Trinitrophenylation of bovine growth hormone.

Bovine growth hormone was chemically modified with picryl sulfonic acid, at pH 8.4 during 2 and 5 min of reaction. The N-terminal residue provides the most reactive amino group followed by the epsilon-amino groups of lysine 179 and lysines 143, 69, 111, 170 and 166 in decreasing order. These results agree with those obtained previously with equine growth hormone, except that residue 156 is not modified in bovine growth hormone. An important decrease in biological activity occurs between 2 and 5 min of reaction without sensible modification in the alpha-helix content of the molecule.     

Modification of lysine 69 reactivity in bovine growth hormone by carbamylation of its N-terminal group

Bovine growth hormone was carbamylated under conditions that assure full reaction of the N-terminal residue. Approximately 28% of lysine 179 and 7% of lysine 143 were also carbamylated. The modified hormone retained an important growth promoting activity and was as effective as the native hormone in competition assays in vivo for the receptors in rat liver. However, a change in its conformation must occur since lysine 69, which is resistant to trinitrophenylation in the native hormone, reacted easily and under mild conditions, in the carbamylated protein, The growth promoting activity and binding properties of the carbamylated and trinitrophenylated hormone were practically nil.     

Role of lysines in ion selectivity of bacterial outer membrane porins

The epsilon-amino groups of available lysine residues of the OmpC, OmpF and PhoE porin proteins of Escherichia coli and of the protein P porin of Pseudomonas aeruginosa, were modified by the bulky reagent trinitrobenzenesulphonic acid. Approximately 78% of the lysines of the anion-selective protein P and PhoE porins were modified whereas only 40-50% of the lysines of the cation selective OmpF and OmpC porins were altered. After modification, the three E. coli porins had very similar high selectivities for cations over anions, in contrast to the native porins which varied 86-fold in ion selectivity. Despite the large size of the trinitrophenyl group attached to modified lysines (i.e., a disc of approx. 0.86 nm diameter X 0.36 nm high) relative to the reported size of the constrictions of the E. coli porins (1.0-1.2 nm diameter), only the anion-selective PhoE porin was substantially blocked after trinitrophenylation. The protein P porin channel was relatively unaffected by trinitrophenylation, in contrast to previous data showing dramatic effects of acetylation of lysines on protein P conductance and selectivity. This favoured a model in which the critical lysines involved in anion binding by protein P were present in a constriction of the channel that was too small for trinitrobenzenesulphonic acid to enter. Overall, the data suggest that both the number and relative position of charged lysines are major determinants of ion selectivity.     

Apolipoprotein E;-low density lipoprotein receptor interaction. Influences of basic residue and amphipathic alpha-helix organization in the ligand

Conserved lysines and arginines within amino acids 140-150 of apolipoprotein (apo) E are crucial for the interaction between apoE and the low density lipoprotein receptor (LDLR). To explore the roles of amphipathic alpha-helix and basic residue organization in the binding process, we performed site-directed mutagenesis on the 22-kDa fragment of apoE (amino acids 1-191). Exchange of lysine and arginine at positions 143, 146, and 147 demonstrated that a positive charge rather than a specific basic residue is required at these positions. Consistent with this finding, substitution of neutral amino acids for the lysines at positions 143 and 146 reduced the binding affinity to about 30% of the wild-type value. This reduction corresponds to a decrease in free energy of binding of approximately 600 cal/mol, consistent with the elimination of a hydrogen-bonded ion pair (salt bridge) between a lysine on apoE and an acidic residue on the LDLR. Binding activity was similarly reduced when K143 and K146 were both mutated to arginine (K143R + K146R), indicating that more than the side-chain positive charge can be important.Exchanging lysines and leucines indicated that the amphipathic alpha-helical structure of amino acids 140-150 is critical for normal binding to the low density lipoprotein receptor.     

Determination of the effect of acetylation of specific lysine residues in human growth hormone on its affinity for somatogenic receptors by an affinity selection technique

