Voet D., Voet Ju.G. Biochemistry

Подождите немного. Документ загружается.

Cuff, J.A. and Barton, G.J., Evaluation and improvement of multi-

ple sequence methods for protein secondary structure predic-

tion, Proteins 34, 508–519 (1999). [The principles behind

Jpred3.]

Das, R. and Baker, D., Macromolecular modeling with Rosetta,

Annu. Rev. Biochem. 77, 363–382 (2008).

DeGrado W.F., Summa, S.M., Pavone,V., Nastri, F., and Lombardi,

A., De novo design and structural characterization of proteins

and metalloproteins, Annu. Rev. Biochem. 68, 779–819 (1999).

Kuhlman, B., Dantas, G., Ireton, G.C., Varani, G., Stoddard, B.L.,

and Baker, D., Design of a novel protein fold with atomic-

level accuracy, Science 302, 1364–1368 (2003). [The design

of Top7.]

Lesk, A.M., Introduction to Bioinformatics (3rd ed.), pp. 333–358,

Oxford University Press (2008).

Mirny, L. and Shakhnovitch, E., Protein folding theory: From lat-

tice to all-atom models, Annu. Rev. Biophys. Biomol. Struct. 30,

361–396 (2001).

Moult, J., Fidelis, K., Kryshtofovych, A., Rost, B., and Tramon-

tano, A., Critical assessment of methods of protein structure

prediction—Round VIII, Proteins 77 (Issue S9), 1–4 (2009).

[The summary article of the issue of Proteins: Structure,Func-

tion, and Bioinformatics that reports the results of CASP8.]

Rose, G.D., Prediction of chain turns in globular proteins on a hy-

drophobic basis, Nature 272, 586–590 (1978).

Tramontano,A., Protein Structure Prediction. Concepts and Appli-

cations, Wiley-VCH (2006).

Zaki, M.J. and Bystroff, C. (Eds.), Protein Structure Prediction

(2nd ed.), Humana Press (2008).

Protein Dynamics

Henzler-Wildman, K. and Kern, D., Dynamic personalities of pro-

teins, Nature 450, 964–972 (2007).

Karplus, M. and McCammon,A., Molecular dynamics simulations

of biomolecules, Nature Struct. Biol. 9, 646–651 (2002).

Palmer,A.G., III, Probing molecular motion by NMR, Curr. Opin.

Struct. Biol. 7, 732–737 (1997).

Protein dynamics, Science 324, 197–215 (2009). [A special section

containing four articles.]

Rasmussen, B.F., Stock, A.M., Ringe, D., and Petsko, G.A., Crys-

talline ribonuclease A loses function below the dynamical

transition at 220 K, Nature 357, 423–424 (1992).

Ringe, D. and Petsko, G.A., Mapping protein dynamics by X-ray

diffraction, Prog. Biophys. Mol. Biol. 45, 197–235 (1985).

Scheraga, H.A., Khalili, M., and Liwo,A., Protein-folding dynam-

ics: Overview of molecular simulation techniques, Annu. Rev.

Phys. Chem. 58, 57–83 (2007).

Conformational Diseases

Booth, D.R., et al., Instability, unfolding and aggregation of hu-

man lysozyme variants underlying amyloid fibrillogenesis, Na-

ture 385, 787–73 (1997); and Funahashi, J., Takano, K., Ogasa-

hara, K., Yamagata, Y., and Yutani, K., The structure, stability,

and folding process of amyloidogenic mutant lysozyme,

J. Biochem. 120, 1216–1223 (1996).

Büeler, H., Aguzzi, A., Sailer, A., Greiner, R.A., Autenreid, P.,

Aguet, M., and Weissmann, C.,Mice devoid of PrP are resistant

to scrapie, Cell 73, 1339–1347 (1993); and Büeler, H., Fischer,

M., Lang, Y., Bluethmann, H., Lipp, H.-P., DeArmond, S.J.,

Prusiner, S.B., Aguet, M., and Weissmann, C., Normal develop-

ment and behaviour of mice lacking the neuronal cell-surface

PrP protein, Nature 356, 577–582 (1992).

Buxbaum, J.N. and Tagoe, C.E., The genetics of amyloidoses,

Annu. Rev. Med. 51, 543–569 (2000).

Caughey, B., Baron, G.S., Chesebro, B., and Jeffrey, M., Getting a

grip on prions: oligomers, amyloids, and pathological mem-

brane interactions, Annu. Rev. Biochem. 78, 177–204 (2009).

Chien, P.,Weissman, J.S., and DePace, J.H., Emerging principles of

conformation-based inheritance, Annu. Rev. Biochem. 73,

617–656 (2004).

Chiti, F. and Dobson,C.M., Protein misfolding, functional amyloid,

and human disease, Annu. Rev. Biochem. 75, 333–366 (2006).

Collinge, J. and Clarke,A.R.,A general model of prion strains and

their pathogenicity, Science 318, 930–936 (2007).

Deleault, N.R., Harris, B.T., Rees, J.R., and Supattapone, S., For-

mation of native prions from minimal components in vitro,

Proc. Natl. Acad. Sci. 104, 9741–9746 (2007).

Geula, C., Wu, C.-K., Saroff, D., Lorenzo, A., Yuan, M., and

Yankner, B.A., Aging renders the brain vulnerable to amyloid

␤-protein neurotoxicity, Nature Med. 4, 827–831 (1998).

Goedert, M. and Spillantini, M.G., A century of Alzheimer’s dis-

ease, Science 314, 777–781 (2006).

Gregersen, N., Bross, P., Vang, S., and Christiensen, J.H., Protein

misfolding and human disease, Annu. Rev. Genomics Hum.

Genet. 7, 103–124 (2006).

Hardy, J. and Selkoe, D.J., The amyloid hypothesis of Alzheimer’s

disease: Progress and problems on the road to therapeutics,

Science 297, 353–356 (2002).

Jackson, G.S. and Clarke, A.R., Mammalian prion proteins, Curr.

Opin. Struct. Biol. 10, 69–74 (2000).

Kajava,A., Squire,J.M., and Parry, D.A.D. (Eds.), Fibrous Proteins:

Amyloids, Prions and Beta Proteins, Adv. Prot. Chem. 73

(2006). [The last four chapters are on various aspects of amy-

loids and prions.]

