BME 201 Practice Exam 2 - i 9 E E e 2 2' E S i E E i i BME...

Info iconThis preview shows pages 1–20. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 14
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 16
Background image of page 17

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 18
Background image of page 19

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 20
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: i 9 E E e 2 2' E S i E E i i BME 201 Biomolecules Fall 2007 Instructor: Gudrun Schmidt Teaching Assistant: Vincent Chan Exam 2 Tue. Oct. 23, 2007 8:30 - 10:00pm EE 270 Name: M5 (Mg- 12. Please place all answers in the space provided after each question. Do not write on the back side or additional paper. No other answers will be graded. No calculators and no cheat sheets allowed. Euaapa 1) Buffers regulate pH in organisms. a) What type of buffer does “nature” use to regulate the pH of human blood? b) Draw the chemical equation (for the buffer system) that is most important for maintaining acid base balance in blood. What does the H of blood depend on? JR c) Draw the titration curve (qualitative) showing the region of maximum buffering capacity. (1) If blood had a normal pH of 6.1 instead of 7.4 would you expect exercise to result in heavy breathing? Discuss Within 4 sentences. “E (02‘ MkMMW—___._M.__...__.M_ dlfl} labial madam“ (at at. 591(WMW4KJW§&WL=LN ‘ - \ - ‘ bank [MLA aha. aaaa amfi Gama ut&.cmuaaaaua2bg§ MW 8% wao?wui -TEl/lou.€l£5a\ (Wig, Lima {5 thou W s .l-'Z.l .1“ «is “Q hamwok Claus ‘ mm W 51211 ngzwaFWlOWZDMSrH“ Ml. 2) Titration curves for amino acids. Draw Within one diagram the titration curves for a) a“simpld‘amino acid, b) an acidic amino acid and c) a basic amino acid (use eg. dotted , dashed lines). Name rel/simple, basic and acidic amino acid and w the molecular structures. 19 05‘“- k ca + ~G__ COO, ‘ \ g \ a H23“ 1 .1 H 4.4%“ “QECQQA + It; U -— a I» coo H E k ‘ a ! pa Pam?” H “(flaunt ‘ I ‘ f if? R f g r MMWNM QM!“ (SJ k W / i a or » EHBRHE 3) A. Draw the ionization steps of a diprotic amino acid. B. What is the isoelectric point? C. What does the isoelectric point have to do with isoelectric focusing? ? Discussr H K i __ C00 ? ([1, (flag + gr '2- r1 i s 63 Hang- Cficmg‘) g mpwc—Cm@+ H 62 a ‘ :4 u air LLLL H’A at W qu&&wfiwufllfiwfl&%%hfl.fl P Lo \ 0], wxev-wkCSy-W, M \ (Dw — MW: wamwfim KM +9.; Limo“ A much W a maqu “W5 to Pi Mommy D. What is the isoelectric point of serine? Calculate using attached table if necessary) f j cal-Ml _ ijflgég- [9 A ‘2— 2' wp~fls~++- '''' ‘tzsf“"‘*“‘“"‘ \ Uh 4) Gel filtration chromatography of a mixture of proteins should release which protein molecules first? Why? Is this separation very specific? Discuss What type of chromatography would you choose for an extremely specific separation? 5) Polypeptides that are linked by disulflde bonds may cause you some trouble during separation. What trouble could that be and how would ou overcome this? a? ’0) w >\e (Um WM” 6 QKMA or 6‘“ WW “S Wigwam? 6) During a tissue engineering experiment, it would be helpfiil for you to Visualize proteins within the cells. How do you proceed? Emmi— wag-MS “493% fossiw We di$w~k¥~skom EWLQMLL wlmtnw We mm. at a, 8) What type of bonds are responsible for the double helix structure of DNA? How would yen a) weaken these bonds and b) how would you “denaturate” or disrupt the double helix? How would you “dissolve” the helix completely? a) f 31w“ M Valli“. ., 9) Hydrophobic Effect: Why is the aggregation ofnonpolar molecules in water favored? What biological relevance do you think this has? Be creative © TW— icbfmkow a» wupceof WGQLwfl-V-a am am wefiwflg wolf an Java-smbb Qua ulfladx‘ous W cam-f \ 5410 (Mg «111140—91 {10w tug-o. TM 0? wuwo‘ek {S ENMRELL 1%)“ Wu“ o‘ofiqflwflok 10) Why do single DNA strands react to form a double helix? Explain from a thermodynamic point of View? SPQW‘da. K ‘ WW 430% mm yam ‘2. 