Interfacial binding of secreted phospholipases A2: more than electrostatics and a major role for tryptophan

428

Interfacial binding of secreted phospholipases electrostatics and a major role for tryptophan Michael H Gelb*, Wonhwa Secreted

A, have

phospholipases

vastly different their binding vesicles

interfacial binding affinity for different

by several

structure/function interfacial binding enzymes

are

orders principles and the

beginning

similar

Chotand

catalytic

sites,

The

that dictate physiological

both the differential function of these

to be unraveled.

Addresses *Departments of Chemistry and Biochemistry, University of Washington, Box 351700, Seattle, WA 98195. USA; e-mail: [email protected] iDepartment of Chemistry (M/C 11 l), University of Illinois at Chicago, 845 West Taylor Street, Chicago, IL 60607-7061, USA; e-mail: [email protected] iDivIsion of Biochemistry and Molecular Biology, School of BiologIcal Sciences, University of Southampton, Bassett Crescent East, Southampton SO1 6 7PX. UK; e-mail: [email protected] Current

Opinion

in Structural

Biology

1999,

Interfacial binding, crystal structures

9:428-432

http://b~omednet.com/elecref/O959440X00900428 8~8Elsevier

Science

Ltd ISSN

0959-440X

Abbreviations EPR electron paramagnetic resonance dissociation constant % sPLA2s secreted phospholipases A,

Introduction Sccrctcd kl)a).

h>~drol~zc rcleasc

the a f‘rcc

cicosanoid

(sPl,A’s)

calcium-rfepcndent .i//-2 f.:rtt!

biosynthesis.

According Figure

A,

phospholipases disulfide-rich,

to amino

are small enxymcs

cstt’!-

of

gll’“erophospholil,icl

acid.

such

;IS :lrachidonic

and

acid scqucnccs

:i

I~sop~~os~~l~~~lipitf

and diaulfide

(-1-I rhat to

acid

David C Wilton and positioning, nine groups of sPLAZs have been described [2,3]. As these enzymes act on naturally occurring phospholipids that have virtually no soluhility in water, they must bind to the membrane interface to gain access to their substrates. ‘I-he purpose of this review is to highlight some recent developments in our understanding of the molecular basis of the association of sPI,AZs tvith the membrane interface.

but

surfaces that modulate kinds of phospholipid

of magnitude.

for [ 1).

numhor

A*: more than

catalytic

site binding

and

‘1‘1~ fact that sPI,.4Zs can act in a highly processkre scooting mode. in which the enzyme remains bound to the lxxiclc and undergoes man); lipolysis reaction cycles. proves that the interfacial binding of the enzyme to the membrane (E -+ RX) and the binding of a substrate moleculc to the catalytic site of the enzyme at the interface (!L” + Sx + K**S”) are distinct steps [4,5]. ‘I’he crysul strllctures of pancreatic, venom and human nonpancrcatic sl’1,42s demonstrate a high degree of structural homology [h]. All contain a catalytic site slot that extends from one surface of the onqme to rhc opposite face (Figurc 1 ). Recent structural studies using a nolrel, electron paramagnetic resonance (Rl’K)-based membrane docking technique have proven that the surface of the enzyme that s\lrrounds the opening to the catalytic site slot constitutcs the interfacial binding surface (i-face) [7”] (l;igure 1). Interfacial binding is thought to seal the cnzymc to the membrane. in order to facilitate phospholipid transfer into the caralycic site Slot. ‘I’llus, the structural studies pro\,ide strong circumstantial evidence for separate the intorfacial binding and catalytic site loadin:: steps. ‘Ike relati\,e importance of electrostatic and

1 1

The I-face of bee venom sPLA2. Stereo pair deplction of the X-ray structure of bee venom sPLA2 [321. A short-chain phosphoiipid analog bound in the catalytic site slot is shown as a black ball-and-stick structure. Residues that lie on the face that includes the opening to the catalytic site slot are basic (arginine and lysine; dark gray), hydrophobic (mid gray) and hydrophilic (hlstidine and threonlne).