Moore, R.A., Taubner, L.M., and Priola, S.A., Prion misfolding

and disease, Curr. Opin. Struct. Biol. 19, 14–22 (2009).

Pan, K.M., Baldwin, M., Nguyen, J., Gasset, M., Serban, A., Groth,

D., Mehlhorn, I., Huang, Z., Fletterick, R.J., Cohen, F.E., and

Prusiner, S.B., Conversion of ␣-helices into ␤-sheet features in

the formation of the scrapie prion proteins, Proc. Natl. Acad.

Sci. 90, 10962–10966 (1993).

Prusiner, S.B. (Ed.), Prion Biology and Diseases (2nd ed.), Cold

Spring Harbor Laboratory Press (2004); and Prion diseases, in

Valle,D. (Ed.), The Online Metabolic & Molecular Bases of In-

herited Disease, http://www.ommbid.com/, Chap. 224.

Rochet, J.C. and Lansbury, P.T., Jr.,Amyloid fibrillogenesis:Themes

and variations, Curr. Opin. Struct. Biol. 10, 60–68 (2000).

Sawaya, M.R., et al., Atomic structures of amyloid cross-␤ spines

reveal varied steric zippers, Nature 447, 453–457 (2007); and

Nelson, R., Sawaya, M.R., Balbirnie, M., Madsen,A.Ø., Riekel,

C.,Grothe,R.,and Eisenberg,D., Structure of the cross-␤ spine

of amyloid-like fibrils, Nature 435, 773–778 (2005).

Selkoe, D.J., Cell biology of protein misfolding: the examples of

Alzheimer’s and Parkinson’s diseases, Nature Cell Biol. 6,

1054–1061 (2004).

Soto, C., Estrada, L., and Castilla, J., Amyloids, prions and the in-

herent infectious nature of misfolded proteins, Trends

Biochem. Sci. 31, 150–155 (2006).

Sparrer, H.E., Santoso,A., Szoka, F.C., Jr., and Weissman, J.S., Ev-

idence for the prion hypothesis: Induction of the yeast [PSI

⫹

]

factor by in vitro-converted Sup35 protein, Science 289,

595–599 (2000).

Tuite, M.F.,Yeast prions and their prion-forming domain, Cell 100,

289–292 (2000).

Weissmann, C., The state of the prion, Nature Rev. Microbiol. 2,

861–862 (2004).

Wiltzius, J.J.W., Sievers, S.A., Sawaya, M.R., Cascio, D., Popov, D.,

Riekel, C., and Eisenberg, D., Atomic structure of the cross-␤

References 321

JWCL281_c09_278-322.qxd 6/1/10 7:35 AM Page 321

spine of islet amyloid polypeptide (amylin), Prot. Sci. 17,

1467–1474 (2008).

Zahn, R., Liu, A., Lührs, T., Riek, R., von Schroetter, C., Garcia,

F.L., Billeter, M., Calzolai, L., Wider, G., and Wüthrich, K.,

NMR solution structure of the human prion protein, Proc.

Natl. Acad. Sci. 97, 145–150 (2000); and Liu, H., Farr-Jones, S.,

Ulyanov, N.B., Llinas, M., Marqusee, S., Groth, D., Cohen, F.E.,

Prusiner, S.B., and James, T.L., Solution structure of Syrian

hamster prion protein rPrP(90–231), Biochemistry 38,

5362–5377 (1999).

Structural Evolution

Bajaj, M. and Blundell, T., Evolution and the tertiary structure of

proteins, Annu. Rev. Biophys. Bioeng. 13, 453–492 (1983).

Dickerson, R.E., Timkovitch, R., and Almassy, R.J., The cy-

tochrome fold and the evolution of bacterial energy metabo-

lism, J. Mol. Biol. 100, 473–491 (1976).

Eventhoff, W. and Rossmann, M., The structures of dehydro-

genases, Trends Biochem. Sci. 1, 227–230 (1976).

Lesk, A.M., NAD-binding domains of dehydrogenases, Curr.

Opin. Struct. Biol. 5, 775–783 (1995).

Moore,A.D., Björklund, Å.K., Ekman, D., Bornberg-Baur,E., and

Elofsson, A., Arrangements in the modular evolution of pro-

teins, Trends Biochem. Sci. 33, 444–451 (2008).

Scott, R.A. and Mauk, A.G. (Eds.), Cytochrome c. A Multidisci-

plinary Approach, University Science Books (1996).

322 Chapter 9. Protein Folding, Dynamics, and Structural Evolution

1. How long will it take the polypeptide backbone of a

6-residue folding nucleus to explore all its possible conformations?

Repeat the calculation for 10-, 15-, and 20-residue folding nuclei.

Why, in the classic view of protein folding, are folding nuclei

thought to be no larger than 15 residues?

*2. Consider a protein with 10 Cys residues. On air oxidation,

what fraction of the denatured and reduced protein will randomly

reform the native set of disulfide bonds if: (a) The native protein

has five disulfide bonds? (b) The native protein has three disulfide

bonds?

3. Why are ␤ sheets more commonly found in the hydropho-

bic interiors of proteins than on their surfaces?

4. Under physiological conditions, polylysine assumes a ran-

dom coil conformation. Under what conditions might it form an ␣

helix?

5. Explain how landscape theory is consistent with the obser-

vation that many small proteins appear to fold to their native con-

formations without detectable intermediates, that is, via two-state

mechanisms.

6. Explain why Pro residues can occupy the N-terminal turn

of an ␣ helix.

7. Explain why ␤ sheets are less likely to form than ␣ helices

during the earliest stages of protein folding.

8. Molten globules are thought to be predominantly stabilized

by hydrophobic forces. Why aren’t hydrogen bonding forces im-

plicated in doing so?

*9. The GroEL/ES cycle diagrammed in Fig. 9-25 only circu-

lates in the clockwise direction. Explain the basis for this irre-

versibility in terms of the sequence of structural and binding

changes in the GroEL/ES system.

*10. Predict the secondary structure of the C peptide of proin-

sulin (Fig. 9-4) using the methods of Chou–Fasman and Rose.

11. As Mother Nature’s chief engineer, now certified as a

master helix builder,you are asked to repeat Problem 8-8 with the

stipulation that the ␣ helix really be helical. Use Table 9-1.

