Group additivity thermodynamics for dissolution of solid cyclic dipeptides into water

11

Z’hermochimica Actu, 172 (1990) 11-20 Elsevier Science Publishers B.V., Amsterdam -

Printed in The Netherlands

GROUP ADDITIVITY

FOR DISSOLUTION

DIPEPTIDES

THERMODYNAMICS

OF SOLID CYCLIC

INTO WATER *

K. P. MURPHY

and S. J. GILL**

Department of Chemistry and Biochemistry University of Colorado Boulder, Colorado 80309-0215

SUMMARY The thermodynamics solids

of dissolution

into water has been studied

energetic

contributions

folding.

The techniques

interpreted

apolar

groups

are described

or enthalpy

observed

compounds

and the results

the

scheme.

protein

are

It is shown that

leads to linear correlations

change with heat capacity

for protein

dipeptide

which determine

in terms of a group additivity

change

have been

in order to establish

of various

this kind of group additivity entropy

of a set of cyclic

denaturation

(either

change)

which

and the dissolution

of

[ll.

INTRODUCTION A knowledge

of the interactions

protein

structures

protein

folding.

many

key features

of a globular

from the anhydrous

liquid amino

phase

initial

involves

interior

phase.

Such studies

acids or amino acid analogues

[2, 31.

Hydrophobic

properties

dissolution

of

hydrocarbon

pure

into solvent

been

examined into

the transfer

water

phase from [4,

TO

whom correspondence should be addressed.

0040-6031/90/$03.50

0 1990 -

Elsevier Science Publishers B.V.

of

the 51.

a

into water Other

* Presented at the 7th International Symposium on Microcalorimetric Applications in Biology, Egham, U.K., 9-11 April 1990, and Dedicated to Ingemar Wadsij on the Occasion of his 60th Birthday. **

and

from

to date have utilized

from an organic

liquids

water,

on the transfer

have included

have

in order to

The unfolding

of amino acid groups

have focused Most studies

structures,

studies

stability.

of the molecule

into water.

and modeling

of protein

compound

protein

the transfer

studies

globular

to understanding

to model

that determine

compound

stabilize

of the complexity

have turned

protein

thus most model some initial

is a prerequisite Because

investigators

probe

which

12 studies

have investigated

gas phase

Considerable interior packing

the thermodynamics

evidence

of a globular density

exists

protein

structures

by X-ray diffraction

a solid model

Consequently,

stability.

Additionally,

the crystal

and this information of results.

dipeptides

and summarize

The cyclic

dipeptides

Figure

1.

The compounds

c(VV),

c(GF),

the letters acids

c(LP),

are discussed

correlations

we have utilized

of these

P=proline,

in

studies

A=alanine,

S=serine,

c(AA),

in

c(LG),

"cycle"

and

for the amino

L=leucine,

and Y=tyrosine).

in terms of a group additivity

scheme and

leads to the entropy

which have been observed

[Is].

illustrated

c(AG),

one letter abbreviations (G=glycine,

it is shown that this group additivity capacity

to be valuable

and c(SY), where the "c" indicates

F=phenylalanine,

The results

of many model

structure

studied were c(GG),

are the standard

seem more

of protein

of a series of solid

the results

all have the basic

in the L conformation

V=valine,

system would

can prove

two techniques

of dissolution

to high

their crystalline

structures

solids are available,

The

is similar

Likewise,

the energetics

in the interpretation

the thermodynamics

[lo].

liquids.

compound

to understand

In this paper we outline

that the

a solid

[123 of proteins

demonstrates

in attempting

cyclic

from the

have been determined

appropriate

studying

to suggest

solids rather than organic

the fact that many protein resolution

however,

is essentially

[ll] and compressibility

to that of organic

nature.

of transfer

[6-91.

for several

- heat systems

[Il.

METHODS ase Ph

alorimetr The molar

enthalpy

the solid to the aqueous

change

for the transfer

phase

can be determined H 0 I

N-

Figure 1. Basic cyclic dipeptide side chain of the amino acids.

of a compound

from

by calorimetry.

E

structure.

