Biochimica et Biophysica Acta (BBA) - Protein Structure最新文献

Rat casein components (C1-, C3A-, C3B- and C4-casein) were extensively purified from rat milk, and the properties of these proteins were compared with those of other caseins including rat C2-casein. C1-casein was precipitated by a low concentration of CaCl₂ (1.5 mM). Both C3A- and C3B-casein were less sensitive to Ca²⁺ than were C1- and C2-casein, and the presence of 20 mM CaCl₂ was required at 37°C for their precipitation. C4-casein was absolutely insensitive to Ca²⁺. This protein exhibited the ability to stabilize all of the other rat casein components against Ca²⁺-dependent precipitation. In addition, C4-casein contained sialic acid, galactose and $\text{N-}\text{acetylgalactosamine}$. Therefore, C4-casein appears to be a bovine ϰ-casein-like protein.

A new rat serum protein has been isolated by affinity chromatography using ethanolamine- or phosphoethanolamine-substituted agarose gels. This protein has the morphological and functional characteristics of serum amyloid P-component and C-reactive protein. It comprises C₅ cyclic symmetry structure with non covalently cross-linked subunits which have calcium-dependent binding sites. We have called this protein rat serum amyloid P-component since it has all the properties typical of human serum amyloid P-component: it is made up of 10 subunits, it contains sialic acid and hexoses, it forms macroscopic polymers and it does not precipitate with pneumococcal C-polysaccharide. Rat amyloid P-component has three remarkable properties. Electron microscopy has shown that apart from pentagonal figures and stacked discs, rat P-component has a C₁₀ cyclic symmetry structure. Rat amyloid P-component has an affinity for specific ligands, such as phosphorylcholine or phosphoethanolamine. These ligands are able to depolymerize self-associated rat P-component. With these characteristics, rat serum amyloid P-component could prove to be an important model in the study of the relations between amyloid P-component and amyloidosis.

The acrylamide quenching of holoazurin B was studied as a function of emission wavelength in order to investigate discrepancies in interpretation of previous fluorescence measurements. A red-fluorescing acrylamide-quenchable component, which may be an impurity, is observed, suggesting that prior studies need to be interpreted with great caution. The presence of this component is not detectable when the 3-fold more fluorescent apo form is studied. Significant acrylamide quenching of apoazurin B is observed. The quenching constant of 5 · 10⁷ M⁻¹ · s⁻¹ at 20°C and the activation energy of 17 kcal/mol obtained are the most extreme values yet reported for a single tryptophan-containing protein. Since azurin B's indole is definitively known to be buried in the hydrophobic interior of the molecule, these results provide additional support for the contention that light-excited proteins undergo structural fluctuations in the nanosecond time range.

The near-ultraviolet CD spectra from 260 to 300 nm of staphylococcal enterotoxins A, B and C were resolved empirically into Gaussian curves. Each spectrum contained the same six components with maximum ellipticities at virtually identical wavelengths. The rotatory strength of the bands of enterotoxin A, however, differed significantly from that of enterotoxins B and C. Each Gaussian curve was identified as corresponding to a prominent CD transition of an aromatic chromophore: two phenylalanine bands from its ¹L_b transition ($\text{0-0}\text{and}\text{0 + 930}\text{cm}{}^{\mathrm{-1}}$); four tyrosine bands from its ¹L_b transition (the 800 cm⁻¹ primary vibronic progression) with the weakest of these overlapping a phenylalanine band; and one ¹L_b tryptophan band. At alkaline pH, tyrosylate CD bands arose, centered at 247–249 nm and at 295–298 nm. The low intensity of the 295–298-nm bands indicated that most of the tyrosine CD in the neutral enterotoxins was contributed by buried residues. The tryptophan contribution to the CD of enterotoxin A was positive, while that of B and C was negative. Also indicative of a different milieu was the ready oxidation of the A toxin by $\text{N-}\text{bromosuccinimide}$ and the unavailability of the single tryptophan residue in the other two toxins to this reagent. The environment of the disulfide bond was markedly diverse in the three enterotoxins. Enterotoxin C was refractory to reduction by mercaptoethanol under conditions where enterotoxins B and A were readily reduced. The contribution of the disulfide of enterotoxin B to its CD spectrum was positive with an intensity of about 9000 deg · cm² · dmol⁻¹ and was centered near 273 nm. The difference spectrum between native and reduced enterotoxin A was much smaller and appeared to be conformational in origin.

The mechanism of precipitation of fibrinogen in the presence of a basic protein, protamine, has been investigated. The precipitation was clearly inhibited by the addition of Fragment D. Photooxidation of fibrinogen abolished the association caused by thrombin, but it did not affect the precipitation brought about by protamine. Not only fibrinogen but also fibrin monomer precipitated in the presence of protamine. These results led to the conclusion that protamine bound directly with D-domains of fibrinogen (of fibrin monomer) to cause the precipitation.