*12. Predict the secondary structure of the N-terminal do-

main of yeast protein disulfide isomerase using Jpred3

(http://www.compbio.dundee.ac.uk/www-jpred/). How does this

prediction compare with the observed structure of this domain

(PDBid 2B5E; the a domain in Fig. 9-17)? [To enter the sequence

of this domain into Jpred3, first point your browser at the PDB

(http://www.rcsb.org/pdb), enter the PDBid 2B5E, click on the

“Sequence Details” tab at the top of the resulting page, and deter-

mine the sequence range of the N-terminal domain.Then click on

the UniProt reference (P17967), scroll down to the sequence, copy

the relevant portion to the Jpred3 input box, edit out everything

but the sequence, and click on the “Make Prediction” button. On

the page that comes up indicating that sequence matches were

found in the PDB, click on the “continue” button. When the Re-

sults page appears (you may have to wait some time for it), click

on “View Simple” to see the Jpred3 prediction.The observed sec-

ondary structure of the N-terminal domain is diagrammed on the

foregoing “Sequence Details” page.]

13. Indicate the probable effects of the following mutational

changes on the structure of a protein. Explain your reasoning.

(a) Changing a Leu to a Phe, (b) changing a Lys to a Glu,

(e) changing a Met to a Pro.

14. Explain why Trp rings are usually completely immobile in

proteins that have rapidly flipping Phe and Tyr rings.

15. Explain why Prn-p

0/0

mice are resistant to scrapie. What

might be the susceptibility of heterozygous Prn-p

⫹/0

mice to

scrapie?

*16. Discuss the merits of the hypothesis that the dinucleotide-

binding domains of the dehydrogenases arose by convergent

evolution.

PROBLEMS

JWCL281_c09_278-322.qxd 6/1/10 7:36 AM Page 322

323

CHAPTER 10

Hemoglobin:

Protein Function

in Microcosm

1 Hemoglobin and Myoglobin Function

A. Heme

B. Oxygen Binding

C. Carbon Dioxide Transport and the Bohr Effect

D. Effect of BPG on O

Binding

2 Structure and Mechanism

A. Structure of Myoglobin

B. Structure of Hemoglobin

C. Mechanism of Oxygen-Binding Cooperativity

D. Testing the Perutz Mechanism

E. Origin of the Bohr Effect

F. Structural Basis of BPG Binding

G. Role of the Distal Histidine Residue

3 Abnormal Hemoglobins

A. Molecular Pathology of Hemoglobin

B. Molecular Basis of Sickle-Cell Anemia

4 Allosteric Regulation

A. The Adair Equation

B. The Symmetry Model

C. The Sequential Model

D. Hemoglobin Cooperativity

Appendix: Derivation of Symmetry Model Equations

A. Homotropic Interactions–Equation [10.22]

B. Heterotropic Interactions–Equation [10.23]

also been successful in explaining the control of enzyme ac-

tivity. The first protein X-ray structures to be elucidated

were those of hemoglobin and myoglobin (Greek: myos,

muscle). This central role in the development of protein

chemistry together with its enzymelike O

-binding proper-

ties have caused hemoglobin to be dubbed an “honorary

enzyme.”

Hemoglobin is not just a simple oxygen tank. Rather, it

is a sophisticated oxygen delivery system that provides the

proper amount of oxygen to the tissues under a wide vari-

ety of circumstances. In this chapter, we discuss hemoglo-

bin’s properties, structure, and mechanism of action, both

to understand the workings of this physiologically essential

molecule and to illustrate the principles of protein struc-

ture that we have developed in the preceding chapters. We

also consider the properties of abnormal hemoglobins and

their relationship to human disease. Finally, we discuss the-

ories of cooperative interactions among proteins, both to

better understand the properties of hemoglobin and to set

the stage for our later consideration of how enzyme action

is regulated.

1 HEMOGLOBIN AND

MYOGLOBIN FUNCTION

Hemoglobin (Hb), as we have seen in Chapters 7 and 8, is

a 65-kD heterotetramer, ␣

␤

(alternatively, a dimer of ␣␤

protomers).The ␣ and ␤ subunits are structurally and evo-

lutionarily related to each other and to myoglobin (Mb),

the 18-kD monomeric oxygen-binding protein of skeletal

and heart muscle (Section 7-3C).

Hemoglobin transports oxygen from the lungs, gills, or

skin of an animal to its capillaries for use in respiration.

Very small organisms do not require such a protein be-

cause their respiratory needs are satisfied by the simple

passive diffusion of O

through their bodies. However, since

the transport rate of a diffusing substance varies inversely

with the square of the distance it must diffuse, the O

diffu-

sion rate through tissue thicker than ⬃1 mm is too slow to

support life. The evolution of organisms as large and com-

plex as annelids (e.g., earthworms) therefore required the

development of circulatory systems that actively transport

and nutrients to the tissues.The blood of such organisms

must contain an oxygen transporter such as Hb because the

solubility of O

in blood plasma (the fluid component of

The existence of hemoglobin (Greek: haimato, blood), the

red blood pigment, is evident to every child who scrapes a

knee. Its brilliant red color, widespread occurrence, and

ease of isolation have made it an object of inquiry since an-

cient times. Indeed, the early history of protein chemistry is

essentially that of hemoglobin. The observation of crys-

talline hemoglobin was first reported by Friedrich Hüne-

feld in 1840, and by 1909 Edward Reichert and Amos

Brown had published a photographic atlas of hemoglobin

crystals from several hundred species. In contrast, it was

not until 1926 that crystals of an enzyme, those of jack bean

urease, were first reported. Hemoglobin was one of the

first proteins to have its molecular mass accurately deter-

mined, the first protein to be characterized by ultracen-

trifugation, the first to be associated with a specific physio-

logical function (that of oxygen transport), and, in

sickle-cell anemia, the first in which a point mutation was

demonstrated to cause a single amino acid change (Section

7-3Aa). Theories formulated to account for the coopera-

tive binding of oxygen to hemoglobin (Section 10-4) have

JWCL281_c10_323-358.qxd 2/24/10 1:57 PM Page 323

Fe(II) (ferrous) oxidation state whether or not the heme is

oxygenated (binds O

The Fe atom in deoxygenated Hb and Mb is 5-coordinated

by a square pyramid of N atoms: four from the porphyrin

and one from a His side chain of the protein. On oxygena-

tion, the O

binds to the Fe(II) on the opposite side of the

porphyrin ring from the His ligand, so that the Fe(II) is

octahedrally coordinated; that is, the ligands occupy the six

corners of an octahedron centered on the Fe atom (Fig.