R1 and R2 can be any

13

Previously

the predominant

This method

requires

then dissolved developed

in a continuous

an alternative

dissolution

of slightly

perturbation titration

mixture

heat

enthalpy

[16].

[17].

of the crystalline

reestablish

solution.

solution.

The

to reestablish

the

is simply the molar

of moles which must dissolve

to

the equilibrium: (1)

q = v K AHO where

K is the solubility

change

for the transition

constant

solution

and AHO is the molar

from solid to solution.

can be used where necessary saturated

to correct

enthalpy

Heats of dilution

for nonideallity

of the

[16, 181.

Since both K and AH~ vary in a known way with temperature, measurements over a range of temperatures can be used to separate the two terms and also to determine the heat capacity change, ACpO. In practice

however,

of the solubility Differential

when it is feasible,

a separate

refractive

The low solubility

index of many solids makes

reliable

solubility

chromophore

is present

difficult.

the molecule,

optical

appropriate.

In the case of the cyclic dipeptides

chromophore

technique

saturated

index

solubilities

for absorbance

con&entration

measurements

One such technique

is weighed.

Another

(DRI) determination

there

in

are is not a

and another

is a dry weight

from a known volume technique,

of

differential

is more versatile

and yields

over a range of temperatures.

The refractive

absence

such as UV absorbance,

in which the dried residue

solution

refractive

When a suitable

techniques,

is required.

determination

determination

is desirable.

determination

suitable

a

into the cell which

the saturated

solid dissolves

times the number

utilizes

solid and saturated

The heat from an injection

change

we have

cell is loaded with an

is injected

by diluting

as additional

Recently the heat of

The PEPC technique

The sample

v, of pure solvent

is measured

equilibrium.

to measuring

is

solids -- phase equilibrium

(PEPC)

the equilibrium

[14, 151.

of a dry sample which

flow of solvent.

soluble

calorimetry

equilibrium

was flow calorimetry weighing

approach

microcalorimeter

A small volume, perturbs

method

the precise

index of a solution

of solute

of specific

which the refractive

in the solution

chromophores.

is proportional

to the

and can be used even in the

A differential

index of the solution

measurement,

is measured

in

with respect

14 ensures

to that of the solvent, concentrations

and eliminates

the non-differential scattering

with excess

This differential refractive

A new sample 2.

approach

solubilities

as it is in absorbance

solutions

measurements,

in the sample cell.

in. optical

a UV setting

optical

adhesive

The solid/solution

is loaded

into the sample cell and pure water

chamber.

A small magnetic

ensures

rapid equilibration

The cell is placed

Water

or a change

in an aluminum

into the chamber

in

block which

steel tubing

from a circulating

and the temperature

(Esco,

(Norland

stir bar in the sample

upon reloading

with l/4 in. stainless

flats

equilibrium

solvent

lengthwise.

[201.

The flats are attached

mixture

the tubing

the

in urea solutions

from Pyrex with 1/16th

Inc., New Brunswick).

thermostated

is

embedded

water bath is flowed through

is monitored

with a thermocouple.

Slit Solvent cham

_ Sample chamber a Figure 2. side view.

a small

through

with another

lens then focuses

As the slit image passes chambers micrometer

(see Figure field.

index cell shown to scale.

light from a Helium/Neon

lens and passed

is collimated A third

b

Differential refractive b) end-on view.

Monochromatic

in

[191 and by Gill et al. to

NJ) at the ends of each chamber.

temperature.

observed light

was used by Debye to measure

of diketopiperazine

to the cell walls using Products,

since

cell design which we have used is shown in Figure

The cell is made

Oak Ridge,

dependence

solid can be performed

index of polymer

determine

over a wide range of

Additionally,

measurement.

is not a problem

equilibration

linearity

the temperature

laser is expanded

a horizontal

lens and passed

with

slit.

