Vol. 46, No. 3, October 1998 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL Pages 509-517
The role of proline in the p r e v e n t i o n of aggregation d u r i n g p r o t e i n folding in v i t r o T. K. S. Kumal; D. Samuel, G. J a y a r a m a n , T. S r i m a t h i and C. Yu*
Department of Chemistry, National Tsing Hua University, Hsinchu, Taiwan Received June 6, 1998 Received after revision, July 6, 1998
Summary: Proline effectively inhibits protein aggregation during the refolding of bovine carbonic anhydrase. Other osmolytes used such as glycine and ethylene glycol fail to exhibit the 'aggregation-blockade' role shown by proline. Results of viscosity and ANS fluorescence (1-anilino-8-naphthalene sulphonic acid) experiments suggest that proline at high concentrations forms an ordered supramolecular assembly. Based on these results, it is proposed that proline behaves as a protein folding chaperone due to the formation of an ordered, amphipathic supramolecular assembly. To our knowledge, this is the first report wherein proline is proposed as a protein folding aid. Key w o r d s : protein aggregation inhibition, carbonic anhydrase, proline, refolding Protein aggregation is a widespread phenomenon that occurs during protein folding in vitro and in vivo (1-4). A kinetic competition between interchain and intrachain interactions is proposed to occur during the refolding of a protein (5). The interchain interactions tend to predominate if the concentration of the refolding protein is high and as a consequence the protein tends to accumulate (6,7). The formation of protein aggregates is often considered to be a non-specific association of the partially folded polypeptide chains through hydrophobic interactions (8). An understanding of the mechanism of association/aggregation is crucial to solving the problem of the formation of inclusion bodies during the overexpression of recombinant proteins in foreign hosts and also in the prevention/cure of several h u m a n diseases which are proposed to be the result of protein aggregation in the cell (9).
Several experimental strategies have been devised to overcome the
aggregation problem (10-12).
However, to-date, no general panacea exists to
alleviate the in vitro problem of protein aggregation.
* To whom all correspondence should be addressed. Fax - 886 35 711082; e-mail -
[email protected] 103%9712/98/150509-09505.00/0 509
Copyright (2) 1998 by Academic Press Australia. All rights qf reptvduction in any form reserved,
Vol. 46, No. 3, 1998
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
The production and accumulation of organic osmolytes are widespread among halophilic and halotolerant plants and microorganisms
(13).
Some organic
osmolytes can accumulate to cytoplasmic concentrations well above 3 mol/kg in water (14).
The accumulated osmolytes are believed to have a protectant role
against the destructive effects of salt, freezing, heating and possibly drying (15). The best known among the osmolytes are proline, betaine and glycine. Recently, it was proposed that proline at high concentrations (> 3 M) stabilized enzymes and also behaved as a protein solubilizing solute (3). The imino acid proline possess some interesting physical properties. Firstly, its solubility in water is remarkably high - as much as 7 M at ambient temperatures
(16).
Secondly, at high concentrations,
proline behaves as a
hydrotrope (17). These unusual properties of proline prompted us to investigate its role as a protein folding chaperone. In this communication, it is shown that proline at high concentrations (> 3 M) forms an amphipathic supramolecular assembly due to which it successfully thwarts the aggregation associated with the refolding of bovine carbonic anhydrase. Materials and Methods Bovine carbonic anhydrase (from bovine erythrocytes), 1-anilino-8naphthalene sulphonic acid (Mg 2ยง salt) and p-nitrophenyl acetate were obtained from Sigma Chemical Co. (St. Louis). Proline was from Lancaster, England. Glycine, ethylene glycol and guanidinium hydrochloride were from Merck Co., USA. All other chemical used were of high quality analytical grade.
Protein unfolding/folding: Carbonic anhydrase was denatured by overnight incubation in 7 M guanidinium hydrochloride in water at room temperature (pH=8.7). Refolding was achieved by rapid dilution in water containing appropriate osmolytes. T u r b i d i m e t r i c m e a s u r e m e n t s : Turbididimetric measurements were performed on a Hitachi (Model U-3300) spectrometer at 600 nm and room temperature.
Carbonic anhydrase assay: The esterase activity of carbonic anhydrase was measured by the hydrolysis of its substrate, p-nitrophenyl acetate (final concentration of 1 mM) monitored at 400 nm and at room temperature in a double beam spectrometer (Hitachi -Model U3300). Appropriate background corrections were made in all measurements.