Interfacial

hydrophobic interactions in interfacial binding must ultimately depend both on the physicochemical composition of the membrane and on the structure of the i-face.

Quantification

of interfacial

binding

of secreted

phospholipases

A,

Gelb, Cho and Wilton

429

example, tryptophan-containing sPI,AZs give strong spectral observables upon binding to anionic vesicles, but binding to phosphatidylcholine vesicles does not always give a response (MH Getb, unpublished data).

binding

Measuring the affinity of sPI,AZs for membrane interfaces can be quite difficult, especially when interfacial binding is high affinity. Furthermore, the products of phospholipid hydrolysis can greatly promote the binding of sPI,42s to zwitterionic phosphatidylcholine vesicles (set, for example, IS]). (:onditions that prevent hydrolysis include the use of nonhydrolyzable phospholipid analogs, such as diether phospholipids, the absence of calcium, the Lose of catalytic site mutants that lack lipolysis activity and the u\c of polymerized phospholipid liposomes 18,9’,10-l3J. Also. it must he remembered that the binding of the substratc within the catalytic site of the interfacial bound enzyme is a component of the apparent intcrfacial 6, (dissociation constant) value. ‘l‘hc USC of polymeri/.cd phospholipid liposomes or the excluGon of calcium prevents the E” + S” + K*eS” step. so that only the E + E:” step is measured. hlost techniques for measuring interfacial binding rcquirc either physical separation of the bound and fret enqmc or the USC of a spectral change that is sensitive to bindinK. Physical separation LLnder ccluiiibrium conditions involves centrifugation to pellet the phospholipid aggregate. ‘l’his separation is only possible w/ith heavier phospholipid aggregates, such as sucrose-loaded large Lmilamcllar vesiclcs or membrane fractions. ‘l’hese \,esicles arc fragile, however. and may not survive centrifugation; the use of polymerized phospholipid \,csiclcs is hclpfLL1 in this regard. I!nfortLmately, small unilamellar vesicles, which arc commonly Llsed for enzyme assays, do not readily scdimcnt. I’hospholipid-CoatcLl beads provide an alternative for presenting a stable phospholipid surface that can bc readilv sedimented [ 111, bL[t it is difficult to know at the prcscnt time whether such artificial surfaces fully reproduct the interfacial protein-binding properties of bilayercd vesicles. Spectral methods for measuring interfacial binding do not rcqLLirr physical separation, but may require the LISC of nonphysiological reporter groL]ps. ‘I‘he availability of :I tryptophan residue (‘Ii-p.3) on chc intcrfacial sLLrf:Lce of the pancreatic enzyme has been Llsed. and the f]L~orcsccncc change upon interfacial binding can be csplaincd by dcsolvation. as the indolc sidechain inserts into the interface [ IS]. An alternative approach in\,olves the detection of rcsonancc energy transfer bctwccn cn/.ymic tryptophans and a fluorescent phospholipid. sLlch as lV-dansy]-pllos.. phatidylethanolamine. that is prcscnt at a fe\l, nlolc percent in \.esicles [lS,lh]. A major problem hvith thcsc methods is that not only mLlst the enzyme contain a tryptophan, btlt the efficiency of the energy transfer is high]) vari:Llilc. depending on the location of the flu()rcsccnt lipid probe in the vicinity of cn/ymc at the intcrt:dcc. ];or

‘I’here is a fundamental difficulty in quantifying interfacial binding if the value of K,i is very low (< lo-’ M). For example, the sensitivity of fluorescent methods is not sufficient to detect the binding of subnanomolar concentrations of enzyme to < 10-7 hl lipid. l’his is a significant problem for the binding of sPI,.43s to anionic \,esicles, for which values of K(, are thought to be lO-x hl or smaller [ 14,151. Improved methods for measuring tight interfacial binding are needed. ‘I’he method of centrifuging polymerized phospholipid vesicles has been taken to its limit of measuring A;/ values in the l-10 I&I range by very careful c]LlantifiCati(Jn of the free enzyme in the supernatant in the presence of tow micromolar concentrations of phospholipid [ 171.

