Chen! stry

THE STRUCTULE GF POOP THS:) TUCO HYDROGEN-BONDED
HaLiIc..L COUPIGUE. TIONS OF WHE POLYPPP. TUE Ch.

By Linus Pauling, Robort Bb. Corey, and Hy Re Brenson*
Gates and Crellin Laboratorios of Chenistry,
Californis Institute of Techn logy, Pusedenz, Cult forniat
Communicated February 28, 1951

During the pest fifteen years we heve been attseline the roblem of the
structure of protedne in seversl ways. One of thear vays is the complete and
“eeurite Cetoruinition of the erystcl structure of aino ectcs, peptides, and
othor sinple substances relited te protcing, in order thot inform:tion chout
intorstomle Cistences, bond angles, und other confisur- tional paraactors
mipht be obt ined thet would vermit the relishle predietion of ressoncble con
floum.tions for the polypeptide chain. We hove now used this inform:tion to
construct tuo reisoncble hyéropenebonded helie: 1 conff;uritions for the poly-
peptdce chain; ve think that it is lfkely thet these conti:ur tions cong tite
an dmportent pert of the structure of both fibrous ond ¢lotilar ordt- ing, us
well ss of aynthatic polypeptides. | le‘ter enmouncing their discovery uss pob-
lished I:st year.

The vroblem thst we hove set ourselves is thet ef PMdine UL heororen
bondud structurcs for a ginple polypeptice chaing in wiich the residues are
ocuivulont (exeept for the <ifferenees in the side chain BR). on azine seid
reaidue (othe: than elyeine) has no symmetry clecentas The comeral oporetion
of conversion of ene residue of a single eahein inte » seeond reatdae coulw lent
Go the Sipst fe cecordingl]y «© rotelion about an aia ceoonp ented by tronolution

alon the oxige Uenee the only conSi-um ions for 2 coin eompotilde ulth our
postulate of equivalence of the residues are holical conftrurstions. For rota~
tional angle 150° the helices] configuretions m:y cegencrate te 2 airple chain
with sll of the princip:l «toms, C, C' (the con foty ston), Ny ami G, in the
same plone.

We aseume thet, because of the resonance cf the double bond between the
ecrbomo:gyeen and carbon-nitrosen pocltions, the confiruretion of each residue

HA

o/ Re is planar. This structure] fe: ture has been verified for each of

the guides that we have studied. Moreover, the resenenoe theory ie now so vell
grounded ond ites experd-ental asubstantiotion so orton sive that these con te no
doubt whatever about ite upplicstion to the Antde croup. The observed Call
cistance, 1e32 a » corresponds to neorly fifty percent double-bend churactcr, and
we mey conclude that rotation by ss much og 10° fron the plunar confirurs tion
vould result in instebility by aewerwle 3-051 mile”, The interatomic distances
and bond angles within the residue are assumed to have the values shown in
Firore 1. These veluca heve been form ated” by consideration of the experi~
mental values found tn the eryatal structure atudics of DL-slanine ? Imthreonine,*
Neucetyllycine,? and anplyeylgl- cine” that hove beer made in our Luborstorics.
Tt is further gsmmed thit eech nitrogen ctem forms a hy voren bond with an oxy-
gen ston of another residue, with the nitrogen-oxyren distence ecunl to 2e72 |
and thet the vector from the nitroren atom to the hycrogen-londed oxygen atom
diss not more than 30° from the Neel direction. ‘he enerry of am Newdlee ena
hycrocen bond is of the order of 8 keal. nole™, an’ gugh grect inst bility
vould rogult from the failure te form those boncs thet we sey be confident of
their presence. The Neewllees0 dlatunce cormot be expected te be exuetly 2672 i,

bat wight devint< somevhst from this value.
oe

Solutden of this problem shovs that there are five and only five oonfigure~
tions for the chain that satisfy the conditions other than that of direction of
the hydrogon bond relative to the Heli dircetieon. ‘These correspond te the vale
uce 165°, 120°, 108°, “ie, and 0.9° for-the for the rotation.] angle. In the
first, third, umd fifth of these structures the 00 croup ia noystively and the
a group positively circetec alen: the helices] axde, talon as the dircetion
corresponding to the sccucnce —CHR—CO——Ni——Cihe of atoms in the peptide ohain,
and in the other two their cirections are reversed. The first three of the
structur:s are unsatisfactory, in that the Heel croup doca not extend in the dix
ection of the oxygen atom at 2672 a; the fourth and fifth are actisfactory, the
angle bitweon the Nw vector and IO voctor being." Wom as? for these tye
structures, respectively. The fourt’ structure hea 3.69 amine acid residues per
taum in the helix, and the f41fth structure bas 5.05 rosicues ocr turm. In the
fourth structure each xoide croup is hycregen-bonded to the third frdide proup
beyond 1t slonz; the helix, end in the Mfth structure eseh is tonded to the
fifth Amide -roup beyond 4¢) ve sic1l coll these structures cither the 3.7-resd-
due structure und the 5Seleros' due atructure, respectively, or “he third-fdce
hy rogen-bonded structure ant the fifthinide hyiregen-ionded structure.

