Advances in Biopolymers - American Chemical Society

---

Chapter 1

New Views of Protein Structure: Implications for Potential New Protein Structure-Function Relationships

Downloaded by UNIV OF PITTSBURGH on June 26, 2013 | http://pubs.acs.org Publication Date: August 28, 2006 | doi: 10.1021/bk-2006-0935.ch001

Protein Structure and Functionality H. M. Farrell, Jr.

1,*

1

P. X. Qi , and V. N. Uversky

2-4

1

Eastern Regional Research Center, Agricultural Research Service,

U.S. Department of Agriculture, 600 Fast Mermaid Lane, Wyndmoor, PA 19038 Department of Biochemistry and Molecular Biology, School of Medicine, Indiana University, Indianapolis,IN46202 Institute for Biological Instrumentation, Russian Academy of Sciences, 142292 Pushchino, Moscow Region, Russia Molecular Kinetics, 6201 La Pas Trail, Suite 160, Indianapolis,IN46268

2

3

4

Recent advances in the field of protein chemistry have significantly enhanced our understanding of the possible intermediates which may occur during protein folding and unfolding. In particular, studies on α-lactalbumin have led to the theory that the molten globule state may be one possible intermediate in the folding of many proteins. The molten globule state is characterized by a compact structure, a high degree of hydration and side chain flexibility, a significant amount of native secondary structure but little tertiary structure, and the ability to react with chaperones. Other partially folded conformations (e.g., pre-molten globule) have also been found. Many proteins known as natively unfolded, intrinsically unstructured, or intrinsically disordered were shown to be highly flexible under physiological conditions. By taking advantage of this "new view" of protein folding, and applying these concepts to engineered macromolecules and food proteins, it may be possible to generate new and useful forms of proteins for the food ingredient, pharmaceutical and nanotechnological markets.

© 2006 American Chemical Society

In Advances in Biopolymers; Fishman, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2006.

1

2

Downloaded by UNIV OF PITTSBURGH on June 26, 2013 | http://pubs.acs.org Publication Date: August 28, 2006 | doi: 10.1021/bk-2006-0935.ch001

Introduction The recent announcement that the human genome has been solved, represents not an end in itself, but a beginning. It is now the work of the protein chemists and structural biologists to translate this linear sequence information into the functional molecular shapes necessary to make significant breakthroughs in biological applications. This task is currently underway, and by the union experimental and theoretical protein studies, we may soon be able to fully understand how a protein structure relates to its biological function. Basic studies on the protein folding problem, are already producing a wealth of fundamental information (7-7/ We must now render this information into new food and technological applications. It has long been known that changing protein structure can alter food functionality. We have often previously thought of these changes as all (completely unfolded) or nothing (completely native) events. The latest research, however, demonstrates a multiplicity of stages in protein folding (unfolding) pathways. The energy profiles suggest that these "molten globule" (1) and "pre-molten globule" states (7) may be stable under certain conditions. If this is the case, then a single whey protein such as alactalbumin may have a myriad of stable intermediate stages which could be "trapped" and thus provide the potential for new functional properties. Similar applications to the major milk proteins (the caseins) are urgently needed. As food protein chemists we must also recall that while native structure arises from sequence, processing treatment has the potential to transform native structure into non-native states. Furthermore, some proteins were shown to avoid complete folding, existing instead as dynamic ensembles of interconverting partially folded structures. This latter fact represents both a challenge and an opportunity, as denatured, or partially denatured proteins, may serve as either a problem or a potentially valuable food ingredient. Examples of these opportunities for the development of novel protein ingredients are given for alactalbumin (α-LA) as "food for thought".

Historical Background of Protein Folding Historically, the central dogma of structural biology is the Anfinsen hypothesis: the linear primary sequence of amino acids of a protein codes for rather specific secondary structural elements, which in turn lead to protein folding and tertiary structure, and ultimately to quaternary structure for complex higher order systems (8). There has been considerable debate in recent years, folding and tertiary structure, and ultimately to quaternary structure for complex higher order systems (8). There has been considerable debate in recent years, not

In Advances in Biopolymers; Fishman, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2006.

3

Downloaded by UNIV OF PITTSBURGH on June 26, 2013 | http://pubs.acs.org Publication Date: August 28, 2006 | doi: 10.1021/bk-2006-0935.ch001

so much on the veracity of the Anfinsen hypothesis, but on the details of the folding mechanisms which bring about protein structure. This debate arises, in part, because until very recently our best information on protein folding came from unfolding studies on small and relatively simple proteins (9). For example, the early studies on lysozyme, as summarized by Arai and Kuwajima (2), indicated a concerted mechanism by which loss of enzyme activity and loss of structure followed nearly similar curves. Figure 1 represents the typical sigmoidal curve found for lysozyme unfolding induced by guanidine HC1 showing loss of structure; this was coincident with loss of enzymatic function (2). These and many other studies could be summarized in eq 1.