Amino acid determinants of sPLA2 interfacial binding as probed by site-directed mutagenesis Extensive mutagenesis studies of catalytic residues in bovine and porcine pancreatic SPINS have been carried out (see, for example, [ 18.19]), but the following discussion will focus on stLLdies that probe the contribution of i-fact residues to interfacial binding. A hallmark of sPI,AZs is their relatively high affinity for anionic versus neutral \,csicles. Most sPI,AZs bind weakly to phosphatidylcholine vesicles (K, in the approximately millimolar range, except as noted below), and binding to anionic phosphatidylmethanol vesicles is high affinity (h’(,< 1OV M) [14,1.5]. ‘I-he i-faces of sPI,AZs contain a collar of hydrophobic residues that surrounds the opening to the catalytic site slot and two or more cationic arginine or lysine residues (FigLLrc 1). As a result of these basic residues. the i-faces of sPI,AZs have a positive etectrostatic potential [20], which could explain the high affinity of these enzymes for anionic interfaces. 13ovinc pancreatic sPI,AZ has 3-l basic residues located on its pL)tative i-fact. Of thcsc, KlO, h;Sh and Kl 16 were shown to make a contribution to interfacial binding, as mutating them individually to glutamates resulted in a loto 1X)-fold reduction in the affinity of binding to anionic phosphatidylglycerol vesicles [21]. A similar effect of charge-rr\-ersa1 mutagenesis was foLmd for the sl'I,& from the venom of the water moccasin snake. in that a singlc. basic-residue mutation resulted in a lo- to 30..fold reduction in anionic vesicle binding [ 101. ‘I’hc double mutant h;7E/I\;lOE: bound SW-fold weaker than the wildtvpc CO phosphatidylglyccrol vesicles, corresponding to a 3.7 kcal/mol loss of intcrfidcial binding energy. ‘l’he location of these two basic residrlcs matches the position of maximal positive clectroststic potential on the surface of the venom #1,X2. IJikc the venom enqime. hum:ln group II:1 sl’I,AZ contains basic residues at positions 7, 10 and l( 1. lvhich are clearly on the i-face. ‘l‘he hLrman enzyme

430

Lipids

contains 11 additional basic residues that lie on rhe same side of the protein as the i-face, but that appear to be ptlshed back away from the planar surface of the enzyme that includes residues 7, 10 and 16 and the opening to the The group Ila triple mutant catalytic site slot. K7E/KlOIC/K16IX binds 300-fold weaker than the wildtype to phosphatidylglycerol vesicles and mutation of groups of basic residues that make up the 11 additional positively charged sites results in a lo- to ZOO-fold reduction in interfacial binding 1171. ‘I’he five-site mutant K7~/K101~/K1h~/K1218/K1271~ binds 2200-fold weaker than the wild-type to anionic vesicles [Z’]. All together. these studies show that the interfacial binding of these sPIA3s has a strong electrostatic component, but other factors, besides electrostatics, are at play. I:or examplc, the five-site group 11~1 sI’I,hZ mutant mentioned above has an o~xzrall negatively charged i-face and yet it \rill displays a rcspectablc binding affinity for anionic xesiclcs (k;, -7 phi). ‘l‘he sl’IA7 from hone), lxx \ cnom has the typical hydrophol~ic-resid~le collar (see above and FigIIre 1) and six basic residues on or near the i-fact. ‘I’he A,, for this enzyme bound to anionic phosphatid~lmethLlno1 J esicles is probably < lo-” Al. because the enqmc operates on these \ esiclcs in the scooting mode (procchsi\-c catalysis) 01x3 scvcral minlites. \cithollt lca~ ing the surface 12.31. Remarkably. a multisite mutant, in which five out of six basic rcsiducs wxxc changed to ,glutamatcs, still opcratcd tjn anionic vcsiclcs in the scooting mode ]‘,‘I. Although it i5 possible that the intcrfacial KC, for bet venom sl’lA2 was incrcascd hy charge-re\,er\al mutation (\xIurs of i\;, h:i\ e not been measured to date becatisc of their low magnitndc). it is clear thar the binding of bee \xnom sl’l,:22 to anionic \,esicles is not tlri\xzn bv electrostatics. ‘l’hlls, there mubt be other fac,tors besides ionic intcrxtions that go\ cm the rclati\xzly high affinity of sl’l,AL~ for anionic LCI-SIIS zu itterionic vesicles. I>iffcrences in the packing propertic\ of l~hosl~h:ltidylcholinc \ erslls phosi’h~ltid!-lniethanol ma!. alter the ability of enz~-mc rcsiducs to pcnctrate into the intcrfxial region of the bila!cr. but s~ich ;I working h\,pothcsis is difficult to c\ aluatc cxlxximcntall~. tZ ncu. magnetic resonance technicluc hay been rcccntl! dc~cloped to dock pcriphcral membrane proteins at the interface. l Cxjntact the interface. ‘I’his result pro\:itlcs 3 structural cxptanation for the modest effect of charge-rc\xzi-sat mut;~Kcncsis on interfacial binding (discussed abo\ cl. ‘l’ltc rtzx)cluc5 that ill-c in dil-cct contact L\itll rhc intcrfacc arc hydrophobic. ‘1:1kcn togcthcl-. the result\ suggest that