ihe-eytinericul eserdinates—of the stoms-of one-sseléue_for_aveh-of-the
twe—structurce tre -pivenin Table—T,-né Drowin o and-photogrephs of the fue

structure are showm in Pi;urcs 2, 3, dy and §.
onion

Fer glycine toth the 3e7aresidue helix and the Seleresidue hel4x equld occur
with either a positive or 9 negutive rotations] translation) thet is, ae either
& positive or « neputive helix, relulive to the positive direetion of the helf-
el] axis -iven by the sommonece of stomp in the pepiice chain. For other artino
seida with the | eonfigurstion, however, the positive helix ond the negative
helix would differ in the position of the aide chains, end it night vell be exe
pected that im each case one sense of the helt: vould be nore ateble than the
other, 44 arbitrary aselgnment of the R groups hus been made in Plurce, 20nd.

The trenslict4on slang the helles] scls in the 3e7ercsicue helix te 147 i,
and thet in the $21 residve helix is 0.9 Re The values for one complete turn
are Sed4 i and 45 a, respectively. These viluos sre cileulates for the hycre-
gen~-bond distance 2672 AS they vould have to be increased by s few cereent, in
eige thet 4 lurper hycregem-tond Clatcnce (2620 S » say) vere presente

The atabliity of our belies] structures in 8 nonecrystulline phese denends
solely on intcrections between adjooent resides, and does not recuire that the
number of recidses cer turn be a retio of a all inteyera.s The velue 3669 reai-
dues per turn, for the third-gnide hydrogen-tended helix, is cost closel; ap-
proxintted by 48 residues in thirtecn cums (36693 rosid.ea aor “urn}, enc the
velue 5 for the other helix is most closely approximated by Se res’ dues in
Sdattn tums. It is te be oxpectsd thet the mmbeor of residues nor urn would
be effected somevhet by change in the hydrogen-.ond Cistunce, onc cleo thet the
interzetion of helies] moleoules with neighbertn;: similar relecules in 9 erystal
would esuce acl termes in the hcliwes, deforming then sli bbly inte confimu~
ations with « r.tional mother of resicues per cume Tor the thind-anice hedroe
gen—bonded helix the ailmplest strnetures of this pert chat we would oredict are

the Dlerosidne 3eturn helix (3067 roa’ dues cor urn), the lércsidue 4~turn
~ ie

helix (3075), cnd the léeresidue Sturn helix (3060). We have found some evidence
initesting that the first and third of these slicht vartonts of this helix exist
in crystelline polypeptidese

These helices] atructures heave not oreviously been coseribed. In adcitten to
the extended polypeptide chain configuration, which for nearly thirty years
hag been ussuned to be present in stretched hair and other protedins with the j=
verctin structure, confirurations for the nelypeptice chein hove been aroposed
by -stbury ond Bell,’ und especially by Bugcine® and ty Bragg, Fendrow, «nd Pore
utse? Huge ins discussed « number of structures trvolvine intra-moleculer hy-
drogen bonds, and Bragc, Kendrow, and Peruts extended the <iscussion to include
sdcitional atructurus, end investigated the comp ti bility of the structurss
with xery diffreetion dete for hemo;lobin and myorlobins None of these suthors
proposed el ther our 367-resicve helix or our 5.l-residue helix. On the other
Hind, we would elininute, by our b.sie povtuletes, sll of the structures prom
poced by them. The re:sen for the “ifference in results obtuined by other in~
vestigatora end by us through casentially sindler urmments is thet both Grapy: and
his colleborators and hug; ins clseusaed in detetl only helical structurrs with
an integr] maker of residues per tum, end vo cover «acuned only & rough ap.
proximst to the roqulrencnts sbont interstomic cletences 3 bong enples, and
ploncrity of the conjugated ide croup, 8a siven by our investipctions ef sin~
pler substences, 6 eortend that these stereochentonl fecturce mst be very
el sely retcined in ctuble confimurations of polypcntide coring tn proteins, and
thet there is re special stebtlity osseetated with an integral number of residues
per turn in the hclicul moleevle. Brug, Kencrow, and Perutgs live described a
stracture topelo;deslly gicllar to owr 3e7=residna helix og a hydrogen-bonded
helix with { residues ocr turns Ip their thorowh exrrerigon of their mocels