510
Vol. 46, No. 3, 1998
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Viscosity m e a s u r e m e n t s : All viscosity measurements were carried out on a manual Ostwald viscometer, at room temperature. The experiment was repeated five times and the viscosity values thus obtained were averaged. A N S - b i n d i n g studies: To appropriate concentrations of proline, 50 ~L of 5 mM 1anilino-8-napthalene sulfonic acid (ANS) was added (final concentration of ANS is 250 pM) and the fluorescence spectra were recorded between 450 nm and 600 nm using an excitation wavelength of 400 nm. All spectra were corrected for the blank. Results and Discussion Bovine carbonic anhydrase (bCAB) is a predominantly ~-sheet protein with a well-defined esterase activity,
bCAB denatured by chaotropic agents, such as
guanidinium hydrochloride or urea, upon refolding is known to irreversibly form high molecular weight, insoluble aggregates. Hence, this protein is an ideal model to study the effects of the osmolytes on the formation of aggregates during protein refolding. Fig. 1 depicts the formation of aggregates upon refolding bCAB at various concentrations in water. Fig. 1 shows t h a t the percentage transmittance at 600 nm of the bCAB protein solutions steadily decreased with increases in protein concentration. This implies t h a t the aggregate formation increased by the increase in the concentration of the refolding protein. The percentage of transmittance at 600 nm (%T600) of the solution containing bCAB refolded in water at 1 mg/ml concentration, is only
13%.
Interestingly, refolding denatured bCAB at various
protein concentrations ranging from 0.1 - 1.0 mg/ml in 5 M proline results in the protein solutions remaining mostly clear.
The %T600 of the solution containing
bCAB refolded at 1 mg/ml is approximately 80%. Thus, the results demonstrate t h a t proline at
5 M effectively subverts the formation of aggregates during the
refolding of denatured bCAB. However the question still remains does proline merely inhibit protein aggregation or do it direct the protein to refold back to its native, biologically active state? In this context, the esterase activity of bCAB refolded in 5 M proline was measured using p-nitrophenyl acetate as the substrate.
Fig. 2 shows t h a t more
t h a n 80% of the expected esterase activity is regained at a protein concentration range of 0.1 - 1.0 mg/ml, wherein putative refolding of the protein (bCAB) was attempted. There was only a marginal decline in the expected enzymatic recovery 511
Vol. 46, No. 3, 1998
BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL
I00
80 60 ==
40 -
20
0 0.2
0.0
0.4
0.6
0.8
1.0
[Carbonic anhydrase] mg]ml Fig. 1 P e r c e n t a g e t r a n s m i t t a n c e a t 600 n m during the refolding of the d e n a t u r e d bCAB in 5 M proline in aqueous solution ([3) and in w a t e r alone (0).
I00
80
.~ >
60
k= < 4O
0
I 0.0
0.2
0.4
0.6
0.8
1.0
[Carbonic anhydrase] mg/ml Fig. 2 E s t e r a s e activity of carbonic a n h y d r a s e upon refolding the d e n a t u r e d bCAB in 5 M proline ([3) and in w a t e r alone (O).
512
Vol. 46, No. 3, 1998
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
with increasing concentrations of the protein used for refolding (Fig. 2). In contrast, bCAB refolded in water exhibited only poor recovery of the enzymatic activity. A maximum of 32% of the expected activity was regained when the protein was refolded at
0.1 mg/ml concentration in water.
At refolding protein test
concentrations beyond 0.4 mg/ml, less t h a n 15% of the expected esterase activity was regained.
It is proposed from the data of Fig.2, t h a t the lower recovery of
enzymatic activity by attempts to refold the protein in water may be primarily due to the formation of aggregates.
Thus, it seems clear t h a t proline at 5 M
concentration not only prevented protein aggregation during refolding but also chaperones the protein to return to its native three dimensional structure. It was of interest to investigate the 'aggregation-inhibition' effect of proline with other known osmolytes such as glycine and ethylene glycol. It was found t h a t these latter osmolytes were ineffective in inhibiting aggregation during the refolding of bCAB (data not shown) as inferred from the %T600 values. In addition, any recovery of the enzymatic activity was feeble when the refolding of the protein (bCAB) was performed with glycine or ethylene glycol.
These results further
confirm t h a t proline behaves as a protein folding aid. Fluorescence experiments were performed inorder to attempt to understand the molecular mechanism whereby proline acts to prevent protein aggregation during the refolding of bCAB. Fig. 3 shows the emission spectrum of native bCAB (1 mg/ml) in water.