10-1). Oxygenation changes the electronic state of the

Fe(II)–heme, as is indicated by the color change of blood

from the dark purplish hue characteristic of venous blood to

the brilliant scarlet color of arterial blood and blood from a

cut finger (Fig. 10-2).

Certain small molecules, such as CO, NO, CN

⫺

, and H

coordinate to the sixth liganding position of the Fe(II) in

Hb and Mb with much greater affinity than does O

. This,

together with their similar binding to the hemes of cy-

tochromes, accounts for the highly toxic properties of these

substances.

The Fe(II) of Hb and Mb can be oxidized to Fe(III) to

form methemoglobin (metHb) and metmyoglobin (metMb).

MetHb and metMb do not bind O

; their Fe(III) is already

octahedrally coordinated with an H

O molecule in the

sixth liganding position.The brown color of dried blood and

old meat is that of metHb and metMb. Erythrocytes contain

the enzyme methemoglobin reductase, which converts the

blood) is too low (⬃10

⫺4

M under physiological condi-

tions) to carry sufficient amounts of O

for metabolic

needs. In contrast, whole blood, which normally contains

⬃150 g of Hb ⴢ L

⫺1

, can carry O

at concentrations as high

as 0.01M, about the same as in air.

Although Mb was originally assumed to store oxygen,

it is now clear that this function is significant only in

aquatic mammals such as seals and whales, which have

Mb concentrations in their muscles 10- to 30-fold greater

than those in terrestrial mammals. It would seem more

likely that Mb’s major physiological role in terrestrial

mammals is to facilitate oxygen transport in rapidly

respiring muscle. The rate at which O

can diffuse from

the capillaries to the tissues, and thus the level of respira-

tion, is limited by oxygen’s low solubility in aqueous solu-

tion. Mb increases the effective solubility of O

in muscle,

the most rapidly respiring tissue under conditions of high

exertion. Hence, in rapidly respiring muscle, Mb functions

as a kind of molecular bucket brigade to facilitate O

dif-

fusion. It therefore came as a surprise when knockout

mice for the gene encoding Mb exhibited no obvious ab-

normalities (except for the pale color of their muscles)

and exhibited normal exercise capacity and response to

low oxygen levels. However, further investigations re-

vealed that these knockout mice had compensatory adap-

tations including increases in their hematocrit (the frac-

tion of blood volume occupied by red blood cells;

normally ⬃45%) and the capillary density in their mus-

cles. Moreover, many of the mutant embryos died in utero

due to cardiovascular defects. Additional physiological

functions for Mb have recently been recognized: the

detoxification of the highly reactive biological signaling

molecule nitric oxide (NO) through its conversion to nitrate

ion (NO

⫺

) under normal conditions and its synthesis from

nitrite ion (NO

⫺

) under hypoxic (having an inadequate

supply of O

) conditions (see below).

In this section, we begin our discussions of hemoglobin

by considering its chemical and physical properties and

how they relate to its physiological function. Hemoglobin

structure and the mechanisms by which it carries out these

physiological functions are discussed in Section 10-2.

A. Heme

Myoglobin and each of the four subunits of hemoglobin

noncovalently bind a single heme group (Fig. 10-1; spelled

“haem” in British English). This is the same group that oc-

curs in the cytochromes (Section 9-6A) and in certain re-

dox enzymes such as catalase. Heme is responsible for the

characteristic red color of blood and is the site at which

each globin monomer binds one molecule of O

(globins

are the heme-free proteins of Hb and Mb). The hetero-

cyclic ring system of heme is a porphyrin derivative; it con-

sists of four pyrrole rings (labeled A–D in Fig. 10-1) linked

by methene bridges.The porphyrin in heme, with its partic-

ular arrangement of four methyl, two propionate, and two

vinyl substituents, is known as protoporphyrin IX. Heme,

then, is protoporphyrn IX with a centrally bound iron

atom. In Hb and Mb, the iron atom normally remains in the

324 Chapter 10. Hemoglobin: Protein Function in Microcosm

Figure 10-1 The heme group. Fe(II)–heme

(ferroprotoporphyrin IX) is shown liganded to His and O

as it is

in oxygenated myoglobin and oxygenated hemoglobin. Note that

the heme is a conjugated system so that, although two of its

Fe¬N bonds are coordinate covalent bonds (bonds in which the

bonding electron pair is formally contributed by only one of the

atoms forming the bond), all of the Fe¬N bonds are equivalent.

The pyrrole ring lettering scheme is shown.

COO

–

OOC CH

Fe(II)

JWCL281_c10_323-358.qxd 2/24/10 1:57 PM Page 324

small amount of metHb that spontaneously forms back to

the Fe(II) form.

a. Myoglobin Both Detoxifies and Synthesizes

Nitric Oxide

NO, which is synthesized in many tissues, functions as a

locally active signaling molecule, most notably to induce

vasodilation (Section 19-1L).Once NO has delivered its mes-

sage, it is important that it be rapidly eliminated to prevent

its interference with subsequent NO signals (or lack of

them). Moreover, NO is a highly reactive and hence toxic

substance. In muscle, under normal O

concentrations, NO

is detoxified through its reaction with oxygenated myoglo-

bin (oxyMb) to yield nitrate ion and metmyoglobin:

Since the metMb is subsequently reduced to Mb through

the action of an intracellular metmyoglobin reductase,

myoglobin functions as an enzyme in this process. Oxy-

genated hemoglobin (oxyHb) likewise detoxifies the NO

that is present in blood.

