Theoretical variation of the Hα chemical shift in Acetyl-glycyl-N-methylamide and oligoglycines with molecular conformation and environment

Vol.

171, No.

September

:3, 1990 29,

BIOCHEMICAL

Theoretical

of the

oligoglycines

II,

with

Gresh’

‘Laboratoire 13, rue Pierre

RESEARCH

COMMUNICATIONS

1211-1216

4,

and

environment

and C. Giessner-Prettre’

et, Marie

Thhorique

Pierre

(URA

077)

Physico-Chimique

Curie,

de Chimie

4, place Jussieu,

in Acetyl-glycyl-N-methylamide

conformation

de Biologie

Universite

July

shift

de Biochimie

Institut

’ Labor&&e

chemical molecular

Nohad

Received

BIOPHYSICAL

Pages

variation and

AND

1990

75005 PARIS,

Organique et Marie

Th&xique Curie,

76252 PARIS

FRANCE (URA

Batiment

CEDEX

506)

F

05, FRANCE

1990

Summary. The sum of the magnetic anisotropy and polarization contributions to the magnetic shielding constants of the a protons is calculated as a fur&ion of the t.orsion angles about, the NC, (4) and C,C’ ($) bonds of the dipeptide. The results show that the polarization or ele&ic field effec,t is several fold larger than the magnetic anisotropy contribution. The calculated variations are large enough to account for the spread of the values measured for these protons in peptides and proteins. The results obtained for polyglydne 12 helices and antiparallel ~7 5heet.s are discussed in relation with molecular conformation and environmental effectson the one hand and experimental dataon the other. 01990 llcademic *rear. 1°C. Introduction.

In peptides

for the determination equation,’

their

for a given

amino

little

attention.

from

which with

as well

es intermolecular

a limited

but

the designat,ion

number

of studies

and ab initio

an :important.

r&es

long-range

make it a good candidate conformation.

and direction

sprad

oxr

several

has shown,

from

parameters.

anisot,ropy’5

correl&x@

the oi proton

The amide

and is highly

p01ar’~,

role in the determination in peptides.

shifts with

the

that we

molecular

as well as

shift is sensitive

group

for an important

of th e a proton

(including

in the peptide’*)

of the H,

shifts

chemical

of the experimental

effects”. empirical

that

calcula,tions”a~‘4,

ppm

have received

contributions’l~lz

side chains present

used

the use of Iiarplus

for e~a.mple~-‘~)

of “environmental

conformational magnetic

through

also to the dependence

effect of the aromatic

under

semi-empirical

magnit,ude

in solution

have been widely

due to the absence of B clear, well established

relation,

(de)shielding

molecular

resonances

have experimental

conformation

the value of the backbone exhibits

conformation

shifts which

This is due mainly

shall put together However,

if the a proton

acid ( - 2.6 ppm in the case of glycine

on intra

particular

of the polymer

chemical

shift/molecular quantity

and proteins

to

of these nmlecules two characteristics

In order conformation

of the variation, to determine

the

we undertook,

Vol.

BIOCHEMICAL

171, No. 3, 1990

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Figure 1. Representati6n of the glyc,ine oligomers dad aeetyl-glyeyl-N-metl,~l~de).

for these nuclei, contributions (Fig.1).

the computation

investigated

( n=l

of the sum of the magnetic

in the case of acetyl-glycyl-N-methylamide

represents the stm-

anisotropy

and polarization

and of few higher

order

oligomers

Computational input and procedures. The proton magnetic shielding constant is calculated as the sum of two terms, namely the sum of the local or atomic magnetic susceptibility t.ensors anisotropy effects’ plus the polarization or electric field effect.‘8 p = 02 + $ This limitation is fully justified in the case of H, of glycyl residues of (Gly),, peptides since there is no aromatic, ring present in the systems that we consider and this type of proton being not directly involved in short-range strong interactions as those engaged in hydrogen bonding, does not require the taking into account of the charge transfer plus exchange term.” Furthermore, in the case of proline, it has been found that these two terms account correctly for the shift variation of the 01 and p protons with the value of +‘.I4 The magnetic anisotropy effect is calculated as a sum of atomic contributions, the susceptibilit,y tensor of the amide group atoms being calculated as pre~eedingly.‘~J~ For the polarization contribution uE, of prolix; the electric field created at the molecule(s) considered, is computed procedure”’ which builds the molecular multipolar expansion (up to quadrupoles) entity si,udied. However, the multipoles OL carbon carrying the proton considered