A technique is described to study the effect of acetylation of individual lysine residues in peptide hormones on the affinity for their receptors, and is illustrated for the case of human growth hormone (hGH) binding to somatogenic receptors. The hGH was partially acetylated with high specific activity [3H]-acetic anhydride and the product ([3H]-Ac-hGH) was incubated with solubilised affinity-purified somatogenic receptors (from male rat liver) in the presence and absence of excess unlabelled hGH. The receptor-bound and unbound labelled hormone were separated by gel filtration and subjected to HPLC tryptic peptide mapping after the addition of cold carrier Ac-hGH. Peaks of [3H] radioactivity were assigned to peptides corresponding to the acetylation of specific lysine residues in the hGH sequence by amino acid analysis and sequencing. Comparison of the relative intensities of corresponding [3H] peaks in the peptide maps of added receptor, bound and unbound [3H]-Ac-hGH, enabled the relative receptor-binding potencies of different acetylated hGH species to be determined. Acetylation of lysine 168 or 172 in hGH greatly decreases its receptor-binding affinity, acetylation of lysine 115 probably causes a minor decrease, whereas acetylation of lysines 38, 70, and the N-terminal amino group have no appreciable effect. Acetylation of lysine 140 causes a significant increase in receptor-binding affinity.     

Primary structure of alpaca growth hormone.

Reduced and carbamidomethylated alpaca growth hormone was submitted to tryptic digestion. Peptides in the mixture were purified by reverse phase HPLC and N-terminal determination and an amino acid analysis of each was performed. Data obtained and the already known primary structure of the equine growth hormone allowed the assembly-by homology-of a definite sequence of amino acids for the polypeptide chain of the protein. Present data provide further information about the relationship between growth factors.     

Studies on the status of free amino and carboxyl groups in cobrotoxin

The status of free amino groups in cobrotoxin was studied by stepwise modification with trinitrobenzene sulfonate. Lys-27 was selectively modified without altering the activity of cobrotoxin. However, complete loss of activity was observed when Lys-27 and Lys-47 were trinitrophenylated, suggesting that the epsilon-amino group of Lys-47 is essential for the activity of cobrotoxin. The alpha-amino group of N-terminal leucine had no correlation with activity, demonstrated by the guanidination of the lysine residues with O-methylisourea followed by trinitrophenylation of the alpha-amino group. The carboxyl groups in cobrotoxin were modified with glycine methyl ester after activation with water-soluble carbodiimide. Six out of seven free carboxyls reacted in the absence of guanidine.HCl without altering the biological activity. When the remaining carboxyl was modified in the presence of 5 M guanidine.HCl, the resulting toxin was devoid of activity. This "buried" carboxyl is essential for activity and was identified as the gamma-carboxyl group of Glu-21.     

Conformation studies of biologically active fragments of bovine growth hormone

Conformations of bovine growth hormone active fragments were studied using far ultraviolet circular dichroism and intrinsic fluorescence emission spectroscopy. The small fragment, A-II (segment 96-133 of bovine growth hormone), undergoes a helix to random coil structural transition between pH 5 and 10 (pKa = 7.15). At pH9, the random coil state of A-II reverts back to helix conformation as ionic strength increases from 0.01 to 1. The A-II fluorophore, Tyr-110, is quenched by a neighboring carboxyl group of Glu-111, but is only slightly affected by the secondary structural transition. The large fragment, A-I (segments 1-95 and 134-191, connected via a disulfide linkage, of bovine growth hormone), is a rigidly structured molecule with a large amount of beta-sheet structure. Trp-86 of A-I was found to reside in an aromatic and hydrophobic amino acid cluster which is only destroyed by a high concentration of denaturant. Based on the primary sequence of bovine growth hormone, conformation predictions were made using the Chou-Fasman method ((1974) Biochemistry 13, 222). Bovine growth hormone helical structures are predicted to be in segments 10-34, 66-87, 111-127, and 186-191, beta-Sheet structures are predicted to be in segments 45-54, 90-94, 101-105, 136-142, 161-165, and 174-179. Tetrapeptides 37-40, 41-44, 60-63, 129-132, 146-149, and 156-159 were predicted to be beta turns. The prediction scheme confirmed several spectroscopic observations, but it did not completely explain the behavior of bovine growth hormone peptide fragments.     

Use of specific lysine modifications to identify the site of reaction between cytochrome c and ferricyanide

The site of the reaction between horse heart ferrocytochrome c and ferricyanide was investigated by measuring the reaction rate of cytochrome c derivatives specifically modified at single lysine residues to form trifluoroacetyl or trifluoromethylphenylcarbamyl amino groups. Cytochrome c derivatives singly modified at lysines 8, 13, 25, 27, 72, 79, and 87 surrounding the heme crevice had rate constants decreased from that of native cytochrome c by factors of 1.29, 2.03, 1.12, 1.35, 1.46, 1.29, and 1.19, respectively. Modification of a given lysine with the bulky trifluoromethylphenylcarbamyl group caused nearly the same decrease in reaction rate as modification with the trifluoroacetyl group, indicating that the effect was due to removal of an electrostatic interaction between the protonated lysine amino group and ferricyanide. Modification of lysines 22, 55, 99, and 100 at the right side, bottom, and back of cytochrome c had no effect on the reaction rate. These results indicate that the reaction site is located at the exposed edge of the heme and that the electrostatic interaction between ferricyanide and cytochrome c is dominated by the lysine amino groups surrounding the heme crevice, which include lysine 86, in addition to the ones listed above. We have used the specific lysine modification results to estimate the contribution of each lysine amino group to the electrostatic interaction and have developed a semiempirical relation for the total electrostatic interaction.     