The slit image

through

the sample cell

the image onto a filar micrometer through

a)

field.

both the sample and solvent

2b), a split image is observed

The displacement

of the sample

in the

image from the

15 reference

(solvent)

refractivity

image is a measure

can be calibrated

of the DRI.

by use of a solution

The molar of known

concentration. The molar temperature

enthalpy

dependence

change

can be determined

of the solubility

from the

and consequently

and PEPC data can be used together

to determine

without

The thermodynamic

a solubility

determined

calibration.

from the data using a non-linear

algorithm

[21] in which multiple

the DRI

K, AHO and ACpO parameters

least squares

are

fitting

data sets can be fit

simultaneously.

RESULTS

AND DISCUSSION

The thermodynamic the eight

cyclic

free energy

change

The eight

in Table

an aromatic

The

state of unit mole

were estimated

by F-testing

studied

demonstrate backbone

the effects

is present

of several

in all the

The amino acids G, A, V, and L all have purely

side chains;

of

1.

level.

The peptide

groups.

compounds.

present

compounds

on a standard

intervals

on a 67% confidence

functional

at 298 K for the dissolution

into water are given

is based

The confidence

fraction. based

parameters

dipeptides

apolar

ring is found in F and Y; an -OH group

in S and Y; and P has a blocked

is

amino group on the peptide

backbone. Grouo additivitv The crystal

structures

motif

of a hydrogen

these

functional

independently has been

bonded

groups

of many cyclic dipeptides chain with apolar

might be expected

channels

show a common [22-241; thus

to contribute

to the observed

observed

thermodynamics. Such group additivity for AHo and ACp o for the dissolution of gases [25]

and for the partial

molar heat capacity

of aqueous

solutions

[26].

The groups which make up the cyclic dipeptides are then the peptide, -CONH-, the hydroxyl, -OH, the aromatic ring, @, and apolar hydrogens,

-CH.

This last group has been shown to be proportional

to the thermodynamic liquids apolar

parameters

for the dissolution

[27] and is known to be proportional surface

molecules

[28].

of one aromatic

area or to the number Note that a benzene ring and six apolar

of apolar

to the accessible

of primary

shell water

ring under this scheme hydrogens

(i.e

consists

six -CH.groups).

16

TABLE.l. Thermodynamic parameters for the dissolution of cyclic dipeptides at The errors shown 298 K. The standard state is unit mole fraction. Taken from are one standard deviation estimates based on F-testing. 1131. -..._ AGO AHO Acpo Compound kJ mol-l

kJ mol-l

J K-l mol-l

c(GG)

14.7a

26.2 f 0.2b fc

-15 & 18b

c(AG)

11.4 + 0.2b

17.7 + 0.3b,C

53 + 44b

c(m)

14.1 * 0.5b

13.7 + 0.1283

100 + 33b

c(LG)

17.0 + 0.3

13.3 It:1.6

230 + 70 320 f 2Sb

c(W)

25.2 f 0.S2

8.1 + 0.2b,C

c(LP)

16.6 * 0.1

7.5 + 0.2

316 + 13

c(GF)

22 + 1

19.3 f. 0.2

320 + 150

c(SY)

22.5 + 0.3

26 + 2

230 + 100

a s.

J. Gill, J. Hutson, J. R. Clopton, M. Downing, (1961). b K. P. Murphy, S. J. Gill, J. Chem. Thermodynamics c Value given is 6% lower than previously published calibration of injection syringe.

For a group additivity compound

is assumed

constituent

contribution

quantities.

For example

The crystal

which

structures

amino group motif

used to obtain The best are given

thermodynamics

observed

In fact we observed

in c(LP) yields

that the

This is in contrast

always

of Table

is

1.

that

the c(GG)

found for the

Likewise,

a structure

to the

in cyclic

the blocked

which differs

from

in key set that was

the group parameters. fit values 2.

for the thermodynamic

The errors

from the fit [21].

compound

positive

[22-241.

of its

fit to the experimental

[30] and it is not included

in Table

determined

of each

The average

of c(GG) reveals

[29].

is almost

side group molecules

the common

apolar

structure

in

is equal to twice

data did not fit the simple correlation

aliphatic peptide

c(GG)

AHO for -CH.

least squares

ring is planar

chair conformation

enthalpy

AHO for

times

four

a linear

diketopiperazine

dipeptide

scheme the thermodynamics

of each group to the overall using

21, 903 (1989). due to an error

to be the sum of the thermodynamics

groups.