calculated using the formula retained in the case the mid CR bond by the electron distribution of from the set of multipoles used in the SIBFA electronic distribution from the superposition of a of ab initio wave functions of fragments of the of t,he centers located on t,he bonds involving the are not introduced in the summation.

The values actually reported on the Figures and in the Tables are those of the (de)shielding (in ppm) produced by the atomic magnetic susceptibility tensors and by the electronic distribution of t,he molecule. Therefore for the three quantities uk: uE and rT, increasing positive values correspond to an increased shielding while negative values correspond to downfield shifts. Before discussing the results obtained we would like to stress that since the present calculations do not include the local contributions to uT the absolute values reported are not numerically relevant while on the contrary their variation with the molecular conformation or intermolecular interactions aa well as the shift difference between the two cr protons of a given residue should, within the limitations (which remain to be delineated) of t,he computational procedure, be fully representative of the corresponding experiment,al quantities.

The L+ and aT isoshielding clearly

(Fig.

2) that

the corresponding evidence

the range

quantity

the dominant

contours

for ac~etyl-glycyl-N-methylamide

of variation

of cT IS more

t,han three

for 01 (3.6 and 1.1 ppm respectively),

effect of the polarization 1212

contribution.

H,o a result

In addition

proton

times

larger

which

show than

puts into

the comparison

of

Vol. 171, No. 3, 1990

BIOCHEMICAL

-180

-90

90

0 cp

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

180

Figure 2. Isoshielding contour maps in acetyl-glycyl-N-methylamide as a function and .+ angles: a). Total shielding c~T. b). Magnetic anisotropy component ux.

these isoshiehlings

with

those of the magnetic

shows that the two contributions

can be of similar

I for a number of typical conformations both shifted downfield by 0.4-0.6 ppm proton

Ham, this is due to both

latter;

for Ham,

respect

to C5 reflect

shorter

a very small amount the length

which,

of the helix

residue

Both

the chemical occur upon

passing

is increased,

from

Table

input,

H aD

contribution

in Table

shifts found

proton

and one carbonyl

H,D

is less shielded

are For

from the

downfield

in C7 with

than

oxygen,

Ham by only

n=3

I

taking

extensions

dipeptidr

fairly

encompassing

and $J=-57^,

energy for the Gly

d $

Aaca’

constants, and H,D.

from n=l

contribution,

stationary.

on amplitudes

o helical

overall

2.20 -0.62 2.82

1.81 -0.76 2.57

3.22 -0.25 3.47

2.20 -0.62 2.82

1.73 -0.62 2.34

0.61 -1.01 1.62

0.00

0.08

2.61

anisotropy,

of c with

rouformations nhclix

1213

variations

evolutions

C!7

their

up to n=13 uE, whereas

The largest

c’s

n7‘ and

for H,D

of -0.4 ppm and 0.6 ppm for

of the helix entail

in the C5, C5, awl

6% crE

Values (in ppm) of the shielding contributions for protons H,t ‘a)A~= UN,,, aH,o

length,

by the polarization

ax, remains

to n=5,

Glycyl

values of Qi=-47”

II show that crT decreases for Ham and increases

term,

Further

uniform

lead to the best conformational

6’

H OL

The

each H,

keeping

are commanded

protons.

-

between

in o helices of increasing

evolutions

shift anisot,ropy

these respective

The values report.ed

a more important

In the C7 conformer,

the values of Table

of the central residues.

side chainsr2

(0.1 ppm).

given our set of standard

tridecapeptide,

magnitude.

ux and cE: with

distances

effect of the aromatic

show that in the C7 conformer the o protons with respect, to their position in the C5 one.

this is due to the sole aE term.

