Affinity capillary electrophoresis study of the linkage existing between proton and zinc ion binding to bacitracin A1

 Funny & Jokes

 5 views
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Description
Measurements by capillary electrophoresis (CE) of bacitracin A1 effective mobility at different pH values permitted to estimate the five acidic dissociation constants and the Stokes radii at different protonation stages of the macrocyclic
Share
Tags
Transcript
  Massimo Castagnola 1,2 Diana Valeria Rossetti 1 Rosanna Inzitari 1  Alberto Vitali 2  Alessandro Lupi 2 Cecilia Zuppi 1 Tiziana Cabras 3 Maria Benedetta Fadda 3 Raffaele Petruzzelli 4 Bruno Giardina 1,2 Irene Messana 3 1 Institute of Biochemistryand Clinical Biochemistry,Catholic University of Rome,Rome, Italy 2 Institute for the Chemistryof Molecular Recognition,National ResearchCouncil (C.N.R.),Rome, Italy 3 Department of Sciences Applied to Biosystems,Cagliari University,Cagliari, Italy 4 Department ofMedical Biosciences,“G. D’Annunzio” University,Chieti, Italy  Affinity capillary electrophoresis study of thelinkage existing between proton and zinc ionbinding to bacitracin A  1 Measurements by capillary electrophoresis (CE) of bacitracin A  1  effective mobility atdifferent pH values permitted to estimate the five acidic dissociation constants andthe Stokes radii at different protonation stages of the macrocyclic dodecapeptide.The p K  a  values were 3.6 and 4.4 for the two carboxylic groups of the lateral chains of D -Asp-11 and  D -Glu-4, respectively, 6.4 for the aza-atom of the imidazole ring ofHis-10, 7.6 for the amino group of  N  -terminal Ile-1 and 9.7 for the   -amino group of D -Orn-7, very close to the values obtained by other researchers by titration experi-ments. Inagreement witharigidmacrocyclicstructurethe Stokesradii ofdifferentpro-tonatedformsrangedonlybetween14.3and14.8Å.Bestfittingproceduresperformedon experimental mobility measured at two different pH values (5.50 and 6.72) in thepresence ofincreasingZn  2 concentrationallowedconfirmingthemodelthatassumesthe binding of Zn  2 to P 0 peptide form with a 1.5  10 3 M  1 intrinsic association con-stant. Following to Zn  2 binding, the p K  a  of the amino group of  N  -terminal Ile-1 isshiftedfrom 7.6to 5.9and the Stokesradius isreduced ofabout3Å. Themean chargeof the bacitracin A  1 -Zn  2 complex resulted   1.67 and   1.12 at pH 5.50 and 6.72,respectively. These results suggest that the amino group of  N  -terminal Ile-1 is notessential for Zn  2 binding. Keywords:  Affinity capillary electrophoresis / Bacitracin / Zinc ion EL 5280 1 Introduction Bacitracins are a class of structurally related macrocyclicdodecapeptides produced by  Bacillus subtilis  and  Bacil- lus licheniformis , with antibiotic activity against Gram-positive organisms [1]. Commercial bacitracin prepara-tions comprise a mixture of many closely related ana-logues, named bacitracin A to I [2], in which bacitracin A  1 (Fig. 1) represents the major component (percentageusually greater than 60%) with the highest activity [3].Bacitracins are not produced by ribosomal synthesis, butthey are formed throughout the assembling of differentfragments by a large synthetase complex [4, 5]. Conse-quently, bacitracins show several unusual structural fea-tures, namely four amino acid residues with a  D -chirality,a macrocyclic ringderivingfrom amide bond between the  -amino group of Lys-6 and the  C -terminal carboxylicgroup of Asn-12 and a thiazoline ring deriving from thecyclization of the cysteine side chain [2]. These peculiarstructural features might protect bacitracin from proteasedegradation [3]. The molecular basis of the powerful anti-microbial bacitracin activity is not well understood. How-ever, a divalent metal ion, probably Zn  2 , seems essentialfor its antibiotic activity [6]. The bacitracin-Zn  2 complex,which shows a 1:1 stoichiometry, is considered responsi-ble for the inhibition of cell wall biosynthesis of Gram-positive bacteria. It should inhibit the dephosphorylation Figure 1.  