The spectrum has an emission maximum at 341 nm.
The
emission spectrum of bCAB in 7 M guanidinium hydrochloride exhibited a wavelength maximum emission shift to 355 nm indicating t h a t the protein is denatured and the hydrophobic tryptophan residues in the protein are exposed to the solvent (18).
The emission intensity of bCAB decreased significantly in 5 M
proline implying t h a t proline quenched the tryptophan fluorescence of the protein. However, the ~max was unchanged and remained centered at 341 nm.
Refolding
denatured bCAB in 5 M proline resulted in a blue shift of the ~max from 355 nm (shown by denatured bCAB) to 341 nm. This implies t h a t the protein was refolded back to the native state with the tryptophan residues moved back into the solvent inaccessible, native micro environment.
513
Vol. 46, No. 3, 1998
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
4000
3000
-
2000
E q~
>
I000
f i
I
I
i
I
320
340
360
380
400
W a v e l e n g t h (nrn)
Fig. 3 Emission spectra of bCAB. (a) bCAB in water; (b)bCAB in 5 M proline; (c) bCAB in 7 M guanidinium hydrochloride (denatured bCAB); (d) denatured bCAB refolded in 5 M proline. The emission spectra were obtained using an excitation wavelength of 293 nm.
A clearer picture regarding the role of proline in the inhibition of aggregation during the refolding may only emerge when the structural properties of proline are more clearly understood.
In this
context,
the
viscosity and
fluorescence
measurements were performed. Fig. 4 shows the change(s) in relative viscosity as a function of increasing concentrations of proline.
The change in viscosity was not
significant up to a proline concentration of 3.5 M.
However, beyond 3.5 M~ the
relative viscosity value was found to exponentially increase with increasing proline concentrations. The anomalous increase in the relative viscosity value of the higher concentrated proline solutions was unexpected for solutes in their monomeric state. The asymptotic rise in viscosity of the solution is reminiscent of those solutes which exhibit the tendency to form high order aggregates (19). As the viscosity profile is an exponential curve, it is possible, that proline at higher concentrations forms a supramolecular assembly. Formation of such multimeric assemblies by proline may account for its protein solubilizing effect(s). It is important to inquire into the nature of the aggregate(s) formed by proline at high concentrations. ANS is an useful probe to monitor the formation of
514
Vol. 46, No. 3, 1998
BIOCHEMISTRY and MOLECULAR BIOLOGYINTERNATIONAL
30 25 20 "i 15 "~ l o 50 0
i
I
1
i
~
l
1
2
3
4
5
6
7
[Osrnolyte] M Fig. 4 Relative viscosity changes of proline. The relative viscosity values represent a mean of five experimental values obtained under identical conditions (25~ 5~
ordered hydrophobic surfaces in the supramolecular assembly of proline (20). Fig. 5 shows the emission maxima of ANS blue shifted by about 25 nm from 524 nm to 419 nm in the proline concentration range of 0 - 4.5 M. Such significant blue shifts in the emission maxima may be attributed to the formation of hydrophobic surfaces in the supramolecular assembly of proline. The emission intensity of ANS is yet one more polarity sensitive property and it presents as a 7-fold increase in the range of proline concentration used (0 - 4.5 M). It is of interest to note t h a t the increase in the emission intensity is not appreciable upto a proline concentration of 2 M (Fig. 5); but beyond this concentration, the increase in emission is steep. This implies
that
the
supramolecular
assembly
steps-up
beyond 2 M
proline
concentration. Thus, the ANS binding experiments demonstrated results t h a t are consistent with proline forming an ordered supramolecular assembly and t h a t the assembly may possess distinct hydrophobic surface(s) conducive for ANS binding. The crystal structure of proline at low resolution is available (21).
The
crystal structure of proline depicts the pyrrolidine rings of proline molecules to be stacked one above the other in the manner of a stacked pile of coins. From the ANS binding experiments presented here, it appears t h a t the stacking of the proline
515
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
Vol. 46, No. 3, 1998
20
530 - 525
c
~J r 00
15
- 520
-
515
-
510
10-
O
- 505
..>
~J ~J
~ 500 I
t
t
t
I
0
1
2
3
4
Osmolyte
495 o
c o n c e n t r a t i o n
Fig. 5ANS binding profiles of proline at various concentrations.