The $\text{M}{}_{\mathrm{<mtext>r</mtext>}}$ 37 000 d-alanyl-d-alanine peptidase excreted by Streptomyces R61 and the $\text{M}{}_{\mathrm{<mtext>r</mtext>}}$ 530 00 d-alanyl-d-alanine peptidase excreted by Actinomadura R39 are both characterized by a very uneven distribution of the basic ($\text{Arg + Lys}$) amino acid residues. Trypsin degradation of the heat-denatured enzymes generates (1) thirteen soluble peptides which contain from 2 to 28 residues in the case of the R61 enzyme and nineteen soluble peptides which contain 2 to 39 residues in the case of the R39 enzyme; and (2) three large segments or core peptides which, irrespective of the enzymes from which they originate, consist of 50–60, 70–80 and 110–120 residues. About 90% of the basic ($\text{Arg + Lys}$) amino acid residues are recovered in the soluble tryptic peptides. The core peptides represent 62% ($\text{M}{}_{\mathrm{<mtext>r</mtext>}}\text{≈ 23 000}$) and 45% ($\text{M}{}_{\mathrm{<mtext>r</mtext>}}\text{≈ 24 000}$) of the untreated R61 and R39 enzymes, respectively. One 28-residue soluble peptide isolated from the R61 enzyme represents the N-terminal portion of the protein whose sequence has been established. The penicillin attachment site of the R61 enzyme has been located in one of the core peptides. For the R39 enzyme, indirect evidence shows that the penicillin binding site is probably within one of the soluble peptides.

The complete primary structure of a lipase (triacylglycerol hydrolase; EC 3.1.1.3) is presented for the first time. The porcine pancreatic enzyme which was investigated is composed of a single chain of 449 amino acids. Upon fragmentation by CNBr, five peptides were obtained. The sequence of four of them (CN I-CN IV) has already been published. The present report deals with the arrangement of the 142 amino acids of the C-terminal peptide CN V, thus completing the analysis of the whole molecule. Special problems resulting from incomplete cleavage of some peptide bonds in CN V and aggregation of large peptides were overcome using Sephadex filtration of succinylated derivatives in 50% acetic acid, antomated sequence analysis of peptide mixtures and subdigestion of material which could not be directly resolved. No obvious homology was found when the sequence of porcine lipase was compared with other protein, including pancreatic phospholipase A2 and colipase from the same species. However, a few similarities which might be significant were detected between the environment and relative position of certain half cystines in lipase and colipase, as well as between two tyrosine-rich regions existing in both proteins.

The rate of tryptic digestion of troponin C has been shown to be dependent on Ca²⁺ (Drabikowski et al., Biochim. Biophys. Acta 490, 216–224). We have characterized the tryptic peptides produced both in the presence and absence of Ca²⁺ using amino acid composition and end-group analyses. In the presence of Ca²⁺ trypsin cleaves TnC at Arg-8, Lys-84 and Lys-88, leading to the formation of two large peptides, one containing the two low-affinity sites (TR₁C), the other, the two high-affinity Ca²⁺-binding sites (TR₂C). In the absence of Ca²⁺ (1 mM EDTA), digestion proceeds much more rapidly and takes place first at Arg-100, followed by Arg-104, Arg-120, Lys-153, Arg-8 and others. The data suggest that the points of cleavage are determined by the Ca²⁺-dependent conformational states of TnC, particularly in the C-terminal half of the protein where the cation is known to induce secondary structure.

The low-frequency regions of the resonance Raman spectra of deoxygenated ferrous forms of soybean leghemoglobin $\text{a}$, horse myoglobin, sperm whale myoglobin, Aplysia myoglobin, stripped normal human hemoglobin (T quaternary form) and of stripped human NESdesArg-hemoglobin (R quaternary form) are reported. Differences observed among these spectra show that the globins of these hemoproteins impose various heme structures. In particular, the variable frequencies of band II (210–224 cm⁻¹) and of band I_b (121–163 cm⁻¹) show that an increase in dioxygen affinity corresponds to a decrease in Fe-N(pyrrole) bond length.

The low-frequency regions of resonance Raman spectra of various low-spin ferrous forms of normal human hemoglobin, soybean leghemoglobin a and of horse myoglobin are reported. Differences observed among the spectra of oxygenated and nitrosyl forms of these hemoproteins show that their globins impose various low-spin heme structures. A quantitative correlation between the variable frequency of resonance Raman band II (215–271 cm⁻¹) and the iron atom-heme plane distance was observed for hemoproteins and heme models, either ferrous or ferric, high-spin or low-spin. From this correlation, the iron atom-heme plane distance should be 0.3 Å in nitrosyl and oxymyoglobin (band II at 256 cm⁻¹) whereas the iron position should be near to or in the heme plane for nitrosyl and oxy forms of hemoglobin and leghemoglobin (band II between 266 and 273 cm⁻¹). A new method is proposed for monitoring the photodissociation processes in ferrohemoproteins.