Structure of bacitracin A  1 . Correspondence:  Prof. Massimo Castagnola, Istituto di Bio-chimica e Biochimica Clinica, Facoltà di Medicina e Chirurgia,Università Cattolica, Largo F. Vito 1, I-00168Roma, Italy E-mail:  m.castagnola@uniserv.ccr.rm.cnr.it Fax:  +39-6-3053598  Abbreviation: ACE , affinity capillary electrophoresis Electrophoresis  2003,  24,  801–807 801 © 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 0173-0835/03/0503–801$17.50  .50/0       C      E    a    n      d      C      E      C  802 M. Castagnola  et al. Electrophoresis  2003,  24,  801–807of the lipid carrier intermediate C 55 -isoprenyl pyrophos-phate [7], by forming with it a further adduct with a 1:1stoichiometry.For this reason, many studies were devoted to the deter-mination of the structure and the physicochemical pro-perties of metal ion-bacitracin complexes. They includespectrophotometric (UV-vis; NMR; electron paramagnet-icresonance (EPR); extended X-rayabsorptionfinestruc-ture (EXAFS)) and titrimetric analyses [8–11]. Even thoughthese approaches permitted to elucidate the structure ofseveral bacitracin-divalent metal ion complexes, some doubts regarding the residues involved in the metal binding are still present [11]. In the present study, affinitycapillary electrophoresis (ACE) was utilized in order toprovide information on the thermodynamics of bacitracin A  1 -Zn  2 complex formation as a function of bacitracinprotonation. ACE is based on the measurement of theeffective peptide mobility at different ligand activity. Inits different options, ACE has been largely used for thestudy of macromolecular interactions and recent reviewsare devoted to explain its mechanism [12, 13]. ACE canoffer precious information concerning the charge and thehydrodynamic (Stokes) radius of the peptide in its differ-ent bounded forms and, therefore, with respect to otheranalytical techniques, it allows insight into the thermo-dynamics of the peptide binding mechanism [14, 15]. 2 Materials and methods 2.1 Reagents  All common reagents were analytical grade. Formic acidwas purchased from Farmitalia-Carlo Erba (Milan, Italy),acetic acid from Baker (Deventer, The Netherlands), zincchloridefromSigma-Aldrich(St.Louis,MO,USA),sodiumhydroxide from BDH Chemicals (Poole, England), sodiumchloride and hydrogen chloride (36% w/w) were from Merck (Darmstadt, Germany). Raw bacitracin, MES, HEPES, CHES, and CAPS were obtained from Sigma- Aldrich. The commercial bacitracin from Sigma-Aldrich isa mixture of many similar peptides in which bacitracin A  1 is the main constituent. This commercial sample was pu-rified by RP-HPLC according to [2], in order to obtainbacitracin A  1 . 2.2 Apparatus The capillary electrophoretic apparatus employed wereeither a Beckman P/ACE System MDQ Capillary Electro-phoresis (Beckman Instruments, Palo Alto, CA, USA)equipped with a diode array detector and Gold Nouveausoftware for automated apparatus control and data ac-quisition,oraBeckmanP/ACE2100instrumentequippedwith the Beckman Gold Series 711 software for auto-mated apparatus control and data acquisition. CElect N fused-silica capillaries were provided from Supelco (Belle -fonte, PA, USA). The HPLC apparatus was a BeckmanGold 125S solvent module equipped with a Model 168diode array detector and Gold Nouveau software. ThechromatographiccolumnwasapreparativeAlltech(Deer-field, IL, USA) Hyperprep PEP-C18, with 8   m particlediameter (dimensions 260  10 mm) protected by a C18preparative refillable guard column. The pHmeter wasthe Orion model 420 (Beverly, MA, USA). 