(O) relative
emission intensity; (e) shift in the emission maximum of ANS. All experiments were conducted using an excitation wavelength of 400 nm. The final concentration of ANS used was 250 pM.
molecules is ordered and possess an amphipathic nature with the imino and carboxyl groups facing on one side of the assembly providing the polar surface and the methylene groups of the pyrrolidine rings constituting the hydrophobic surface. Protein aggregation is believed to be due to the intermolecular hydrophobic interactions among the folding molecules.
One of the ways to effectively block
protein aggregation during refolding would be to evolve a mechanism to effectively bind to the hydrophobic groups t h a t tend to expose solvent during protein refolding. In this respect, proline appears to be a very effective molecule in this regard. Due to the possible amphiphilic nature of the supramolecular assembly, proline appears to provide hydrophobic surfaces which interact with the aggregation prone, solvent exposed hydrophobic residues, during the folding of the protein. In such an event, protein aggregation is effectively blocked. Thus, the results presented in this paper, demonstrate for the first time, that proline, in vitro, behaves as a protein folding chaperone. It is possible that, in vivo, under water stress conditions, accumulation of proline may not only have an
516
Vol. 46, No. 3, 1998
BIOCHEMISTRYond MOLECULAR BIOLOGYINTERNATIONAL
osmoregulatory role in the cytoplasm, but also assist some proteins to fold by inhibiting aggregation (denaturation) during the protein folding process.
Acknowledgments This work was supported by the National Science Council of Taiwan (NSC 86-2113-M007-003) and the Dr. C. S. Tsong Memorial Medical Research Foundation (VGHTH 86-0112) grants.
References 1. Goldberg, M. E., Rudolph, R. and Jaencke, R. (1991) Biochemistry 32, 38773886. 2. Kumar, T. K. S., Yang, P. W., Lin, S. H., Wu, C. Y., Lei, B., Lo, S. J., Tu, S. C. and Yu, C. (1996) Biochem. Biophys. Res. Commun. 219, 450-456. 3. Yang, P. W., Kumar, T. K. S., Jayaraman, G. and Yu, C. (1996) Biochem. Mol. Biol. Int. 38, 393-399. 4. Sivaraman, T., Kumar, T. K. S., Huang, C. C. and Yu, C. (1998) Biochem. Mol. Biol. Int. 44, 29-39. 5. Cleland, J. L. and Wang, D. I. C. (1990) Biochemistry, 29, 11072-11078. 6. Sivaraman, T., Kumar, T. K. S., Jayaraman, G. and Yu, C. (1997) J. Protein Chem., 16, 291-297. 7. Sivaraman, T., Kumar, T. K. S., Jayaraman, G., Han, C. C. and Yu, C. (1997) Biochem. J. 321, 457-464. 8. De Young, L. R., Dill, K. A. and Fink, A. L. (1993) Biochemistry 31, 1234512352. 9. Kumar, T. K. S., Lee, C. S. and Yu, C. (1996) Adv. Exp. Med. Biol., 391, 115-129. 10.Samuel, D., Kumar, T. K. S., Jayaraman, G., Yang, P. W. and Yu, C. (1997) Biochem. Mol. Biol. Int. 41, 235-242. 11. Shastry, M. C. R. and Effink, M. R. (1996) Biochemistry, 35, 4094-4101. 12.Rozema, D. and Gellman, S. H. (1996) Biochemistry, 35, 15760-15771. 13.Yanley, P. H., Clark, M. E., Hand, S. C., Bowlus, R. D. and Somero, G. N. (1982) Science, 217, 1214-1222. 14. Galinski, E. A. (1995) Experimentia, 49, 487-494. 15.Lin, T. Y., and Timasheff, H. (1994) Biochemistry, 33, 12695-12701. 16. Schobert, B. and Tschesche, H. (1978) Biochim. Biophys. Acta, 541, 270-277. 17. Srinivas, V. and Balasubramanian, D. (1995) Langmuir, 11, 2830-2833. 18.Lakowicz, J. R. (1983) Principles of fluorescence spectroscopy, Plenum Press, New York. pp. 354-359. 19.Moroi, Y. (1992) Micelles: Theoretical a~d applied aspects, pp. 167-180, Plenum Press, New York, USA. 20. Stryer, L. (1968) Science, 162, 526-528. 21.Kayushina, R. L. and Vainshtein, B. K. (1983) Kristallographia, 834, 28-30.
517