AHO for the -CONH- plus

determined

J. Phys. Chem. 65, 1432

dissolution

contribution

group

are one standard

As observed

contributions

deviation

from other

as

studies

of

[25, 271, the -CH and @ groups make a

to ACpO.

The polar

groups,

in contrast,

make

17 a negative

contribution,

contribution

of -OH is less certain

data set this group structure this

as has also been observed

of this compound

reason

indicate

however,

is only present reveals

the -OH group values

the tentative

AG"/kJ

since in the present

in c(SY) and the crystal one water per molecule

are given

AHO/kJ

mol-I

in parentheses

to

mol-I

of dissolution

for cyclic

ASO/J K-l mol-I ACP~/J

K-I mol-I

-CONH-

1.6 ? 1.2

11.0 f 1.0

32 + 7

-CH

1.3 f 0.2

-0.8 + 0.2

-7 f 1

27 f 1

@

5.7 + 1.8

5.7 f 1.8

-9 f 11

158 + 13

(-OH)

(0.3 * 1.1)

(3.4 + 1.1)

(10.4 f 7)

(-45 + 8)

There

are several

from the parameters peptide

on the interior

hydrogen

is involved

contributions structures. stabilizing

can be drawn

point

bonds,

the native

is that the stabilizes

will be exactly

by Privalov

are the principal

however

and Gill

compensated

of native

protein

groups provide

force since the AGO value of 1.3 is multiplied

of apolar

hydrogens.

Additionally,

This is possibly

aromatic-aromatic

interactions

and Petsko

Finally,

[321.

to the crystal,

aromatic

groups

a strong by the strongly

due to the favorable

which have been observed

the -OH appears

but as mentioned

by a

it does agree with

[311 that enthalpic

determinant

It is also seen that the apolar

favor the solid state.

the

state of globular

to a widely held view that a hydrogen

of a protein

reached

-56 + 8

which

in hydrogen

bond to water upon unfolding;

the conclusion

number

conclusions

One important

state and, by analogy, This is in contrast

proteins. bond

important

in Table 2.

which

group,

crystalline

For

[23].

interpretation.

TABLE 2 Group contributions to thermodynamics Adapted from [13]. dipeptides. Group

The

[251.

to provide

by Burley

some stability

above, this conclusion

is quite

tentative. Correlations

of ASQ and AHQ with AC,Q

We have previously apolar

compound

noted

dissolution

[l] that protein

have the common

denaturation

feature

and

of a reference

temperature, T*, at which all compounds of a given class have the same value of AS0 designated as AS*. Furthermore, it has also been observed

for protein

denaturation,

that the denaturational

AH0

18 values, normalized to the number of amino acid residues, also attain At the temperature a common value, AH* near this same temperature. effect, attributed to the interaction of apolar with water, is zero and thus the intercept values, AH* and

T* the hydration groups

As*f represent [l, 311.

the thermodynamic

Insight

can be found First,

into the thermodynamic

in the details

298 K, versus

of a convergence

(2) and

= AH* + Acpo

( T - T* )

ASO

= As*+ A$0

where the starred

ln

value

dipeptide

observations

properties.

temperature

can be

T, typically

[l].

This can be

(3).

(2)

terms

(3)

are the thermodynamic

T*.

values

at the

Such a plot will have an intercept

of the

(AH* or AS*) and a slope of T-T* or In T / T*. the occurrence

of the linear correlation

AHO with ACPO can be seen to result observed

groups

( T / T* )

temperature

Second,

of these

ACPn for a given class of compounds

seen from Equations

common

basis

AHO or AS0 for a given temperature,

AHO

reference

of the polar

of group additivity

the observation

noted by plotting

contribution

for the cyclic

dipeptide

the total AS0 value

for the purely

apolar,

of AS0 and

from the group additivity

dissolution.

For any given cyclic

is the sum of the group terms.

non-aromatic

compounds

Thus,

AS0 can be represented

as: AHO = 2 AHO(_~~NH_) +

N-CH

(4)

AH"(-CH)

AS0 = 2 As'c-co~~-) + N-c~ AS"c-c~) where N-c- is the number group values.