2.3 Methods The electrophoretic runs were carried out with CElect N(Supelco) silica capillaries, 27 cm length, 20 cm at thedetection window, and 75   m ID. The constant appliedvoltagewas10,12and14kVandtherunningtemperaturewas 20  C. Sample solutionswere usuallyprepared from astock solution(6.8 mg/mL) diluted to a finalconcentrationof 1.8 mmol/L with the running buffer. Common buffers(sodium formate, pH 2.6–4.3, sodium acetate, pH 4.5–5.9, MES, pH 6.4–6.8, HEPES, pH 7.3–8.2, CHES,pH 8.8–9.3, CAPS, pH 9.7–10.9) of constant 10 mmol/Lionic strength were prepared in the pH range of 2.60–10.93 according to reported procedures [15–17] andadded with 30 mmol/L NaCl. Therefore, the total ionicstrength of the buffers used was always 40 mmol/L. Dur-ing the experiment of Zn  2 binding, ZnCl 2  was added ata concentration ranging from 1.0 to 10.0 mmol/L andNaCl concentration was accordingly decreased to main-tain a constant ionic strength of 40 mmol/L. Injections,performed by pressure for 1 s, corresponded to about4.0 nL. The reference detection absorbance wavelengthwas 214 nm. Mobility values were always the mean of atleast triplicate measurements, some performed at differ-ent separation voltage. Effective bacitracin A  1  mobilitywas obtained subtracting the value of the electroosmoticflow, measured by acetanilide. Experimental electropho-retic mobility values were analyzed according to leastsquare fitting procedures utilizing the equations de-scribed in Section 3. 3 Results Figure 2 reports the effective bacitracin A  1 mobility meas-ured atdifferentpHvaluesintherangeof2.60–10.93.Theeffective mobility values were determined by subtractingthe electroosmotic flow (EOF) mobility, determined bythe acetanilide marker, from the observed mobility. Asan example, Fig. 3 shows two electrophoretic runs per-formedatbasicandacidicpHvalues.Mobilitywasmeas-  Electrophoresis  2003,  24,  801–807 ACE study of Zn-bacitracin binding 803 Figure 2.  Experimental effective mobility of bacitracin A  1 (    ) at different pH values. The continuous thick curvewas obtained by fitting experimental data according toEqs. (1)and (2).Thincurves represent the theoretical con-tribution to the total mobility of the five bacitracin A  1 charged species. Figure3.  Electrophoretic runsofbacitracin A  1 performedat pH 5.00 (bottom) and 10.93 (top). For separation con-ditions refer to Section 2.3.ured using different running buffers having constant ionicstrength,intheaimtoavoid theinfluenceofionicstrengthmodifications (see Section 2). Analysis of the dependence of bacitracin A  1  mobility frompH permitted to obtain informationon peptide proton dis-sociation constants. In fact, since proton binding kineticsis fast with respect to the electrophoretic separation time,the effective mobility measured can be considered asweighted mean mobility of the  k   different protonatedspecies [14], such that:  ep    q 6   i  ki  1  Z  i  r  i  x  i  (1)where  q  is the electron charge (1.60  10  19 C),    is theviscosity (8.95  10  4 N  s  m  2 at 25  C), and  Z  i ,  r  i  and  x  i are the electrolyte elementary charge (dimensionless),the Stokes radius and the molar fraction of the  i  -th spe-cies, respectively. The ionizable residues of bacitracin A  1 in order of increasing proton affinity are the following:(i) and (ii) the two carboxylic groups of the lateral chainsof  D -Asp-11 and  D -Glu-4, (iii) the aza-atom of the imida-zole of His-10, (iv) the amino group of  N  -terminal Ile-1and, (v) the   -amino group of  D -Orn-7. The aza-atomof thiazoline ring is not titrable over pH 2.5 [9], whereasthe   -amino group of Lys-6 and the carboxylic group of C -terminal Asn-12 are amide-bound. Therefore, the fiveionizable groups can generate six (   k   ) different chargedspecies with a charge ranging from  3 to  2. The molarfraction of these different charged species is linked to five K   macroscopic dissociationconstantsand to H  concen-tration by the following relationship:  x  i   H     k  i     i  1    i  ki  1 H     k  i     i  1   (2)where   (i  1)  is obtainable by:   i  1       j  i  1 j  0 K   j  (   0  = 1) (2   )Introducing