Paradoxically, the tissue damage caused by ischemia (in-

adequate blood flow, such as in a heart attack or stroke) is

often exacerbated when blood flow to the deprived tissues

is reestablished. This so-called reperfusion injury is caused,

in part, by the highly destructive reactive oxygen species

(ROS), such as the superoxide ion (O

) and the hydroxyl

radical (ⴢOH), initially produced by hypoxic mitochondria

when the blood supply is restored. Nitrite ion is normally

present in blood and tissues at low (micromolar) concen-

trations. Under hypoxic conditions, deoxyMb catalyzes the

formation of NO from nitrite ion:

NO suppresses the ROS-producing mitochondrial electron-

transport chain (Chapter 22) and thereby protects against

reperfusion injury. Indeed, the administration of nitrite ion

⫺

⫹ H

⫹

⫹ Mb S NO ⫹ OH

⫺

⫹ metMb

ⴢ

⫺

NO ⫹ MbO

S NO

⫺

⫹ metMb

to the hearts of normal mice immediately before their

blood supply was cut off for 30 minutes provided a signifi-

cant protective effect against tissue damage, but afforded

no such protection to the hearts of Mb-knockout mice.

b. Other Globins and Oxygen-Transport Proteins

Hb and Mb belong to the globin superfamily, whose

members occur widely in all three kingdoms of life, where

most participate in a variety of enzymatic and O

-sensing

functions. Evidently, Hb’s O

-transport function is a rela-

tively recent evolutionary adaptation.

Invertebrate hemoglobins vary in quaternary structure

from dimers to as many as 180 subunits. The larger of these,

which are known as erythrocruorins (Latin: cruor, blood),

are extracellular proteins rather than being packaged

within cells; their large sizes (up to 3.5 million kD) permit

them to be retained within their circulatory systems. The

related chlorocruorins (Greek, chloros, pale green), which

occur in the blood of certain annelids, contain a porphyrin

that differs from protoporphyrin IX by the replacement of

its A-ring vinyl group with a formyl group. Consequently,

chlorocruorins are green when deoxygenated and light red

when oxygenated.

Antarctic icefish, the only adult vertebrates that lack

hemoglobin—their blood is colorless—are viable because

of their reduced need for O

at low temperatures combined

with the relatively high aqueous solubility of O

at the

⫺1.9°C temperature of their environment (recall that the

solubilities of gases increase with decreasing temperature).

Many invertebrates lack Hb but instead produce one of

two alternative types of O

-transport proteins:

1. Hemocyanins, which occur in many species of

mollusks and arthropods, form a family of large, multi-

meric, extracellular proteins that differ in their primary

through quaternary structures. However, they have highly

similar O

-binding sites that each contain two Cu ions.

Hemocyanins are blue in complex with O

and colorless

otherwise.

2. Hemerythrins, which are intracellular proteins, occur

in only a few species of marine worms. Their O

-binding

sites contain two non-heme Fe ions. Hemerythrins are

violet-pink in complex with O

and colorless otherwise.

Vertebrates also express two recently discovered glo-

bins: neuroglobin, which is present mainly in brain, retina,

and endocrine tissues, and cytoglobin, which occurs in

most tissues. Neuroglobin protects neurons (nerve cells)

from damage under conditions of hypoxia, most likely by

preventing reperfusion injury in much the same way as

does myoglobin in muscle. Cytoglobin may have similar

functions.

B. Oxygen Binding

The binding of O

to myoglobin is described by a simple

equilibrium reaction

Mb ⫹ O

Δ MbO

Section 10-1. Hemoglobin and Myoglobin Function 325

Figure 10-2 The visible absorption spectra of oxygenated and

deoxygenated hemoglobins.

Relative absorbance

Wavelength (nm)

480 500 520 540 560 580 600 620 640 660

Oxyhemoglobin

Deoxyhemoglobin

JWCL281_c10_323-358.qxd 2/24/10 1:57 PM Page 325

with dissociation constant

[10.1]

(biochemists usually express equilibria in terms of dissoci-

ation constants, the reciprocals of the more chemically tra-

ditional association constants). The O

dissociation of Mb

may be characterized by its fractional saturation, ,de-

fined as the fraction of O

binding sites occupied by O

[10.2]

Since O

is a gas, its concentration is conveniently ex-

pressed by its partial pressure, pO

(also called the oxygen

tension). Equation [10.2] may therefore be expressed:

[10.3]

Now define p

as the value of pO

when that is,

when half of myoglobin’s O

-binding sites are occupied.

Substituting this value into Eq. [10.3] and solving for K

yields K ⫽ p

. Hence our expression for the fractional sat-

uration of Mb finally becomes

[10.4]

a. Hemoglobin Cooperatively Binds O

Myoglobin’s O

-dissociation curve (Fig. 10-3) closely

follows the hyperbolic curve described by Eq. [10.4]; its

is 2.8 torr (1 torr ⫽ 1 mm Hg at 0°C ⫽ 0.133 kPa; 760

torr ⫽ 1 atm). Mb therefore gives up little of its bound O

⫽

⫹ pO

⫽ 0.50,

⫽

K ⫹ pO

⫽

[MbO

]

[Mb] ⫹ [MbO

]

⫽

]

K ⫹ [O

]

K ⫽

[Mb][O

]

[MbO

]

over the normal physiological range of pO

in blood (100

torr in arterial blood and 30 torr in venous blood); for ex-

ample, at pO

⫽ 100 torr and 0.91 at 30 torr. In

contrast, hemoglobin’s O

-dissociation curve (Fig. 10-3),

which has a sigmoidal shape (S shape) that Eq. [10.4] does

not describe, indicates that the amount of O

bound by Hb

changes significantly over the normal physiological range

of pO

in blood. For example, at 100 torr and

0.55 at 30 torr in whole blood for a difference in of 0.40.

Mb therefore binds O

under conditions in which Hb re-

leases it. Thus, the two proteins form a sophisticated O

transport system that delivers O

from lung to muscle

(where pO

may be ⬍20 torr). Hemoglobin’s sigmoidal

-dissociation curve is of great physiological importance;

it permits the blood to deliver much more O

to the tissues

than it could if Hb had a hyperbolic O

-dissociation curve

with the same p

(26 torr; dashed curve in Fig. 10-3). Such a

hyperbolic curve has at 100 torr and 0.54 at 30

torr for a difference in of only 0.25.

A sigmoidal dissociation curve is diagnostic of a cooper-

ative interaction between a protein’s small molecule bind-

ing sites; that is, the binding of one small molecule affects

the binding of others. In this case, the binding of O

in-

creases the affinity of Hb for binding additional O

.The

structural mechanism of hemoglobin cooperativity is dis-

cussed in Section 10-2C.

b. The Hill Equation Phenomenologically Describes

Hemoglobin’s O

-Binding Curve

The earliest attempt to analyze hemoglobin’s sigmoidal

-dissociation curve was formulated by Archibald Hill in

1910. We shall follow his analysis in general form because

it is useful for characterizing the cooperative behavior of

oligomeric enzymes as well as that of hemoglobin.