hydrophobic residues are better able to penetrate into the interfacial region of phosphatidylmethano1 vesicles than into the interfacial region of phosphatidylcholine vesicles. l-or human group Ila, water moccasin and pancreatic sPI.,4Zs. one might anticipate, according to the mutagenesis studies described above, that the basic residues are in direct contact with the interface, but there are no direct structural data that relate to this issue. Not al1 sPI,A& bind weakly to phosphatidylcholine vesicles, A hallmark of the cobra venom sI’lA2s is their relatively high affinity for zwitterionic vesicles (KCl for A,‘ujc/ inr///n///~z~r(~ sI’l,.42 of X x 10-7 hl [&I)). These enzymes have l-3 tryptophans located on their putative iface\, and numerous early studies have shown that chemical modification of tryptophan drastically lowers the acti\,ity of these enzymes on pl~osphatidylcholine vcsiclcs (sec. for example. [2.5,&i]). ‘I’herc is mounting cvidencc that tryptophan is a key residue in supporting lipid-protcin interactions. t-or cxamplc. spectroscopic studies have shown that simple indole analogs preferentially partition into the interfacial region of phospholipid hilaycrs (the region that includes the ;giyccrol backbone. the cstcrs that link fatty acyl chains to glycerol and the first 1-L (:H, Kroups of the fatty acyl chains) 1271. ‘l’he powcrfuI cffcct of tryptophan on interfacial binding is highlighted I)> studies with a murant of human group IIa sl’lA2 containing tryptophan replacing ;I Ixline located on the N-terminal helis that forms part of the i-fact. ‘I’hc \..iC\ mutant i5 200- to 1000-fold more acti\x than the wild-t!lx on phosl~hatidylcholinc \,csiclcs [2X**]. Nthoiigh thcrc arc no direct interfacial binding data a\xilablc for these proteins. the large kinetic effect of addinK the i-fact tryptophan is probably the result of enhanced \,esiclc affinity, because both the \r,ild-type and the mutant sho\t comparat~lr activity on anionic vesicles. Also, whereas wild-t\-1x group Ila slJl~,4.! dissociates from phosphatidylmethanol vesicles with ;I half-time of a few minutes, the \..:I!’ mutant remains bollnd for more rhan 30 min and undergoes intcrfacial catalysis in the scootinK mode (AlK Jain. 1xrs”nal conliiilinic;ition). I,ikewisc. bovine pancreatic. ~I’IA.? opcratcs in the scooting mode on anionic \ csiclcs, t)ut the \\:.3:2 nitjtant sho\\,s intorvcsic-lc csch;ingc o\‘er minutes [ S]. ‘l‘hese results shon, that tr)l)tophan aIso enhance\ intcrfacial binding to anionic \,csicles. Rcccnt stud&. in M hich cobra \ cnom trvptophans ha\.c been mutated, support the chelnical modification studies and show that the indolc ring makes :I siiI~atantial contribution to interfacial binding (LV (:ho CrN/.. personal cominunicatic,il). ‘I‘hc rccentl\, disco\xxd human group \’ sl’IA2 beha\.es \crv diffcrcnrlv than the group Ila cnzvme with respect to intert’xial binding. \Vhcrca\ group Ila st’IA.2 displays c\trcmrl\ poor activic\ on phosl’hatidvlcholinc \xxiclcs. the group \’ enz\mc is 70(N)-fold more :icti\ c :~nd both cn/yn~cs bind tightly to anionic xesiclcs ].W]. ‘I’he i-face of ~ri~iip 1. #IA.! contain5 ;I trvptophan and mlich fe\Vcr