Eqs. (2) and (2’) into Eq. (1) and performing abest fitting procedure on the experimental data reportedin Fig. 2, the acidic dissociation constants were obtained(Table 1). They are in a satisfactory agreement with thevalues obtained by Scogin and colleagues [9] by titrationexperiments, which are reported in the table for compar-ison. The best fitting procedure indicated that the Stokesradius of the different charged species is comprised be-tween 14.3–14.8 Å in the whole pH range investigated.These results are in agreement with a rigid peptidemacrocyclic structure. Figure 2 also demonstrates thedifferent contributions of the five charged species to theexperimental effective mobility obtained using the baci-tracin A  1  acidic constants of Table 1.The study of Zn  2 binding to bacitracin A  1  was performedat pH 5.50 and 6.72. One of the ACE runs performed at8mmol/LZn  2 ,pH5.50,isreportedinFig.4.Theeffectivemobilityvaluesmeasured atdifferentZn  2 concentrationsat the two chosen pH values are shown in Fig. 5. Toensure a constant ionic strength during the experiments  804 M. Castagnola  et al. Electrophoresis  2003,  24,  801–807 Table 1.  Proton dissociation constants, Zn  2 associationconstants and Stokes radii of all the bacitracin A  1  species, obtained by fitting experimentaldata of Figs. 2 and 5 according to Eqs. (1–5) Bacitracin A  1 species(Proton dissocia-tion) p K  a) (Zn  2 binding) K  ass  ( M  1 )Stokesradius (Å)P  3 3.6  0.2 (4.1) 14.3  0.2P  2 4.4  0.2 (4.4) 14.5  0.2P  1 6.4  0.2 (6.7) 14.6  0.2P 0 7.6  0.2 (7.7) (1.5  0.2)  10 3 14.5  0.2P  1 9.7  0.2 (10.0) (7.5  0.4)  10 4 14.4  0.2P  2 7.5  10 4 (n.d.) 14.8  0.2(PZn)  2 5.9  0.2 11.5  0.4(PZn)  1 9.7 (n.d.) 11.8  0.4(PZn) 0 11.8 (n.d.) a) Inparentheses the values obtained by titrationaccord-ing to [9] are given.n.d., not determinable (see Section 3)of Zn  2 binding, the NaCl used in the metal free experi-ments (30 mmol/L) was progressively decreased accord-ing to the increase of ZnCl 2  concentration. Due to enoughfast binding kinetics, the mobility determined at differentZn  2 concentration can be also fitted according to thegeneral equation (1). However, in the presence of Zn  2 ,other charged species, deriving from the interaction ofthe metal ion with the different charged bacitracin A  1  spe-cies, should be considered as contributes to the effectivemobility. Figure4.  Electrophoretic runsofbacitracin A  1 performedat pH 5.50 in the absence of Zn  2 ions (top) and at8.0mmol/LZn  2 (bottom). For separation conditionsreferto Section 2.3. Figure 5.  Experimental effective mobility of bacitracin A  1 (    ) at pH 5.50 and 6.72 in the presence of different Zn  2 concentrations.Thecontinuouscurvesrepresenttheoret-ical mobility values obtained by fitting experimental dataaccording to Eqs. (1–5). The dashed curve representsthe theoretical effective mobility at pH 6.72 assumingthat the p K  4  value does not change on Zn  2 binding.Using the data obtained on free peptide as reference, thesimplestmodelable tosatisfactorilyfitexperimental data, thatcontemporaneouslyprovidedacceptable Stokesradii, indicated that Zn  2 ion binds to the uncharged bacitracin A  1  (P 0  ), thus generating a double-positive charged com- plex.Thiscomplexhasstilltwodissociablegroups,namelythe N  -terminal amino group of Ile-1 and the   -amino groupof  D -Orn-7, with K   4  and  K   5  acidic constants, respectively.The equilibria present in solution can be described by thescheme: P  3 K  1   P  2 K  2   P  1 K  3   P 0  PZn   2 K  4  K   4  P  1  PZn   1 K  5  K   5  P  2  PZn  0 (3)Interactionbetween P 0 andZn  2 isregulatedbytheeffec-tive association constant: K  0ass   PZn    2   P 0    Zn  2    (4)Utilizing this constant and the acidic dissociation con-stants of Table 1, the molar fraction of each peptide spe-cies can be obtained by relationships similar to Eqs. (2)and (2   ). For instance, the molar fraction of (PZn)  