(5)

of apolar hydroqens

Similarly,

and subscripts

the total ACPo is given

indicate

as:

(6)

ACpo = 2 AcpOt-co~~-) f N-CH AcpOt-c~) Eliminating resulting AH'=2

N-c~ in (4) or

equations

(5) and

(6) and setting

the

equal to each other yields:

AH"~-CoNH-~-ACpo~-CoNH-~ ! (8)

Equations the bracketed

(7) and terms

(8) have the same form as

can be identified

It can thus be seen that the value

from the ratios

of the apolar

apolar

heat capacity

and AS* are determined

or entropy

of T* results change

change and that the intercept by polar

(3) and

with AH* and T-T* and with AS*

and In (T/T*).

enthalpy

(2) and

group contributions.

to the

quantities,

AH*

19 in summary

it is seen that the thermodynamics

of solid cyclic

dipeptides

follow a group additivity

and that this group additivity correlations

results

which have been observed

the dissolution thermodynamics

of apolar

Acknowledaement grant CHE-891145

in the

to T* provides

of the apolar thermodynamics

We wish to acknowledgement and NIH grant GM44276

relationship

thermodynamic

in protein

model compounds.

with reference

the contribution

of dissolution

denaturation

Description a means

and

of the

of separating

from the total values.

the support

for various

of NSF

aspects

of this

work.

REFERENCES K.P. Murphy, P.L. Privalov and S.J. Gill, Common Features of Protein Unfolding and Dissolution of Hydrophobic Compounds, Science, 247 (1990) 559-561. 2. Y. Nozaki and C. Tanford, The Solubility of Amino Acids and Two Glycine Peptides in Aqueous Ethanol and Dioxane Solutions, J. Biol. Chem., 246 (1971) 2211. 3. J.L. Fauchere and V. Pliska, Hydrophobic parameters n of amino acid side chains from the partitioning of N-acetyl-amino acid amides, Eur. J. Chem.-Chim. ther., 18 (1983) 369-375. 4. S.J. Gill, N.F. Nichols and I. Wadso, Calorimetric Determination of Enthalpies of Solution of Slightly Soluble Liquids I. Application to Benzene in Water, J. Chem. Thermodynamics, 7 (1975) 175. 5. S.J. Gill, N.F. Nichols and I. Wadsb, Calorimetric Determination of Enthalpies of Solution of Slightly Soluble Liquids II. Enthalpy of Solution of Some Hydrocarbons in Water and Their Use in Establishing the Temperature Dependence of Their Solubilities, J. Chem. Thermodynamics, 8 (1976) 445-452. 6. R. Wolfenden, Interaction of the Peptide Bond with Solvent Water: A Vapor Phase Analysis, Biochem., 17 (1978) 201-204. I. R. Wolfenden, L. Andersson, P.M. Cullis and C.C.B. Southgate, Affinities of Amino Acid Side Chains for Solvent Water, Biochem., 20 (1981) 849-850. 8. G. Olofsson, A.A. Oshodj, Qvarnstrtjm and I. Wadso, Calorimetric Measurements on Slightly Soluble Gases in Water. Enthalpies of Solution of Helium, Neon, Argon, Krypton, Xenon, Methane, Ethane, Propane, n-Butane and Oxygen at 288.25, 298.15, and 308.15 K., J. Chem. Thermodynamics, 16 (1984) 1041-1052. 9. S.F. Dee and S.J. Gill, Heats of Solution of Gaseous Hydrocarbons in Water at 15, 25, and 35 "C, J. Sol. Chem., 14 (1985) 827-836. 10. C. Chothia, Principles that Determine the Structure of Proteins, Annu. Rev. Biochem., 53 (1984) 537-572. 11. B. Lee and F.M. Richards, The Interpretation of Protein Structures: Estimation of Static Accessibility, J. Mol. Biol., (1971) 12. B. Gavish, E. Gratton and C.J. Hardy, Adiabatic Compressibility of Globular Proteins, Proc. Natl. Acad. Sci USA, 80 (1983) 750754. 1.