01 for Ham and Oe for H,o.

When

anisotropy

of the 4

ux.

and polarizntion,

bE,

Vol. 171, No. 3, 1990 a more limited

BIOCHEMICAL

amplitude

for these respective which

amounts

to the further n=13.

(attaining

protons).

opposed

evolutions

RESEARCH COMMUNICATIONS

-0.15 and 0.30 ppm upon

On the other

to 2.4 ppm for n=3

The extension

AND BIOPHYSICAL

hand

passing

the largest

value

drops by 1 ppm upon passing of gE for Ham and H,D,

of the helix

length

to n=5,

decreases

up to 21 residues

from n=5

to n=13

of Ao=~H~L-uH,D and then, owing

down

to 0.9 ppm for

does not modify

significantly

this result. Information

concerning

from

2D NMR

recent

F of CAMP cecropin

receptor

protein

values of Au

ations/variations

tridecamer.

the possible

a hydration

reported

whereas

theoretical

Au values

reconciled that

with

proton

terms,

with

a larger

protons

belonging

bonding

region,

Experimental BUSI, ppm,

prot,ons.

t.o the two bracketing

residues,

III show

shifted for glycine Growth

clearly

that

Table II. Vahws of the noclcar amino arid rrsidue embedded

fLc

K?D

H,D

1 3.19 -0.26 3.45

3

UT ux

0.88 -0.93

uE A0

In

constants,

we

helix with

its

of values.

terms.

shifted,

As a result,

the

and can be readily

From Table

a downfield

which

downfield

is also downfield

III we see also

shift of 0.35 ppm for

the anisotropy

one; on the other

and polarization

hand,

are not. inside

within

p sheet stretches

(Y’, cecropin one found

calculated

the Ha~/Ha~

the bidentate

H-

in proteins,

e.g.,

A ‘, are in the range

for a helices.

for the unhydrated

i

-0.19 3.56

5 2.95 -0.15 3.10

1.81

1.01 -0.89 1.90

2.31

2.36

3.3i

effects.

a large

-0.2 and 0.1 ppm,

magnetic shielding constants of protons H,L in a helicrs of increasing length, encompassing residues (qk-4i”, $=-57’)

3 ux 2

n

residues Factor

to the corresponding if the values

shielding

by fluctu-

upfield.

Transforming

a range identical

Proton

from both

the former

values of 6H,

human

or environmental

has undergone

j? sheet produces

from

are negligibly

the exper-

be modified

and polarization range

This shift results

contribution

H,L

between

experimental

of the antiparallel

Hu~/Ha~

‘,

of 0.9 ppm is calculated

on the H,

term.

anisotropy

are now comprised

the corresponding

the formation

the central

of both

3, helix

values.

III show that

less, because

side chain

residue,

t,o the first shell, of the Gly tridecamer

shift of 1.2 ppm due, again, to the polarization but significantly

a value

of ACT could

of the latter

is available

of myohemerythrin

that for a given

this value

above-mentioned

in Table

show

neighbouring

incidence

C$and $ angles held at their The values

studies

0.15 ppm”,

study, limited

stretches

‘, the lac repressor

6! a nonadecapeptide

However

of the q& $ angles,

to evaluate

undert,ook

lo These

in cy helical

such as BPTI

5, CYpurothionin

never exceed

in the best. Q helical

for Gly residues

of small proteins

A ‘, and rat gallanin.

imental

order

the values of&H, studies

The

helix

values of Table lead to an unac-

and 11,~ of a central frcw n=l to n=13

-0.19 3.OG

9 2.87 -0.22 3.08

II 2.81 -0.22 3.02

1.58 -0.85 2.43

1.73 -0.87 2.59

1.74 -0.89 2.63

1.80

1.84

-0.89

-0.89

2.69

2.74

1.37

1.14

1.13

1.01

0.90

1214

2.8i

3.7-4.2

13 2.74 -0.22 2.96

Vol.