Interfacial

basic residues than the i-face of group Ila sPLA2. Mutation of the i-face tryptophan to alanine renders group V sP122 W-fold less active on phosphatidylcholine vesicles [30’].

binding

8.

of secreted

phospholipases

A, Gelb,

431

Cho and Wilton

Othman R, Baker S, LI Y, Worrall AF, Wilton DC: Human non-Pancreatic (group II) secreted phospholipase A, expressed from a synthetic gene in Escherichia co/i: characterisation of Nterminal mutants. Biochim Siophys Acta 1996,1303:92-l 02.

9. .

Conclusions

and future

directions

WC are beginning to understand the molecular determinants of interfacial binding by sPLAZs. Clearlp for some enzymes, electrostatic interactions involving basic XIlinO acids play an important role in allowing high-affinity binding to anionic vesicles. Further work is needed to understand the role of hydrophobic residues in supporting interfacial

binding.

~1’1 ,X2. interfacial ‘I’hib for

especially

‘liyptol)han binding

i\ most which

and

\vith

interfacial

that

carr!.

arc go\xxned

~xxqxxtirs.

out

mcInt)raneb.

(;roup

\’ sl’l,A2

11~1 erlq~me

groiip

mcmhrancs

[.%Y].

grtJul>

‘I‘hilS.

on

arachidonic

\’ acid

References

the

arc

more

rich

in

actixc

than

published

within

/\\,itccrionic player

13.

LI Y, Tsat M-D: Phospholipase A, engineering. histidine catalytic diad also plays an important J Am Chem Sot 1993,115:8523-8526.

14.

Kim Y, Llchtenbergova kinetic and binding beads. Anal Biochem

15.

Ramlrez Proteins

F, Jaln MK: Phospholipase 1991~ 9:229-239.

A, at the

16.

Burack WR, Gadd ME, Blltonen A,: identification of an inactive B;ochemMry 1995, 34:14819-l

RL: Modulation membrane-bound 4828

1 7.

Smtko Y, Kodun RS, Han SK, Othman R, Baker SF, Molini B, Wllton DC, Gelb MH, Cho W: Mapping the interfacial binding surface of human secretory class Ila phospholipase A,. Biochemfstry 1997, 36:14325-I 4333.

18.

Kulpers OP, van den Bergh CJ, Verhelj HM, de Haas GD: Probing the mechanism of pancreatic phospholipase A, with the aid of recombinant DNA techniques. Adv Exp Med Biol 1990. 279:65-84.

19.

Sekar K, Btswas R. LI Y, Tsar M-D. Sundarallngam M: Structures the catalytic site mutants D99A and H48a and the calcium-loop mutant D49E of phospholipase A,. Acta Crysfallogr D - Bb/ Crystsllogr 1999, 55:443-447.

20.