Consider a protein E consisting of n subunits that can

each bind a molecule S, which, in analogy with the sub-

stituents of metal ion complexes, is known as a ligand. As-

sume that the ligand binds with infinite cooperativity,

that is, the protein either has all or none of its ligand-

binding sites occupied, so that there are no observable in-

termediates ES

,ES

, etc.The dissociation constant for this

reaction is

[10.5]

and, as before, its fractional saturation is expressed:

[10.6]

Combining Eqs. [10.5] and [10.6] yields

which on algebraic rearrangement and cancellation of

terms becomes the Hill equation:

[10.7]Y

⫽

[S]

K ⫹ [S]

⫽

[E][S]

[E](1 ⫹ [S]

>K)

⫽

n[ES

]

n([E] ⫹ [ES

])

K ⫽

[E][S]

[ES

]

E ⫹ nS Δ ES

⫽ 0.79

⫽ 0.95

⫽ 0.97

326 Chapter 10. Hemoglobin: Protein Function in Microcosm

Figure 10-3 Oxygen-dissociation curves of Mb and of Hb in

whole blood. The normal sea level values of human arterial and

venous pO

values are indicated.The dashed line is a hyperbolic

-dissociation curve with the same p

as Hb (26 torr). See

the Animated Figures

1.00

0.80

0.60

0.40

0.20

0.0

0 20 40 60 80 100 120

Hemoglobin

in whole blood

Myoglobin

Venous pressure

Arterial pressure

(torr)

JWCL281_c10_323-358.qxd 2/24/10 1:57 PM Page 326

which, in a manner analogous to Eq. [10.4], describes the

degree of saturation of a multisubunit protein as a function

of ligand concentration.

Infinite ligand-binding cooperativity (n equal to the

number of protein subunits), as assumed in deriving the

Hill equation, is a physical impossibility. Nevertheless, n

may be taken to be a noninteger parameter related to the

degree of cooperativity among interacting ligand-binding

sites rather than the number of subunits per protein. The

Hill equation then becomes a useful empirical curve-fitting

relationship rather than an indicator of a particular model

of ligand binding. The quantity n, the Hill constant, in-

creases with the degree of cooperativity of a reaction and

thereby provides a convenient, although simplistic, charac-

terization of a ligand-binding reaction. If n ⫽ 1, Eq. [10.7]

describes a hyperbola, as do Eqs. [10.3] and [10.4] for Mb,

and the ligand-binding reaction is said to be noncoopera-

tive. A reaction with n ⬎ 1 is described as positively coop-

erative: Ligand binding increases the affinity of E for fur-

ther ligand binding (cooperativity is infinite in the limit

that n is equal to the number of ligand-binding sites in E).

Conversely, if n ⬍ 1, the reaction is termed negatively co-

operative: Ligand binding reduces the affinity of E for sub-

sequent ligand binding.

c. Hill Equation Parameters May Be

Graphically Evaluated

The Hill constant, n, and the dissociation constant, K,

that best describe a saturation curve can be graphically de-

termined by rearranging Eq. [10.7] as follows:

1 ⫺ Y

⫽

[S]

K ⫹ [S]

1 ⫺

[S]

K ⫹ [S]

⫽

[S]

and then taking the log of both sides to yield a linear

equation:

[10.8]

The linear plot of log[Y

/(1 ⫺ Y

)] versus log[S], the Hill

plot, has a slope of n and an intercept on the log[S] axis of

(log K)/n (recall that the linear equation y ⫽ mx ⫹ b de-

scribes a line with a slope of m and an x intercept of ⫺b/m).

For Hb, if we substitute pO

for [S] as was done for Mb,

the Hill equation becomes:

[10.9]

As in Eq. [10.4], let us define p

as the value of pO

. Then, substituting this value into Eq. [10.9],

so that

[10.10]

Substituting this result back into Eq. [10.9] yields

[10.11]

(Note: Equation [10.4] is a special case of Eq. [10.11] with

n ⫽ 1). Equation [10.8] for the Hill plot of Hb therefore

takes the form

[10.12]

so that this plot has a slope of n and an intercept on the log

axis of log p

Figure 10-4 shows the Hill plots for Mb and Hb. For Mb

it is linear with a slope of 1, as expected.Although Hb does

log

1 ⫺ Y

b⫽ n log pO

⫺ n log p

⫽

(pO

)

⫹ (pO

)

K ⫽ (p

)

0.50 ⫽

)

K ⫹ (p

)

⫽ 0.50

⫽

(pO

)

K ⫹ (pO

)

log a

1 ⫺ Y

b⫽ n log[S] ⫺ log K

Section 10-1. Hemoglobin and Myoglobin Function 327

Figure 10-4 Hill plots for Mb and

purified (“stripped”) Hb. Note that this is

a log–log plot. Hence the horizontal axis,

occurs where

(and pO

⫽ p

).Y

>(1 ⫺ Y

) ⫽ 1

log[Y

>(1 ⫺ Y

)] ⫽ 0,

100

0.1

0.01

0.99

0.5

10 100 100010.10.01

1 –

the last

oxygen

bound

the first

oxygen

bound

Myoglobin

slope = 1

Upper asymptote

Hemoglobin

slope 3.0

Lower asymptote

0.9

0.1

0.01

(torr)

JWCL281_c10_323-358.qxd 2/24/10 1:57 PM Page 327

not bind O

in a single step as is assumed in deriving the

Hill equation, its Hill plot is essentially linear for values of

between 0.1 and 0.9. Its maximum slope, which occurs

near pO

⫽ p

is nor-

mally taken to be the Hill constant. For normal human Hb,

the Hill constant is between 2.8 and 3.0; that is, hemoglobin

oxygen binding is highly, but not infinitely, cooperative.

Many abnormal hemoglobins exhibit smaller Hill con-

stants (Section 10-3A), indicating that they have a less than

normal degree of cooperativity. At values near 0, when

few Hb molecules have bound even one O

molecule, the

Hill plot of Hb assumes a slope of 1 (Fig. 10-4, lower

asymptote) because the Hb subunits independently com-

pete for O

as do molecules of Mb. At values near 1,

when at least three of each of hemoglobin’s four O

-binding

sites are occupied, the Hill plot also assumes a slope of 1

(Fig. 10-4, upper asymptote) because the few remaining un-

occupied sites are on different molecules and therefore

bind O

independently.