171, No. 3, 1990

BIOCHEMICAL

TaLle

AND BIOPHYSICAL

111. Glycyl

RESEARCH COMMUNICATIONS

triderapeptide a). a helix

Unhydrated Residue

i- I 2.77 -0.22 2.99

JLI.

JJd

helix

Hydrated

i

it-l 2.77 -0.22 2.99

2.74

-0.22 2.96

helix

i-l 1.61 -0.27 1.88

i 1.51 -0.2.5-l 1.76

i+l 1.82 -0.26 2.08

1.81

1.R4

1.83

1.53

1.67

l.i6

-0.90

-0.89

-0.90

-0.96

-0.96

-0.95

2.71 0.96

2.74 0.90

2.73 ON

2.48 0.08

2.F3 -0.16

2.71 O.OG

b). p sheet MOllOlllPI Residue

i-l 1.99 -0.52 2.71

II aL

Antiparallrl

i 1.99 -0.72 2.71

iS 1 1.99 -0.72 2.71

i-l 2.02 -0.77 2.79

dimer i 1.65 -0.95 2.60

i+l 2.02 -0.76 2.79

Values of the nuclear mmgnrtic shielding constants of 11,~ and 11,~ protons of the central residue (i=7) and of the two residues upstream and downstream of it. Investigated conformations arc: a). A regular a helix (a#~= -47”, $ = -NO), et‘th er unhydrated or hydrated with a first shell of water nw~lccules. h). A p sheet conformer, in both the monomeric and antiparallel dimrric forms (same notations as in Table II). For the p sheet conformer, only the H,L proton shifts CUP given, since the valrws of the corresponding H,D are identical

ceptable

discrepancy

both

protons

H,

the /3 sheet. subject latter, with

with

Further

with

in such structured

the hydrated

one give for

for the protons

of

1.5-1.8 ppm versus 1.7-2.0 ppm in the

measured

experimentation

should

those concerning

shielding,

the corresponding

to available

elaborations

flexibility

data,

are very close to those computed

to the intermolecular

in agreement, respect

experimental

values of GT which

values.

lends credence

This additional

to the stability

incorporate

the effects of local

oligopeptide

stretches,

consistency

of our procedure.

sequence

and conformational

and work is under

progress

along t,hese

lines. Conclusions.

The results

underline

the instrumental

the value

of H,

of glycyl length

proton

shielding

oligopeptides,

constants

effects.

anisotropy

region

contributions

to nuclear

magnetic

procedure,

an accurate

nuclear addition,

magnetic

shielding

it can easily

time an accurate,

initio” trqat

confers

We wish

of the polarization

SCF multipoles formulae

oligomers evaluation

14 ,

in the determination

of

the sensitivity

of aH,

shift of H, protons

to chain

which

are inside

of long-range/environmental

to underline

that with

term is warranted.

based on non-empirical size, whilst

of both intramolecular

conformation

its effect adds up to t,hat

used by the SIBFA

of very large

1215

those of ref.

In the o helical

the importance

shielding.

constants

simultaneous

that

a downfield

confirming

representation

to “ab

contribution

In the p sheet conformation,

term entailing

H-bonding

in line with

in peptides.

it, is this very term

the bidentate

by the recourse

investigation,

role of the polarization

and environmental

of the chemical

of the present.

the present

This is enabled

method

and to the

developments.” providing

In

at the same

(conformational)

and/or

Vol.

171, No. 3, 1990

intermolecular to diverse report,ed

BIOCHEMICAL

(environmental) oligopeptides

energies.

and oligomeric

AND BIOPHYSICAL Extensions stretches

RESEARCH COMMUNICATIONS

of this novel SIBFA/Shifts in proteins

are underways

procedure and will

be

separately.

Acknowledgments. It is a pleasure to thank Marie-Claude Schauer and Denis Girou for their kind help into obtaining the isoshielding contours of Figure 2. We also wish to thank the Groupement Scientifique C.N.R.S. - I.B.M. France, for computer time on an IBM 3090, on which the present calculations were performed.