Native

τ

1

Unfolded

1

1

1

1

0)

1

Γ

GdnHCI Concentration (M) Figure L Guanidine HCl induced unfolding of lysozyme showing coincident loss of enzyme activity and of structure by as measured by f , the apparent fraction denatured. Open symbols represent NaCl only and closed represent NaCl + CaCl . Changers in CD were monitored at 289 nm (Π, A) 255 nm (Φ, O) and 222 nm (Τ, V / Data from Arai and Kuwajima (2) reproduced courtesy of Academic Press, San Diego, CA. Copyright 2000. app

2

By inference, protein folding was thought to be the reverse reaction. Indeed the argument was made that kinetically the two processes might be microscopically reversible, and that the native state could represent a singular folded state with minimum potential energy.

In Advances in Biopolymers; Fishman, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2006.

4 α-Lactalbumin and the Molten Globule State The "new view" o f protein folding began to emerge, in part, from experimental observations on the milk Ca -binding protein, α-lactalbumin (aL A ) (70, 11). Physical studies on this protein appeared to indicate that some measurable structural loss might occur at different stages in the unfolding process. Conventional Fourier transform infrared (FTIR) and far-UV circular dichroism (CD) spectroscopies are used to follow losses of secondary structural elements, but near-UV C D can also provide important information on protein tertiary structure. In a typical globular protein, steric restrictions as well as hydrogen bonding can limit the mobility of aromatic side chains such as Trp, Tyr and Phe, thus causing a near-UV C D spectra. Unrestricted rotation of aromatic groups does not in general give rise to a C D spectrum. Figure 2 shows that for guanidine denaturation of α-LA, loss of aromatic dichroism preceded loss of secondary structure. In a similar fashion the typical near-UV C D spectra of bovine α-LA is diminished at elevated temperatures (2) or low p H (11). The near-UV C D spectrum is significantly deminished at 50°C, whereas measures of changes in secondary structure reveal higher transition temperatures.

Downloaded by UNIV OF PITTSBURGH on June 26, 2013 | http://pubs.acs.org Publication Date: August 28, 2006 | doi: 10.1021/bk-2006-0935.ch001

2+

ι—ι

1

1

1

1

1

1

1

GdnHCI Concentration (M)

Figure 2. Guanidine HCl induced unfolding of α-LA; the f (fraction denatured) shows non-coincident loss of protein structure by CD. Loss of aromatic dichroism was measured at 270 nm (o); loss of α-helix was flollowed at 222 nm (Δ). Figure as credited in Figure 1. app

Differential scanning calorimetry (DSC) studies of α-LA by Dolgikh et al. (11) showed that the denaturation temperature is about 60°C (Figure 3). When C a is removed from α-LA there is a loss of tertiary structural stability and the denaturation temperature lowers to about 20°C (Figure 3). Curiously, on acid denaturation at p H 2.0, much secondary structure is retained but a complete loss of the characteristic thermal transition of α-LA occurs (dotted line, Figure 3). The lack of coincidence of loss of aromatic C D and secondary structure began the search for a broader meaning of eq 1. 2+

In Advances in Biopolymers; Fishman, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2006.

20

40

60

60

Downloaded by UNIV OF PITTSBURGH on June 26, 2013 | http://pubs.acs.org Publication Date: August 28, 2006 | doi: 10.1021/bk-2006-0935.ch001

Temperoture l*C)

Figure 3. Dependence ofpartial heat capacity (Cp) on temperature for aLA as holoprotein (—) apoprotein (dark dashed line) and acid denatured protein (flat, light dashed line). Data from Dolgikh et al (12) reproduced courtesy of Springer- Verlag Heidelberg, Germany. Friere and his colleagues (13) introduced the concept of 3D D S C by which the temperature of denaturation was studied as a function of guanidine concentration. Surprisingly, 1 M guanidine alleviated almost all of the typical enthalpic thermal transition found for α-LA in the absence of guanidine, yielding a D S C pattern similar to that found at p H 2.0 (Figure 3). From the data of Figure 2, it can be seen that at 1 M guanidine, there has been some change in the near-UV C D spectrum, but little change in secondary structure. Studies on many other proteins now reveal that multi-step mechanisms of protein unfolding may be the rule, rather than the exception (1, 14). The condition in which a partially denatured protein may exhibit a high degree of segmental motion (i.e., loss of aromatic dichroism) while retaining a significant amount of secondary structure has been termed the molten globule state, but this too turns out to be an over simplification as we shall see.

Multi-State Unfolding of Globular Proteins and Pre-Molten Globule State Subsequent studies revealed that the molten globule is a member of the realm of partially folded conformations as globular proteins may exist in at least four different conformations: native (ordered), molten globule, pre-molten globule, and unfolded (1, 7, 15). For example, using a combination of several spectroscopic techniques, such as intrinsic and A N S fluorescence, near- and farUV C D spectra, with size exclusion chromatography it has been shown that the equilibrium GdmCl-induced unfolding of β-lactamase (16), carbonic anhydrase (17), creatine kinase (18), and NAD -dependent DNA-ligase from Thermus scotoductus (19) is a sequential process, which involves the formation of at least +

In Advances in Biopolymers; Fishman, M., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2006.

6 two partially folded intermediates, molten globule (MG) and pre-molten globule (PMG) states. Figure 4 summarizes the results of these studies for DNA-ligase and visualizes the four-state N