20 13. K.P. Murphy, Probing the Energetics of Protein Stability through Solid Model Compound Studies, Ph. D. thesis, University of Colorado (1990). Siebold, Flow Calorimeter Cell for Measuring 14. S.J. Gill and M.L. Rev. Sci. Instr., 47 (1976) 1399Heats of Solutions of Solids, 1401. vessel for 15. S.-O. Nilsson and I. Wadsb, A flow-microcalorimetric solution of slightly soluble solids, J. Chem. Thermodynamics, 18 (1986) 1125-1133. 16. K.P. Murphy and S.J. Gill, Calorimetric Measurement of the Enthalpy of Dissolution of Diketopiperazine in Water as a Function of Temperature, Thermochim. Acta, 139 (1989) 279-290. and S.J. Gill, A Twin 17. I.R. McKinnon, L. Fall, A. Parody-Morreale Titration Microcalorimeter for the Study of Biochemical Reactions, Anal. Biochem., 139 (1984) 134-139. 18. K.P. Murphy and S.J. Gill, Thermodynamics of Dissolution of Cyclic Dipeptide Solids Containing Hydrophobic Side Groups, J. 21 (1989) 903-913. Chem. Thermodynamics, 19. P. Debye, Light Scattering in Solutions, J. Appl. Phys., 15 (1944) 338-342. 20. S.J. Gill, J. Hutson, J.R. Clopton and M. Downing, Solubility of Diketopiperazine in Aqueous Solutions of Urea, J. Phys. Chem., 65 (1961) 1432-1435. 21. P. Bevington, Data Reduction and Error Analysis for the Physical Sciences, McGraw-Hill, New York, 1969. 22. L.E. Webb and C.-F. Lin, Conformations of Cyclic Dipeptides. Structure of cycle-Glycyl-L-tyrosyl (L-3-(4_Hydroxybenzyl)-2,5piperazinedione), J. Amer. Chem. Sot., 93 (1971) 3818-3819. 23. L.E. Webb and C.-F. Lin, Crystal Structures and Conformations of the Cyclic Dipeptides cycle-(Glycyl-L-tyrosyl) and cycle-(LSeryl-L-tyrosyl) Monohydrate, J. Amer. Chem. Sot., 95 (1973) 6803-6808. Studies 24. E. Benedetti, R.E. Marsh and M. Goodman, Conformational on Peptides. X-Ray Structure Determinations of Six N-Methylated Cyclic Dipeptides Derived from Alanine, Valine, and Phenylalanine, J. Am. Chem. Sot., 98 (1976) 6676-6684. 25. S. Cabani, P. Gianni, V. Mollica and L. Lepori, Group contributions to the thermodynamic properties of non-ionic organic solutes in dilute aqueous solution, J. Solution Chem., 10 (1981) 563-595. 26. N. Nichols, R. Skold, C. Spink and I. Wadso, Additivity relations for the heat capacities of non-electrolytes in aqueous solution, J. Chem. Thermodynamics, 8 (1976) 1081-1093. 27. S.J. Gill and I. Wadso, An Equation of State Describing Hydrophobic Interactions, Proc. Natl. Acad. Sci., 73 (1976) 2955-2958. 28. W.L. Jorgensen, J. Gao and C. Ravimohan, Monte Carlo Simulations of Alkanes in Water: Hydration Numbers and the Hydrophobic Effect, J. Phys. Chem., 89 (1985) 3470-3473. 29. R.B. Corey, The Crystal Structure of Diketopiperazine, J. Am. Chem. Sot., 60 (1938) 1598-1604. 30. I.L. Karle, Crystal Structure and Conformation of the Cyclic Dipeptide cycle-L-Prolyl-L-leucyl, J. Amer. Chem. Sot., 94 (1972) 81-84. 31. P.L. Privalov and S.J. Gill, Stability of Protein Structure and Hydrophobic Interaction, Adv. Protein Chem., 39 (1988) 191-234. 32. S.K. Burley and G.A. Petsko, Aromatic-Aromatic Interaction: A Mechanism of Protein Structure Stabilization, Science, 229 (1985) 23-28.