2 isobtainable by:  x  PZn    2   H    2 H    2  K   4  H      K   4 K   5   K  0ass  Zn  2    3  H    2 H    5   1  H    4   2  H    3   1   K  0ass    Zn  2    3  H    2   4  H       5 (5)  Electrophoresis  2003,  24,  801–807 ACE study of Zn-bacitracin binding 805Molar fractions of the other forms can be calculated fromequations similar to Eq. (5). Best fitting procedures per-formed utilizing Eq. (1) and considering the species ofEq. (3) provided the theoretical curves of Fig. 5. From thefit an association constant for the (PZn)  2 complex (  K  0ass  )of 1.5  10 3 M  1 was obtained. Moreover, the Stokes radiiof (PZn)  2 and (PZn)  1 complexes were found to be 11.5and 11.8 Å, respectively (Table 1) suggesting that thebinding of metal ion to bacitracin A  1  generates a morecompact structure with respect to that one of the freepeptide. The best fitting procedure indicated also that asensible modification of the  N  -terminal amino group dis-sociation constant occurred after the formation of thecomplex, being the p K   4  value about 5.9 (Table 1). Onthis assumption, the mean charge of the complex com-puted at saturating Zn  2 amounts resulted   1.67 and  1.12at pH5.50and6.72,respectively. Thecurve ofthe-oretical mobility at pH 6.72, obtained supposing that the N  -terminal amino group did not change its acidity aftermetal ion binding, is reported in Fig. 5 as a dashed line,showingtheinabilityofthisassumptiontofitexperimentaldata.The linkage existing between the acidity dissociationconstants and the Zn  2 binding association constant atthe P 0 form allowed to obtain the values of the  K   1ass  and K   2ass  constants (Table 1), namely the Zn  2 binding asso-ciation constants at the P  1 and P  2 forms, respectively.In the absence of experimental information, the  K   5  acidicconstant of   -amino group of  D -Orn-7 was supposedunchanged after Zn  2 binding and the Stokes radius of(PZn) 0 was assumed equal to that one of (PZn)  1 . Thestandard free energy variations of the reactions shownin Eq. (3), which involve the nine charged bacitracin A  1 species, are reported in Fig. 6.In the pH range 4–8,the bindingof Zn  2 ionto the P 0 formof bacitracin A  1  provided a contemporaneous release ofprotons, measured in studies of other groups by titrationexperiments [9]. In fact, if Zn  2 concentration increasedall the positive charged forms are shifted towards the P 0 form with the loss of their protons. Nonetheless, due tothe low peptide concentration (about 0.8 mmol/L) withrespect to the buffering capacity of the separation buffer,this phenomenon cannot affect mobility measurements.Themolesofprotonreleasedpermoleofpeptideatsatur-ating amounts of the metal ion were computed fromtheconstantsreported inTable 1.OurresultsarereportedinFig.7incomparisonwiththoseobtainedbyScoginandcolleagues by titration experiments [9]. Also in this figurea dashed line represents the moles of proton releasedper mole of peptide under the not acceptable hypothesisthat  N  -terminal amino group did not modify its acidityconstant on Zn  2 binding. Figure 6.  Standard free energy variations (   G 0  ) of thereactions occurring in solution involving the six Zn  2 freeand three Zn  2 bounded bacitracin A  1  species, obtainedfrom the dissociation and association constants reportedin Table 1. Figure 7.  Moles of proton released for moles of bacitra-cin A  1  at a saturating amount of Zn  2 (continuous line).The dashed curve represents the theoretical values ex-pected, hypothesizingthatthep K  4 valuedoesnotchangeon Zn  2 binding. (    ) Experimental values obtained fromtitration studies of Scogin and colleagues [9].The dependence of bacitracin A  1  effective mobility onZn  2 concentration (Fig. 5) can be also described by anapparent pH-dependent association constant  K  H . Theinverse of this apparent constant corresponds to the
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks