Water binding by glycogen molecules

258

Biochimica et Biophysica Acta, 543 (1978) 258--263

© Elsevier/North-Holland Biomedical Press

BBA 28625 WATER BINDING BY G L Y C O G E N MOLECULES

T. BRITTAIN and R. GEDDES Department of Biochemistry, University of Auckland, Auckland (New Zealand)

(Received January 26th, 1978)

Summary L o w temperature NMR spectra have been obtained of the water b o u n d to glycogen. These data have allowed the evaluation of the amount of water b o u n d and the energy and entropy associated with this bonding. High molecular weight glycogen (approx. 1 • 109) exhibits water binding properties analogous to those previously found for other glycoproteins. Low molecular weight glycogen (approx. 1 • 10~), however, shows anomalous binding characteristics, with large amounts of associated "non-freezing" water. These findings are disCussed in terms of previously proposed molecular architecture. Introduction The a m o u n t of water b o u n d b y certain proteins and a glycoprotein has recently been measured b y NMR [1]. It was implied that the hydration of the Carbohydrate portion of the glycoprotein was 0.66 g per g carbohydrate, which is relatively close to the 0.59 g per g carbohydrate previously reported for Agarose [2]. Additionally, a recent h y d r o d y n a m i c study of liver glycogen [3] has reported 0.3--1.1 g water associated with each g polysaccharide. Since this latter study measured water b o u n d under specific hydrodynamic conditions (capillary viscometry, zonal centrifugation, etc.), and since this t y p e of b o u n d water is some sort of average, it could well be different under varied hydrodynamic conditions [4]. Therefore it seemed of interest to investigate the hydration of glycogen b y NMR. Further, it proved impossible to estimate the a m o u n t of water b o u n d to low molecular weight glycogen b y hydrodynamic means, and as this is probably the most c o m m o n molecular size in the cytosol [5] it was also of great interest to ascertain this quantity. Experimental Livers were quickly removed from rabbits (New Zealand White), which had been given an overdose of Nembutal ( A b b o t t Laboratories Ltd, Naenae, New

259 Zealand) and plunged into liquid nitrogen. Glycogen was then isolated from the tissue by cold-water extraction [5,6] and fractionated on a sucrose density gradient [3,5]. Fractions were combined to provide both " l o w " and " h i g h " molecular weight samples. Approximate molecular weights were calculated as detailed in reference [3]. Glycogen concentrations were determined either by an iodine-iodide reaction (in the presence of sucrose) [6] or by a standard anthrone m e t h o d (after extensive dialysis of the glycogen samples). All samples were made 10 mM in KC1 prior to the recording of NMR spectra. The samples were placed into 10 mm commercial NMR tubes and a sample of deutero-acetone in a coaxial 3-mm tube acted as an external reference lock. Spectra were recorded using a JEOL FX60 60 MHz spectrometer. In order to obtain Fourier transform spectra the samples were pulsed 30 times using a 10 kHz spectral width. The samples were maintained to within +0.5°C of the required temperature by use of the JEOL NM5471 temperature programmer. At each stage of cooling, 20 min were allowed for equilibration of the sample after the temperature probe had attained the required reading. The line widths reported here represent the width (in Hz) at half height of the spectral curves. Areas were calculated as the product of the line width and height and were converted to hydration values by comparison with a standard water sample, in this case bovine serum albumin. The value of 0.37 g water/g protein a t - - 3 5 ° C was used for bovine serum albumin [ 1 ].

Results and Discussion Fig. 1 shows clearly t h a t even though the bulk water in the samples used has frozen at --15°C, a measurable NMR water signal is still apparent in glycogen. This is consistent with previous data on a variety of macromolecules which have been shown to possess a certain a m o u n t of " b o u n d " water which does n o t freeze at temperatures down to --35°C [1,7--9]. However, differences are apparent in the non-freezing water in the samples. Fractions C and A of glycogen (Mr of approx. 1 . 1 0 7 and 1 . 1 0 9 , respectively [3]). The smaller line width obtained for fraction A shows a greater mobility of the associated water compared to fraction C. The line widths of the various low molecular weight fractions of glycogen show a linear dependence on temperature in Fig. 2. From these Arrhenius plots it is apparent that whole glycogen and fraction C have activation energies similar to bovine serum albumin. Fig. 3 shows that the a m o u n t of water associated with these fractions is very similar to t h a t of bovine serum albumin [1], being approx. 0.35 g water/g glycogen at --35 ° C. Molecular weight fractions B (Mr approx. 5 • 107 [3]) and A, however, show NMR signals different from bovine serum albumin {Fig. 1). Fig. 2 shows t h a t the line widths for these fractions are much smaller than those of bovine serum albumin at the same temperature, and are almost temperature independent. Analysis of the areas of the NMR curves for fractions B and A show a level of h y d r a t i o n of approximately 0.8 g water/g glycogen and 9 g water/g glycogen, respectively, at --35 ° C. These data suggest t h a t in the sample of whole glycogen the contributions arising from fraction A in particular must be very small. This

260

~)

//1'/'%'1J1

~35o ~L<

(b)

p2~50 tt,:

j

•

~_

_

F i g . 1. (a) T h e N M R s p e c t r u m o f f r a c t i o n C o f g l y c o g e n a t 2 . 6 m g / m l a n d - - 1 5 . 5 ° C , ( b ) T h e N M R s p e c o t r u m o f f r a c t i o n A o f g l y c o g e n o b t a i n e d a t 1.1 m g / m l a n d - - 2 9 . 5 C. T h e ba~ i n b o t h c a s e s r e p r e s e n t s a s p e c t r a l w i d t h o f 2 5 0 Hz. P e a k A is f r o m t h e d e u t e r o a c e t o n e l o c k a n d P e a k B is w a t e r c o n t a m i n a t i o n o f the deuteroacetone

used.

is n o t unreasonable when consideration is taken of the overall molecular weight distribution of glycogen (see Fig. 2 and ref 3). The whole glycogen must represent the average of the properties of the whole range of molecular weight fractions present; contributions from fractions B and A would grossly lower the temperature dependence of the line widths and increase the level of hydration, if present at any appreciable level. These interesting findings for the very high molecular weight fractions were more closely analysed. Eyring plots of the line width data were constructed in order to obtain activation enthalpies (/-/¢) and entropies (S ~) (Table I). Once again, these data show that whole glycogen and fraction C possess properties very similar to those found for bovine serum albumin, which shows an activation enthalpy of 5.5 kcal/mol and an activation energy of cal/degree per mol, in close agreement with previous determinations [8,10]. The activation enthalpy found for fraction A is very small, whilst the activation e n t r o p y is large and negative. The values q u o t e d for fraction A should be taken as an indication

261 3.2

2-8

¢--

2-4

¢D rCD

2.0 o

I

I

3"8

4.0

I 4.2

PT (" I'~3) F i g . 2. A r r h e n i u s p l o t s o f t h e v a r i a t i o n o f l i n e w i d t h s o f t h e v a r i o u s s a m p l e s as a f u n c t i o n o f t e m p e r a t u r e . W h o l e g l y c o g e n : c o o l i n g , o ; h e a t i n g , e . b o v i n e s e r u m a l b u m i n , ~ ; G l y c o g e n f r a c t i o n C, O; f r a c t i o n A , o; a n d f r a c t i o n B , ~.

only of the size of these quantities for the almost complete temperature independence of the NMR signals necessarily implies a large error in the evaluated parameters. Even so, these data lead us to the conclusion that the large amounts of water associated with fraction B and especially fraction A are bound much more weakly than is the case for fraction C, probably due to lower hydrogen bonding, and also that the water is present in a less structured environment. It will be noted that fraction A (Fig. 1) has an associated small side band in

1.5

1-0

B

()o

c<39 O') O 03

o"

0.5

03

I

I

I

l

0

-10

- 20

- 30

Temp(°c) F i g . 3. T h e a m o u n t o f b o u n d w a t e r f o r w h o l e g l y c o g e n o n c o o l i n g ( o ) a n d h e a t i n g ( o ) is s h o w n t o g e t h e r w i t h f r a c t i o n C (A). T h e a m o u n t o f w a t e r is n o r m a l i s e d t o a s a m p l e c o n c e n t r a t i o n of 20 mg/ml of bovine serum albumin with a water content of 0.37 g/g glycogen at --35°C.

262

TABLE I ACTIVATION PARAMETER

OF WATER BINDING TO GLYCOGEN

E a ¢ is o b t a i n e d f r o m F i g . 2 a n d A H ~e a n d AS p f r o m E y r i n g p l o t s . Molecule

Ea ¢ (kcal/mol)

AH ¢ (kcal/mol)

Whole glycogen Fraction C Fraction B Fraction A

12.6 11.0 ~ 1.0 ~ 1.0

12.7 11.9 -~ 1.6

AS ¢ (cal/degree per mol) 9.7 6.2 -~--40

its NMR spectrum and it is tempting to suggest the presence of two types of water. However, for the above analysis only the larger peak has been treated. The variety of this data may be viewed as arising from the differences in structure and molecular weights of the various fractions of glycogen. It has been shown that whole glycogen, as prepared, consists mainly of spheres or small groups of spheres, each sphere with a molecular weight of approx. 1 • 107 [3], and the higher molecular weight fractions consist of aggregations of these spheres. Fraction A consists of molecules of Mr approx. 1 • 109 composed of a cluster of approx. 150 protein-linked spheres, each of a molecular weight of approx. 1 • 107 [3]. Thus in the low molecular weight fractions, which are easily permeable to water [ 3,11], the residual NMR signal at low temperatures probably represents a surface layer of b o u n d water. In the case of such examples as fraction A the large amount of very mobile water may represent either or both of the two possible cases. The aggregation of the large number of spheres may well lead to a very large volume of interstitial water being trapped within the glycogen. Alternatively the less permeable: spheres, which constitute the higher molecular weight spheres, may have trapped within them a large a m o u n t of non-freezing water. These results complement the reported [3] levels of water associated hydrodynamically with glycogen. It is n o t e w o r t h y that the level reported here is significantly higher than those found by hydrodynamic investigations. This is n o t unreasonable, since other authors have pointed o u t that the water apparently associated with macromolecules under dynamic conditions may well vary with the m e t h o d of investigation used [4,12]. Acknowledgments We would like to thank Ms. C. Thoreau for expert technical assistance in the preparation of fractionated glycogen, and Mr D. Calvert and Associate-Professor B.R. Davis for help in attaining the NMR spectra. R. Geddes also gratefully acknowledges the support of the Medical Research Council of New Zealand. References 1 H a s c h e m e y e r , A . E . V . , G u s c h l b a u e r , W. a n d de V r i e s , A . L . ( 1 9 7 7 ) N a t u r e 2 6 9 , 8 7 - - 8 8 2 D e r b y s h i r e , W. a n d D u f f , I . D . ( 1 9 7 6 ) F a x a d a y Disc. 57, 2 4 3 - - 2 5 4 3 G e d d e s , R. H a r v e y , J . D . a n d Wills, P . R . ( 1 9 7 7 ) B i o c h e m . J. 1 6 3 , 2 0 1 - - 2 0 9

263 40gston, A.G. (1953) Trans Faraday Soc. 49, 1481--1489 5 G e d d e s , R. a n d S t r a t t o n , G.C. ( 1 9 7 7 ) C a r b o h y d r a t e Res. 57, 2 9 1 - - 2 9 9 6 G e d d e s , R. a n d R a p s o n , K . B . ( 1 9 7 3 ) F E B S L e t t . 3 1 , 3 2 4 - - 3 2 6 7 De Vries, A . L . ( 1 9 7 1 ) S c i e n c e 1 7 2 , 1 1 5 2 - - 1 1 5 5 8 K u n t z , I.D. a n d B r a s s f i e l d , T.S. ( 1 9 7 1 ) A r c h . B i o c h e m . B i o p h y s . 1 4 2 , 6 6 0 - - 6 6 4 9 C o o k e , R. a n d K u n t z , I.D. ( 1 9 7 4 ) A m . R e v . B i o p h y s . B i o e n g . 3, 9 5 - - 1 2 6 1 0 K u n t z , I.D. a n d K a u z m a n n ( 1 9 7 4 ) A d v . P r o t . C h e m . 28, 2 3 9 - - 3 4 5 11 S t e t t e n , M . R . a n d K a t z e n , H . M . ( 1 9 6 1 ) J. A m . C h e m . S o c . 8 3 , 2 9 1 2 - - 2 9 1 8 1 2 Mehl, J . W . , O n c l e y , J . L . a n d S i m b a , R . ( 1 9 4 0 ) S c i e n c e 9 2 , 1 3 2 - - 1 3 3