Chemical modification of rhodopsin and its effect on regeneration and G protein activation

The studies reported are concerned with the functional consequences of the chemical modifications of the lysines and carboxyl-containing amino acids of bovine rhodopsin. The 10 non-active-site lysine residues of rhodopsin can be completely dimethylated and partially acetimidated (8-9 residues) with no loss in the ability of the proteins to activate the G protein when photolyzed or to regenerate with 11-cis-retinal. These modifications do not alter the net charge on the protein. Surprisingly, heavy acetylation of these lysines (eight to nine residues) with acetic anhydride, which neutralizes the positive charges of the lysine residues, yields a modified rhodopsin fully capable of activating the G protein and being regenerated. It is concluded that the non-active-site lysine residues of rhodopsin are not importantly and directly involved in interactions with the G protein during photolysis. However, this is not to say that they are unimportant in maintaining the tertiary structure of the protein because heavy modification of these residues by succinylation and trinitrophenylation produces proteins incapable of G protein activation, although the succinylated protein still regenerated. The active-site lysine of rhodopsin was readily modified and prevented from regenerating with 11-cis-retinal and with o-salicylaldehyde and o-phthalaldehyde/mercaptoethanol, two sterically similar aromatic aldehyde containing reagents which react by entirely different mechanisms. It is suggested that rhodopsin contains an aromatic binding site within its active-site region. Monoethylation, but not monomethylation, of the active-site lysine also prevented regeneration.(ABSTRACT TRUNCATED AT 250 WORDS)     

Modification of the amino acid residues in the serum albumin of oncological patients

A modified form of albumin isolated from the blood of oncological patients was studied. A modification of N-terminal amino acid, 50% tyrosine groups, free SH-group of cysteine and lysine group in the fourth position in the N-terminal sequence of amino acids was discovered. The possibility of a post-translation modification of serum albumin in disease in individual amino acid groups is discussed.     

EFFECT OF ACETYLATION AND METHYLATION ON THE SWEETNESS INTENSITY OF THAUMATIN I

The lysine residues in thaumatin I were chemically modifiedby acetylation with acetic anhydride and by reductive methylation,under various conditions. The acetylated and methylated thaumatinswere isolated by ion-exchange chromatography. The number ofremaining free amino groups was determined by trinitrophenylation. At least four acetylated thaumatins with either one, two, threeor four acetylated amino groups were obtained as well as onemethylated thaumatin with six dimethyl lysine residues and onemonomethyl lysine residue. The sweetness intensity of the acetylated thaumatins decreasedwith the increasing number of acetylated amino groups; the sweettaste had disappeared completely when four amino groups wereacetylated. The methylated thaumatin with seven modified lysineresidues had a sweetness intensity practically equal to thatof the original thaumatin. The total net change, i.e. the isoelectric point of thaumatin,might play a role in the physiological behaviour of thaumatincausing a sweet taste sensation.     

Biological activity of a fluorescein human growth hormone derivative prepared by specific covalent labeling of lysine-70

Modification of human growth hormone (hGH) with a low equimolar concentration of fluorescein isothiocyanate (FITC) yielded a derivative containing 1 mol of fluorescein/mol of protein. The site of modification was identified as lysine-70. Lysine-70 of hGH is about 3-fold more reactive than a "normal" lysine in a protein, having pseudo-first-order kinetics Kobs = 110 +/- 7 M-1 min-1 at pH 10.5. The pKa of the lysine was estimated to be 10.7, within the normal range of normal epsilon-lysine moieties in proteins. This higher chemical reactivity seems to favor selective labeling of this moiety at low FITC concentrations. To obtain monomodified derivatives, hGH was derivatized with 0.6 equiv of FITC, and the modified derivatives were separated from unreacted hormone by means of HPLC using a Mono Q column. Its biological activity, determined by Nb2 bioassay, decreased to 40%, and its affinity toward lactogen receptors in Nb2 cells and toward somatogen receptors in bovine liver decreased respectively to 30% and 20%. The present study indicates that out of the seven amino groups of human growth hormone, the epsilon-amino group of lysine-70 is excessively reactive toward FITC. Second, this particular amino group contributes to receptor binding and receptor activation. Lysine-70 is located in the loop between the first and second helix and close to the carboxy-terminal end of the first helix. This contribution is most likely the result of the formation of an electrostatic interaction between the hormone and the receptor.(ABSTRACT TRUNCATED AT 250 WORDS)     

Role of individual amino acids of apolipoprotein A-I in the activation of lecithin:cholesterol acyltransferase and in HDL rearrangements

The central region of apolipoprotein A-I (apoA-I), spanning residues 143--165, has been implicated in lecithin:cholesterol acyltransferase (LCAT) activation and also in high density lipoprotein (HDL) structural rearrangements. To examine the role of individual amino acids in these functions, we constructed, overexpressed, and purified two additional point mutants of apoA-I (P143R and R160L) and compared them with the previously studied V156E mutant. These mutants have been reported to occur naturally and to affect HDL cholesterol levels and cholesterol esterification in plasma. The P143R and R160L mutants were effectively expressed in Escherichia coli as fusion proteins and were isolated in at least 95% purity. In the lipid-free state, the mutants self-associated similarly to wild-type protein. All the mutants, including V156E, were able to lyse dimyristoylphosphatidylcholine liposomes. In the lipid-bound state, the major reconstituted HDL (rHDL) of the mutants had diameters similar to wild type (96--98 A). Circular dichroism and fluorescence methods revealed no major differences among the structures of the lipid-free or lipid-bound mutants and wild type. In contrast, the V156E mutant had exhibited significant structural, stability, and self-association differences compared with wild-type apoA-I in the lipid-free state, and formed rHDL particles with larger diameters. In this study, limited proteolytic digestion with chymotrypsin showed that the V156E mutant, in lipid-free form, has a distinct digestion pattern and surface exposure of the central region, compared with wild type and the other mutants. Reactivity of rHDL with LCAT was highest for wild type (100%), followed by P143R (39%) and R160L (0.6%). Tested for their ability to rearrange into 78-A particles, the rHDL of the two mutants (P143R and R160L) behaved normally, compared with the rHDL of V156E, which showed no rearrangement after the 24-h incubation with low density lipoprotein (LDL). Similarly, the rHDL of V156E was resistant to rearrangement in the presence of apoA-I or apoA-II. These results indicate that structural changes are absent or modest for the P143R and R160L mutants, especially in rHDL form; that these mutants have normal conformational adaptability; and that LCAT activation is obliterated for R160L.Thus, individual amino acid changes may have markedly different structural and functional consequences in the 143--165 region of apoA-I. The R160L mutation appears to have a direct effect in LCAT activation, while the P143R mutation results in only minor structural and functional effects. Also, the processes for LCAT activation and hinge mobility appear to be distinct even if the same region of apoA-I is involved. -- Cho, K-H., D. M. Durbin, and A. Jonas. Role of individual amino acids of apolipoprotein A-I in the activation of lecithin:cholesterol acyltransferase and in HDL rearrangements. J. Lipid Res. 2001. 42: 379--389.     

Reactions of the amino groups of RNAse A: III. The existence of pH-dependent values of pK

The rates of the trinitrophenylation of the amino groups of ribonuclease A (RNAse) with the specific reagent trinitrobenzene sulfonic acid have been studied at 27°C, between pH 7.0 and 9.9. From the variation of the velocity constants with pH it has been shown that the reaction is biphasic in the sense that for each amino group two pKs have been found: one (pK = 7.3–7.52) in the range of pH between 7.0 and 8.3 and the other (pK = 9.28–9.69) in the pH range 8.5–9.9. It is pointed out that when the experimental conditions approached one another, there was agreement between the pK values obtained from titrimetric and kinetic studies. Evidence is presented from the literature concerning the validity of the pK value near 7.5 for the ε-amino groups in RNAse. The studies were repeated with performic acid oxidized RNAse and the 10 ε-amino groups were found to be monophasic with pK values between 8.01 and 8.10. The α-amino group of the N-terminal lysine was biphasic with a pK of 7.26 (pH range 7–8) and 8.13 (pH range 8.2–9.5).     

Reactivity and ionization constants of the lysine residues in apovitellenin i of emu egg yolk low-density lipoprotein by competitive labelling.