Scott DL, Mandel for the interfacial B!ophys J 1994,

lipid>. in rcleaqing

reading the annual

Annand RR, Kontoylanni MH: Active site of bee histidine-34, aspartate-64, 35:4591-4601.

the cell

~xxx~iicticm.

and recommended

12.

binding

of mammalian

I~B\ Ix ;I key

for eicosanoid

Papers of parttcular Interest. have been hIghlIghted as:

ib that

interfacial

leaflet

Yu B-Z, Berg OG, Jain MK: The divalent cation is obligatory binding of ligands to the catalytic site of secretory phospholipases AP. Biochemistry 1993, 32:6485-6492.

functions

sl’l.hl seemsbest fit to

is moth outer

v hich

sl’IA-7

i-&x

[.il]. as this cwvine disl)lavs interfacc5. siich a% hactcrial

protein anionic

11.

multiple

ph~sic,lo,gical IIa

Han SK, Yoon ET, Scott DL, Slgler PB, Cho W: Structural interfacial adsorption. J Biol Chem 1997, 272:3573-3582.

\wiclcs.

hypothesis

in their

group

10.

vesicles.

contain

working

venom

lacking

\lammals

1)~ cliffercnces

act a\ an antibacterial ;I hil?;h affinity for

bee

in promoting

of sI’IA2s

different

example.

I.‘or

of

charge-ncutl-al

bincling

is IOU affinity. ~OI-ms of sPl,A7s. A rcasonal,lc cnqmrs

case rcsicluc

phosphacid~lcholinc

tr~l~tol>hans these

the

as a ke);

co anionic

aI)parent

the

in

emerges

Ghomashchl F, Lin Y, Yu B-Z, Annand RA, Jain MK, Gelb MH: Interfacial recognition by bee venom phospholipase AZ: insights into nonelectrostatic molecular determinants by charge reversal mutagenesis. Biochemisfry 1998, 37:6697-6710. This study provides clear evidence for the importance of a nonelectrostatic component In Interfacial binding.

period

of review. 21.

of special interest **of outstanding interest

l

1.

Kudo I, Murakaml phospholipase

2.

Dennis EA: The growing transduction enzymes.

3.

Cupilard L, Koumanov K, Matt’ei Cloning, chromosomal mapping, human secretory phospholipase 272.15745-l 5752.

4.

lain MK. Rogers J, Jahaglrdar DV. Marecek JF, Ramlrez R: Kinetics interfacial catalysis by phospholipase A2 in intravesicle scooting mode, and heterofusion of anionic and zwitterionic vesicles. Bmchun Bmphys Acta 1986, 860:435-447.

5.

Gelb MH, Jaln MK, Hanel AM, Berg 0: The interfacial enzymology of glycerolipid lipases: lessons from secreted phospholipases Anr~u Rev Biochem 1995, 64:653-688.

6.

Scott

DL, Slgler

secretory

M, Ham S, lnoue K: Mammalian A,. Blochlm B,ophys Acta 1993, phospholipase Trends B~ochem

PB: Structure

phospholipases

non-pancreatic 1170:2 17-231.

and

catalytic

for the

M, Penzotti JE, Dudler T, Lybrand TP, Gelb venom phospholipase A,: the role of and tyrosine-87. Biochemistry 1996. IO. The asparatestructural role.

L, Snitko Y. Cho W: A phospholipase assay using phospholipid-coated 1997, 250:109-l 16.

AM, Slgier PB. Homg binding of secretory 67:493-504.

bilayer

A2 hydrophobic interface.

of phospholipase state.

B: The electrostatic phospholipases

Dua R, Wu S-K, Cho W: A structure-function study pancreatic phospholipase A, using polymerized-mixed liposomes. J Biol Chem 1995. 270:263-268.

of

basis A,.

of bovine

Koduri RS, Baker SF, Snitko Y, Han SK, Cho W, Wllton DC, Gelb MH: Action of human group Ila secreted phospholipase A, on cell membranes: vesicle but not heparinoid binding determines rate of fatty acid release by exogenously added enzyme. J B/o/ Chem 1998. 273:32142-32153. This study shows that the residues that are Important for the InterfacIal bindtrig of secreted phospholipases A, to vesicles are important for the action of the enzyme on living cell membranes.

