G. L., and Snell, E. E. (1984) J. Biol. Chem. 259, 2833-2839), while the class 2 cysteinyl residue is present at position 71, adjacent to a hydrophobic portion of the same chain. Cysteinyl peptides identical with or homologous to the class 2 cysteinyl peptide of the Lactobacillus 30a enzyme were isolated from the alpha subunits of histidine decarboxylases from Lactobacillus buchneri and Clostridium perfringens, respectively. The L. buchneri enzyme also contained a peptide homologous to the class 1 cysteinyl peptide from Lactobacillus 30a. However, no corresponding peptide was present in the enzyme from C. perfringens, in which the second cysteinyl residue of the alpha chain occupies position 3, very near the essential pyruvoyl residue. This enzyme, unlike those from Lactobacillus 30a or L. buchneri, also contains one cysteinyl residue in its beta chain. Although Cys 147 is an active site residue in histidine decarboxylase from Lactobacillus 30a, the absence of a corresponding residue in the C. perfringens enzyme confirms previous indications (Recsei, P. A., and Snell, E. E. (1982) J. Biol. Chem. 257, 7196-7202) that this SH group is not essential for decarboxylase action.     

Analysis of the in vivo phosphorylation state of protein phosphatase inhibitor-2 from rabbit skeletal muscle by fast-atom bombardment mass spectrometry

A new procedure has been developed for identifying phosphoserine residues in proteins, and is used to analyse the in vivo phosphorylation state of inhibitor-2. The method employs reverse-phase liquid chromatography to resolve phosphorylated and dephosphorylated forms of peptides and fast-atom bombardment mass spectrometry (FABMS) to identify phosphorylated derivatives. The positions of phosphorylation sites within peptides are located by gas-phase sequencer analysis after conversion of phosphoserine residues to S-ethylcysteine. The phosphorylation sites on inhibitor-2 were identified as serines-86, -120 and -121, the three residues phosphorylated in vitro by casein kinase-II. Serine-86 was phosphorylated to 0.7 mol/mol and serines-120 and -121 each to 0.3 mol/mol. These values were not altered significantly by intravenous injection of adrenalin or insulin. No phosphate was present in the region comprising residues 1-49, even after injection of adrenalin, demonstrating that inhibitor-2 is not a substrate for cyclic AMP-dependent protein kinase in vivo. The absence of phosphotyrosine also indicated that inhibitor-2 is not a physiological substrate for the insulin receptor. Surprisingly, no phosphate was present at threonine-72, the residue phosphorylated in vitro by glycogen synthase kinase-3, after injection of either propranolol, adrenalin or insulin. The implications of this finding for the in vivo activation of protein phosphatase 1I (the 1:1 complex between inhibitor-2 and the catalytic subunit of protein phosphatase-1) are discussed. FABMS analysis of inhibitor-2 confirmed the accuracy of the primary structure reported previously, and showed that the only post-translational modifications were an N-acetyl moiety and the three phosphoserine residues. FABMS also demonstrated the presence of an additional serine residue at the C-terminus, and showed that 50% of isolated inhibitor-2 molecules lack the C-terminal Ser-Ser dipeptide.     

A potent new class of reductively activated peptide gene delivery agents

A new class of peptide gene delivery agents were developed by inserting multiple cysteine residues into short (dp 20) synthetic peptides. Substitution of one to four cysteine residues for lysine residues in Cys-Trp-Lys(18) resulted in low molecular weight DNA condensing peptides that spontaneously oxidize after binding to plasmid DNA to form interpeptide disulfide bonds. The stability of cross-linked peptide DNA condensates increased in proportion to the number of cysteines incorporated into the peptide. Disulfide bond formation led to a decrease in particle size relative to control peptide DNA condensates and prevented dissociation of peptide DNA condensates in concentrated sodium chloride. Cross-linked peptide DNA condensates were 5-60-fold more potent at mediating gene expression in HepG2 and COS 7 cells relative to uncross-linked peptide DNA condensates. The enhanced gene expression was dependent on the number of cysteine residues incorporated, with a peptide containing two cysteines mediating maximal gene expression. Cross-linking peptides caused elevated gene expression without increasing DNA uptake by cells, suggesting a mechanism involving intracellular release of DNA triggered by disulfide bond reduction. The results establish cross-linking peptides as a novel class of potent gene delivery agents that enhance gene expression through a new mechanism of action.