Competitive labelling with[14C]acetic anhydride over a range of pH values has been used to explore the surface topography of the apovitellenin I moiety in emu egg yolk low-density lipoprotein. The reaction of the lysine xi-amino groups with acetic anhydride has been related to pH in a set of titration curves; from these, the reactivities relative to alanine and the ionization constants of all but the amino terminal lysines have been determined. All lysines have near normal pKa values around 10, and lower than normal reactivities (except the amino terminal lysine). At pH values above 10, the titration curves show breaks where the epsilon-amino groups become much more reactive, except for lysine 71 which in this regard behaves like a normally ionizing lysine in not showing a discontinuity. Most of the basic residues in this apoprotein may occur clustered at the surface of the molecule. This accounts best for the observed low reactivities and pKa values. The amino terminal lysine residue is presumably completely exposed to the aqueous environment.     

Somatostatin analogs which define the role of the lysine-9 amino group

Structure-activity studies of the lysine residue in the highly active cyclic hexapeptide somatostatin analog cyclo(Pro-Phe-D-Trp-Lys-Thr-Phe) confirm the importance of the lysine amino group for biological activity through the loss of activity seen on replacement of lysine by ornithine, arginine, histidine and p-amino phenylalanine. Three analogs containing thialysine, gamma- and delta-fluorolysine were equipotent to the parent as inhibitors of insulin, glucagon, and growth hormone release. The pKa's of the amino groups in these equiactive peptides ranged from 8.23-9.4. The lack of a correlation between the basicity of the amino groups and the biological activities suggests that deprotonation is not required for biological activity.     

Specific modification of free lysine amino groups of histidine decarboxylase from Micrococcus sp. n. by trinitrobenzene sulfonic acid]

It has been found that 14 lysine residues are accessible for trinitrobenzene sulfonic acid (TNBS) in the molecule of histidine decarboxylase (HDC). The other 62 lysine residues in the molecule of native HDC are masked and inaccessible for TNBS. It is demonstrated that the SH- and alpha-amino groups of methionine are not modified by TNBS. A correlation between the decarboxylase activity of the enzyme and the degree of its trinitrophenylation has been studied. HDC, whose molecule contains 3--9 TNP groups, retains up to 90--97% of its initial activity. Trinitrophenylation of 14 lysine residues induces inactivation of HDC by 33--34%, which probably depends on conformational changes or steric hindrances, occurring in the catalytic site of the modified active centre of HDC. Using circular dichroism and fluorescence methods as well as disc-electrophoresis in polyacrylamide gel, it has been shown that trinitrophenylation does not cause any significant changes in the enzyme structure. The TNP groups have been found to be localized in the large and small subunits of the HDC molecule.     

Proteasome-mediated degradation of p21 via N-terminal ubiquitinylation

We examined the mechanism responsible for the degradation of p21, a negative regulator of the cell division cycle. We found that p21 proteolysis requires functional ubiquitin and Nedd8 systems. Ubiquitinylated forms of p21 and p21(K0), a p21 mutant missing all lysines, are detected in vivo and in vitro, showing that the presence of lysines is dispensable for p21 ubiquitinylation. Instead, the free amino group of the N-terminal methionine of p21 is a site for ubiquitinylation in vivo. Although wild-type p21 is more abundantly ubiquitinylated than p21(K0) mutant due to the presence of internal lysine residues, their rates of proteolysis are indistinguishable. These results demonstrate that proteasomal degradation of p21 is regulated by the ubiquitin pathway and suggest that the site of the ubiquitin chain is critical in making p21 a competent substrate for the proteasome.     

N-terminal basic amino acids are not required for translocation and processing of preproparathyroid hormone

An N-terminal deletion mutant of preproparathyroid hormone that contains a single basic amino acid, lysine, in the N-terminal domain of the signal peptide is translocated across the endoplasmic reticulum membrane similarly to intact preproparathyroid hormone. To examine the function of charged residues preceeding the hydrophobic core, the lysine was replaced by an uncharged (methionine) or negatively charged (glutamic acid) amino acid. The translocational activity of the mutant signal peptides was assayed in a reticulocyte lysate system containing chicken oviduct microsomal membranes. Altering the net charge of the N-terminal domain did not abolish signal sequence activity, although the efficiency of translocation was decreased for the mutant with a glutamic acid substitution. Posttranslational, ribosome independent, translocation was observed for all the mutants tested, with the same dependence on N-terminal charge but with much lower efficiency than cotranslational translocation. These studies show that the presence of basic amino acids in the N-terminal domain of a eukaryotic signal sequence is not required for its activity.