Extrapolating the lower asymptote in Fig. 10-4 to the

horizontal axis indicates, according to Eq. [10.11], that

⫽ 30 torr for binding the first O

to Hb. Likewise, ex-

trapolating the upper asymptote yields p

⫽ 0.3 torr for

binding hemoglobin’s fourth O

.Thus the fourth O

to bind

to Hb does so with 100-fold greater affinity than the first.

This difference, as we shall see in Section 10-2C, is entirely

due to the influence of the globin on the O

affinity of heme.

It corresponds to a free energy difference of 11.4 kJ ⴢ mol

⫺1

between binding the first and binding the last O

to Hb

(Section 3-4A).

More sophisticated mathematical models than the Hill

equation have been developed for analyzing the coopera-

tive binding of ligands to proteins. We examine some of

them in Section 10-4.

[(Y

⫽ 0.5; Y

>(1 ⫺ Y

) ⫽ 1],

d. Globin Prevents Oxyheme from Autooxidizing

Globin not only modulates the O

-binding affinity of

heme, but makes reversible O

binding possible. Fe(II)–heme

by itself is incapable of binding O

reversibly. Rather,in the

presence of O

it autooxidizes irreversibly to the Fe(III)

form through the intermediate formation of a complex

consisting of an O

bridging the Fe atoms of two hemes.

This reaction can be inhibited by derivatizing the heme

with bulky groups that sterically prevent the close face-

to-face approach of two hemes. Such picket-fence

Fe(II)–porphyrin complexes (Fig. 10-5), which James Coll-

man first synthesized, bind O

reversibly. The back side of

this porphyrin is unhindered and is complexed with a sub-

stituted imidazole in a manner similar to that in Mb and

Hb. In fact, the O

affinity of the picket-fence complex is

similar to that of Mb.Thus, the globins of Mb and Hb func-

tion to prevent the autooxidation of oxyheme by surround-

ing it, rather like a hamburger bun surrounds a hamburger,

so that only its propionate side chains are exposed to the

aqueous solvent (Section 10-2B).

C. Carbon Dioxide Transport and the Bohr Effect

In addition to being an O

carrier, Hb plays an important

role in the transport of CO

by the blood. When Hb (but

not Mb) binds O

at physiological pH’s, it undergoes a

conformational change (Section 10-2Ba) that makes it a

slightly stronger acid. It therefore releases protons on

binding O

where n ⫽ 0, 1, 2, or 3 and x 0.6 under physiological con-

ditions. Conversely, increasing the pH, that is, removing pro-

tons, stimulates Hb to bind O

(Fig. 10-6). This phenomenon,

whose molecular basis is discussed in Section 10-2E, is

known as the Bohr effect after Christian Bohr (the father

⬇

Hb(O

)

⫹ O

Δ Hb(O

)

n⫹1

⫹ xH

⫹

328 Chapter 10. Hemoglobin: Protein Function in Microcosm

0 20 40 60 80 100 120 140

1.0

0.8

0.6

0.4

0.2

(torr)

pH 7.6

pH 7.2

pH 7.4

Figure 10-5 A picket-fence Fe(II)–porphyrin complex with

bound O

. [After Collman, J.P., Brauman, J.I., Rose, E., and

Suslick, K.S., Proc. Natl. Acad. Sci. 75, 1053 (1978).]

Figure 10-6 Effect of pH on the O

-dissociation curve of Hb:

the Bohr effect. The vertical dashed line indicates the pO

actively respiring muscle tissue. [After Benesch, R.E. and

Benesch, R., Adv. Prot. Chem. 28, 212 (1974).]

See the

Animated Figures

JWCL281_c10_323-358.qxd 2/25/10 10:51 AM Page 328

of the pioneering atomic physicist Niels Bohr), who first

reported it in 1904.

a. The Bohr Effect Facilitates O

Transport

The ⬃0.8 molecules of CO

formed per molecule of O

consumed by respiration diffuse from the tissues to the

capillaries largely as dissolved CO

as a result of the slow-

ness of the reaction forming bicarbonate:

This reaction, however, is catalyzed in the erythrocyte by

carbonic anhydrase (Fig. 8-41). Accordingly, most of the

in the blood is carried in the form of bicarbonate (in

the absence of carbonic anhydrase, the hydration of CO

would equilibrate 100-fold more slowly, so bubbles of the

only slightly soluble CO

would form in the blood and

tissues).

In the capillaries, where pO

is low, the H

⫹

generated by

bicarbonate formation is taken up by Hb, which is thereby

induced to unload its bound O

. This H

⫹

uptake, moreover,

facilitates CO

transport by stimulating bicarbonate forma-

tion. Conversely, in the lungs, where pO

is high, O

binding

by Hb releases the Bohr protons, which drive off the CO

These reactions are closely matched, so they cause very lit-

tle change in blood pH.

The Bohr effect provides a mechanism whereby addi-

tional O

can be supplied to highly active muscles. Such

muscles generate acid (Section 17-3A) so fast that they

lower the pH of the blood passing through them from 7.4

to 7.2. At pH 7.2, Hb releases ⬃10% more O

at the ⬍20

torr pO

in these muscles than it does at pH 7.4 (Fig. 10-6).

b. CO

and Cl

⫺

Modulate Hemoglobin’s O

Affinity

modulates O

binding directly and by combining re-

versibly with the N-terminal amino groups of blood proteins

to form carbamates:

The conformation of deoxygenated Hb (deoxyHb), as we

shall see in Section 10-2Ba, is significantly different from

that of oxygenated Hb (oxyHb). Consequently, deoxyHb

binds more CO

as carbamate than does oxyHb. CO

⫹

, is therefore a modulator of hemoglobin’s O

affinity:A

high CO

concentration, as occurs in the capillaries, stimu-

lates Hb to release its bound O

. Note the complexity of

this Hb⫺O

⫺CO

⫺H

⫹

equilibrium: The protons released

by carbamate formation are, in part, taken up through

the Bohr effect, thereby increasing the amount of O

that

Hb would otherwise release. Although the difference in

binding between the oxy and deoxy states of hemoglo-

bin accounts for only ⬃5% of the total blood CO

, it is nev-

ertheless responsible for around half of the CO

trans-

ported by blood. This is because only ⬃10% of the total

blood CO

turns over in each circulatory cycle.