with Patterson j ojcetions for heaoplehin and myoglobin they elirinoted this
oof
structure, and drow the cautious conclusion th t the evidence favors the none
helicul Zeresidue folded aekerctin conficuration of astbury ond Bell, in which
only one third of the exrbonyl and #eine groups ure involved in intra-molcoular
hyvregen-bond formations

It is our opinion that the structure of aficorstin, afeyosin, and aimilar
fibrous proteins is closely represented by our 3e7-residue helix, ond that this
helix slse constitutes an important structural fexture in hero lobin, myoglobin,
and other glolallar prot«inas, ue well aa of aynthetiec polypeptides. ‘co believe
thet the $Sel—rosidue helix is represented in nature hy supercontreetcd -cretin
and supercontracted myosin.s The evidence leuding us to there conelustions will
be presented in loter piperte

Gur work hoa been aided by grants from The fockefeller Found:tion, The Het.
jonel Foundstien for Infantile Pars] ais, and The U.3. Public Health Service.
Many caloul:tions were curriocd out by Pre 35, “einbaun.

Summary. - Two hycrogem-bonded b.clicel stroctur:s for a pelypcptive chain
hove been found in viich the residues are stereochar call: ecuivelent, the intom
atomic distenees en’ bend ancles heve velues found in emino celiay peosticoa,
unm other sirple substanecs relate! io vroteinc, enc the oonjnceicd Auide systen
ds plancrs In ene streeturo, vith 367 residues ser burn, esch @ rborg) ond inino
group te attcehed by 4 hyvregen bond to the complenentary sreup in the third
Snide croup resovec fron it in the polypeptic¢e chein, and in the otf-cr struc.

ture, with Sel recicuea ner turn, each is bonded to the fiitth p ide craup.
Teo
Referonoes

*Present sddmog, Poword University, Veahington, Dale

teontr: tition Ho. 15386

Tpwiing, Ieey ond Cory, ReBay Jo ge Guu doce, 2s 6219 (1990).

“coroy, habe, ord Donotue, Je, Jo ae Gheme Soces 72, 2699 (1950)

Wavy, Hoses arc) Gorey ReBey Lbhiley G3, 2005 (2041)} Donotwe, Je, ibid,
Tes 9.9 (1950).

4Snoa-er, DeP., Donohue, Je, Schom rer, Vey omc Corey, Eebe, Ladd, Zé,
2328 (1950).

carpenter, GeBe, and Donohue, Jeo, tbide, Ta, 2215 (1960)«

Shoat, Deke, ond Moore, Velde, ipide, J]. 2616 (1949).

7 atiury, ele, ond Nell, Pete, Heture, LUZ, 696 (29/1)6
Srucginsy Mele, Cheme Seve, 2g; 195 (1943).

g . ae a “4 rt es a,
Parsee, Le, Kerwlvw, 3.Coy ond Perwts, MePe, Prooce Hays Sogey 2203. 071 (1950).
Legends for Mepures

Fige le = Dimensionea of the polypeptide cheine
Fige 2e = The helix with 367 reeicGucs per turn.
Fige 3e ~ The helix with Sel resicuocs par Lurte
Fige 4a ~ Plun of the 3e7=resicue holix.s
Fige Se - Plen of the Selerceidue helix,
°

T27A

 

Fige 1. Dimensions of the polypeptide chain.
Fig. 3

On
= A®--- 2G ee Pe Os? Qo
Pk Tok i seehe. nobigo.ebpo—-eo

Qi--0& “i fe y SFOS: oa y8 =o oe £ Er 5 of £
© op 08 ob. ob. 7 8 Koo Q-0 ye oF eee pied J
BG" - Boo BBs; ee hs OBO, GACY
ee = x 2 roo ne So adel Yoo
2 Si iho d2bo eo

ot ee
Fig. 4. Plan of the 3.7-residue helix.
 

Fig. 5. Plan of the 5.1-residue helix.