A, superfamily of signal So 1997, 22:1-Z.

mechanism Chem 1994,

of

22 .

MG, Lazdunskl M, Lambeau G. and expression of a novel A,. J Biol Chem 1997,

A,. Adv Protein

aspects

23

Yu B-Z Ghomashchl F. Cafal Y, Annand RA. Berg OG. Gelb MH, Jaln MK: Use of an imperfect neutral diluent and outer vesicle layer scooting mode hydrolysis to analyze the interfacial kinetics and substrate preferences of bee venom phospholipase A,. Biochemistry 1997, 36:3870-3881.

24.

Jaln MK, Egmond MR. Verhelj HM, Apltz-Castro R, Dqkman Haas GH: Interaction of phospholipase A, and phospholipid bilayers. Biochim Biophys Acta 1982, 688:341-348.

25.

Chang LS, Kuo KW, Chang CC: Identification of functional involvement of tryptophan residues in phospholipase A, from Naja naja atra (Taiwan cobra) snake venom. B;och/m Biophys Acta 1993. 1202:216-220.

26.

Takasakl C, Sugama A. Yanagita A, Tamlya N, Rowan EG, Harvey AL: Effects of chemical modifications of Pa-l 1, a phospholipase A, from the venom of Australian king brown snake (Pseudechis ausfralis), on its biological activities. 7i,x/con 1990, 28:107-l 17.

of

A,.

of 45:53-88.

7. ..

Lln Y, Nielsen R. Murray D, Mailer C. Hubbel WL. Robinson BH, Gelb MH: Docking of phospholipase A, on membranes using electrostatic potential-modulated spin relaxation magnetic resonance. Science 1998. 279.1925-l 929. This article describes the only known method for docklng peripheral membrane prolelns on membranes and determlmng thelr depth of penetration.

R, De

432

27.

Lipids

K, Asuncron-Punzalan E, London E: Anchoring of tryptophan and tyrosine analogs at the hydrocarbon-polar boundary in model membrane vesicles: parallax analysis of fluorescence quenching induced by nitroxide-labeled phospholipids. Biochemistry 1995, 34:15475-l

28. ..

5479.

SF, Othman R, Wilton DC: Tryptophan-containing mutant human (group Ila) secreted phospholipase A, has a dramatically increased ability to hydrolyze phosphatidylcholine vesicles and cell membranes. Biochemishy 1998,

Baker

tryptophan-enhanced

rnterfacial

of

the structural

30. .

mutagenesis,

and functional

roles of N-terminal

31.

Weiss

activity

L, Munoz

NM,

Leff AR, Wilton

in high membrane binding of human group V phospholipase

Chem 1999,271 :I 1881-l 1888. underscores the fact that different mammalian A, can have distinct inter-facial binding properties. J, Wright

G, Bekkers

AC, van den

Bergh

secreted

CJ, Verheij

DC,

and A,. phos-

HM:

Conversion of pig pancrease phospholipase A, by protein engineering into enzyme active against Escherichie co/i treated with the bactericidal/permeability-increasing protein. f Biol Chem 32.

residues

X-ray, and NMR. Biochemistry

3417322-7334.

Han SK, Kim KP, Koduri R, Bittova Gelb MH, Cho W: Role of Trp31

J Biol This study pholipases

brnding

Liu X, Zhu H, Huang B, Rogers J, Yu B-Z, Kumar A, Jain MK, Sundaralingam M, Tsai M-D: Phospholipase A2 engineering.

Probing

1995,

proinflammatory

37:13203-13211. Thus study illustrates the powerful, of secreted phospholrpases A,. 29.

with site-directed

Kachel

1991,

266:4162-4167.

Scott

DL, Otwinowski

bee-venom analogue.

Z, Gelb

MH,

Sigler

PB: Crystal

phospholipase A, in a complex Science 1990, 250:1563-l 566.

structure of with a transition-state