References 1. a)Bystrov, V. F. (1976) Prog. NMR Spectrosc. 10, 41-81; b) Wiithrich, K. (1986) NMR of Proteins and Nucleic Acids, Wiley, New York. 2. a) Pardi, A., Wagner, W., and Wiithrich, K. (1983) Eur. J. Biochem. 137, 445-454; b) Chazin, W., Goldenberg, D., Creighton, T. and Wiithrich, K. (1985) Eur. J. Biochem. 152, 429-437. 3. a) Zuiderweg, E., Kaptein, R., and Wiithrich, K. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 5837-5841; b) Zuiderweg, E., Kaptein, R., and Wiithrich, K. (1983) Eur. J. Biochem. 137, 279-292. 4. a) Strop, P., WdI er, G., and Wtithrich, K. (1983) 3. Mol. Biol. 166,641-667; b) Williamson, M., Havel, T., and Wiithrich, K. (1985) J. Mol. Biol. 182, 295-315. 5. Clore, A.M., Gronenborn, A.M., Brunger, A., and Karplus, M. (1985) J. Mol. Biol. 186, 435-455. 6. Clore, G.M., Sukumaran, D., Gronenborn, A.M., Teeter, M., Whitlow, M., and Jones, B. (1987) J. Mol. Biol. 193, 571-578. 7. Brown, S., Mueller, L., and Jeffs, P. (1989) Biochemistry 28, 593-599. 8. Dyson, H., Rance, M., Houghten, R., Wright, P., and Lerner, R. (1988) J. Mol. Biol. 201, 201-217. 9. Holak, T., Engtrom, A.? Kraulis, P., Lindeberg, G., Bennich, H., Jones,T.A., Gronenbom, A.M., and Clore, G.M. (1988) Biochemistry 27, 7620-7629. 10. Wennerberg, A., Cooke, R., Carlquist, M., Rigler, R., and Campbell, I. (1990) Biochem. Bjophys. Res. Comm., 186, 1102-1109. 11. a) Sternlicht., W., Wilson, H. (1967) B iochemistry, 6, 2881-2892; b) Kalman, J. R., Blake, T. J., Williams, D. H., Feeney, J., Roberts, G. C. K. (1979) J. Chem. Sot. Perkins Trans. 1,1313-1321; Jardetzky, O., Roberts, G. C. K. (1981) NMR in molecular biology, Academic Press, New York. 12. a) Perkins, S. J., Dwek, R. A. (1980) Biochemistry, 19, 245-258; b) Giessner-Prettre, C., Pullman, B. (1981) Biochem. Biophys. Res. Corn., 101, 921-926. 13. Wagner, G.! Pardi, A., W&rich, K. (1983) J. Am. Chem. Sot., 105, 5948-5949. 14. Giessner-Prettre, C., Cung, M. T., Marraud, M. (1987) Eur. J. Biochem., 163, 79-87. 15. Tigelaar, H. I,.: Flygare, W. H. (1972) J. Am. Chem. Sot. 94, 343-346. 16. Kurland, R. J., Wilson, E. B. (1957) J. Chem. Phys. 27, 585-890. 17. Pople, J. A. (1962) Disc. Faraday Sot. 34, 7-14. 18. Buckingham, A.D. (1960) Can. J. Chem. 38, 300-307. 19. a) Giessner-Prettre, C., Ferchiou, S. (1983) J. Magn.Reson. 55, 64-77; b) Giessner-Prettre, C. (1984) J. Biomol. Struct. Dyn, 2, 233-248; (1986) ibid. 4, 99-110. 20. Pullman, A., Berthod, H., Giessner-Prettre, C., Hinton, J. F.: Harpool, D. (1978) J. Am. Chem. Sot. 100, 3991-3994. 21. a) Gresh, N., Claverie, P., and Pullman, A. (1984) Theoret. Chim. Acta, 66, l-20; b) Gresh, N., Pullman, A., and Claverie, P. (1985) Theoret. Chim. Acta, 67, 11-32. 1216