⫺

is also bound more tightly to deoxyHb than to

oxyHb (Section 10-2E). Accordingly, hemoglobin’s O

affinity also varies with [Cl

⫺

]. freely permeates theHCO

⫺

R¬NH

⫹ CO

Δ R¬NH¬ COO

⫺

⫹ H

⫹

⫹ H

O Δ H

⫹

⫹ HCO

⫺

erythrocyte membrane (Section 12-3D), so that once

formed, it equilibrates with the surrounding plasma. The

need for charge neutrality on both sides of the membrane,

however,requires that Cl

⫺

, which also freely permeates the

membrane, replace the that leaves the erythrocyte

(the erythrocyte membrane is impermeable to cations).

Consequently, [Cl

⫺

] in the erythrocyte is greater in the ve-

nous blood than it is in the arterial blood. Cl

⫺

is therefore

also a modulator of hemoglobin’s O

affinity.

D. Effect of BPG on O

Binding

Purified (stripped) hemoglobin has a much greater O

affinity than does hemoglobin in whole blood (Fig. 10-7).

This observation led Joseph Barcroft, in 1921, to speculate

that blood contains some other substance that complexes

with Hb so as to reduce its O

affinity. In 1967, Reinhold

and Ruth Benesch demonstrated that this substance is

D-2,3-bisphosphoglycerate (BPG)

[previously known as 2,3-diphosphoglycerate (DPG)].

BPG binds tightly to deoxyHb in a 1:1 mole ratio (K ⫽

1.5 ⫻ 10

⫺5

M) but only weakly to oxyHb. The presence

of BPG therefore decreases hemoglobin’s oxygen affinity

by keeping it in the deoxy conformation; for example, the

of stripped hemoglobin is increased from 12 to 22 torr

by 4.7 mM BPG, its normal concentration in erythrocytes

–

D-2,3-Bisphosphoglycerate (BPG)

OPO

2⫺

OPO

2⫺

HCO

⫺

Section 10-1. Hemoglobin and Myoglobin Function 329

Figure 10-7 Comparison of the O

-dissociation curves of

“stripped” Hb and whole blood in 0.01M NaCl at pH 7.0. [After

Benesch, R.E. and Benesch, R., Adv. Prot. Chem. 28, 217 (1974).]

1.0

0.8

0.6

0.4

0.2

–1

012

log

(torr)

Stripped

hemoglobin

Whole

blood

JWCL281_c10_323-358.qxd 2/25/10 10:52 AM Page 329

(similar to that of Hb). Organic polyphosphates, such as

inositol hexaphosphate (IHP)

and ATP, also have this effect on Hb. In fact, in birds, IHP

functionally replaces BPG and ATP does so in fish and in

most amphibians. The ⬃2 mM ATP normally present in

Inositol hexaphosphate (IHP)

⫺2

OPO

2⫺

OPO

2⫺

OPO

2⫺

HOPO

2⫺

mammalian erythrocytes is prevented from binding to Hb

by its complexation with Mg

2⫹

BPG has an indispensable physiological function: In arte-

rial blood, where pO

is ⬃100 torr, Hb is ⬃95% saturated

with O

,but in venous blood, where pO

is ⬃30 torr,it is only

⬃55% saturated (Fig. 10-3). Consequently, in passing

through the capillaries, Hb unloads ⬃40% of its O

. In the

absence of BPG, little of this O

is released since hemoglo-

bin’s O

affinity is increased, thus shifting the O

-dissociation

curve significantly toward lower pO

(Fig. 10-8, left).

and BPG independently modulate hemoglobin’s

affinity. Figure 10-8 indicates that stripped Hb can be

made to have the same oxygen-dissociation curve as the

Hb in whole blood by adding CO

and BPG in the concen-

trations found in erythrocytes (the pH and [Cl

⫺

] are also

the same). Hence, the presence of these four substances in

whole blood—BPG, CO

⫹

, and Cl

⫺

—accounts for the

-binding properties of Hb.

a. Increased BPG Levels Are Partially Responsible

for High-Altitude Adaptation

High-altitude adaptation is a complex physiological

process that involves an increase in the amount of hemo-

globin per erythrocyte and in the number of erythrocytes.

It normally requires several weeks to complete. Yet, as is

clear to anyone who has quickly climbed to high altitude,

even a 1-day stay there results in a noticeable degree of

adaptation.This effect results from a rapid increase in the

erythrocyte BPG concentration (Fig. 10-9; BPG, which

cannot pass through the erythrocyte membrane, is synthe-

sized in the erythrocyte; Section 17-2Hb).The consequent

decrease in O

-binding affinity, as indicated by its elevated

330 Chapter 10. Hemoglobin: Protein Function in Microcosm

Figure 10-8 The effects of BPG and CO

, both separately and

combined, on hemoglobin’s O

-dissociation curve compared with

that of whole blood (red curve). In the Hb solutions, which were

0.1M KCl and pH 7.22, pCO

⫽ 40 torr and the BPG concentration

was 1.2 times that of Hb. The blood had pCO

⫽ 40 torr and a

plasma pH of 7.40, which corresponds to a pH of 7.22 inside the

erythrocyte. [After Kilmartin, J.V. and Rossi-Bernardi, L.,

Physiol. Rev. 53, 884 (1973).]

See the Animated Figures

1.0

0.8

0.6

0.4

0.2

0 10203040506070

Stripped Hb

Hb + BPG +

Hb + BPG

Hb + CO

Whole blood

(torr)

160

140

120

100

–1 0 1 2 3 4 5 6/0 1 2 3

BPG concentration

( g phosphorus

–1

blood) μ

Time (days)

(torr)

Sea

level

Sea

level

4530 m above

sea level

BPG concentration

Figure 10-9 The effect of high-altitude exposure on the p

and the BPG concentration of blood in sea level–adapted

individuals. The region on the right marked “Sea level” indicates

the effects of exposure to sea level on high altitude–adapted

individuals. [After Lenfant, C.,Torrance, J.D., English, E., Finch,

C.A., Reynafarje, C., Ramos, J., and Faura, J., J. Clin. Invest. 47,

2653 (1968).]

JWCL281_c10_323-358.qxd 2/24/10 1:57 PM Page 330