9% ‘ ° 4 "" (Amt Lam g‘HMal/a stkeflww‘m fa +341 stfiw, 3&1;me 6)st 6: ZEN; 0} ___> S M‘ W W}. [at aim/4,01 SWOWCX‘ "'0 Maia, ll 11) a) Name and draw an amino acid with two chiral centers? b) What do lysinenand arginine have in common? c) Why is proline unique among the 20 amino acids? (1) What type of amino acid can be oxidized to form covalent cross-links within a protein chain structure and what type of bonds are these? Write down the oxidation reaction (cross-linking reaction) for two such amino acid molecules. Ch) its-saw (iul H w-OUSUQ'iF‘b‘ MS (9%“ _. a, wee) i (\- Oh (Blip —d——Qeo@ ( ‘ e9 its -2}??? —— 2e S S“. : i S Q” \ I 12 Gin, @ l Lou69 (la; “by {C(— GQJfbp » c — £0003 Hr l H, 12) Show all possible forms of cysteine in solution: Q00 W 1 ®4+3U r (‘14 06“ SH f 13 UEJE‘JMEE‘)§'E: 13) What are the physical parameters of a protein that control its migration during electrophoresis? 9G1 (th (JAN-06L 14) Why can Edman degradation not be used effectively with very long peptides? How would you proceed with analyzing very long peptides? 14 Amino Acid' pKa Values'fhttQ://mvw.cem.msu.edul~cem25215997/ch24/ch24aa.htmlJ it Amino Acid iia—carboxylic acidHocmminoHSide chainj IIAJEHJILWJLL_WJI9§LJI ______: JAIginiggmmwg j_____l_%m_. _ aAspartl A . [2.23 “9.21" [wwwme lPhenylaiafliqgjjggg mmmm mm “9.24 }[_______W_m§ IP£0_I@__§i2-00 WW1! Wu. Serine uw;l2.21fl___m_mmng;l—ijl wmmj' mmbflwhuv‘b-m—u .......—M..“_*M.._._...._____. WWW? l i A structural model for Alzheimer's fl-amyloid fibrils based on experimental constraints from solid state NMR Aneta T. Petkova*, Yoshitaka Ishii“, John J. Balbach", Oleg N. Antzutkinf, Richard D. Leapman5, Frank Delaglio*, and Robert Tycko‘l'n *Laboratory of Chemical Physics, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892-0520; *Department of Chemistry, Luiea University of Technology. Luiea, Sweden 5-97187; and §Division of Bioengineering and Physical Science, Office of Research Services, National institutes of Health, Bethesda, MD 20892-5766 Communicated by David R. Davies, National Institutes of Health, Bethesda, MD, November 1, 2002 (received for review August 30, 2002) We present a structural model for amyloid fibrils formed by the 40-residue B-amyioid peptide associated with Alzheimer’s disease (A3145). based on a set of experimental constraints from solid state NMR spectroscopy. The model additionally incorporates the cross-:3 structural motif established by x-ray fiber diffraction and satisfies constraints on A131-“ fibril dimensions and mass—per—length deter- mined from electron microscopy. Approximately the first 10 residues of A3140 are structurally disordered in the fibrils. Residues 12-24 and 30—40 adopt B—strand conformations and form parallel E-sheets through intermolecular h dro en bonding. Residues 25—29 contain a bend of the peptide backbone that brings the two B-sheets in contact through sidechain-sidechain interactions. A single crOSs-B unit is then a double-layered B-sheet structure with a hydrophobic core and one hydrophobic face. The only charged sidechains in the core are those of D23 and K28, which form salt' bridges. Fibrils with minimum mass-per-length and diameter consist of two cross-,6 units with their hydrophobic faces juxtaposed. myloid fibrils are filamentous structures, with typical diame- ters of a10 nm and lengths up to several micrometers, formed by numerous peptides and proteins with disparate sequences and molecular weights. Biomedical interest in amyloid fibrils arises from their occurrence in amyloid diseases (1), including Alzheimer’s disease, type 2 diabetes, Huntington’s disease, and prion diseases. Current interest in the molecular structures of amyloid fibrils additionally arises from fundamental Questions regarding the mo- lecular mechanism of amyloid formation and the nature of the intermolecular interactions that stabilize these structures for an efiremely diverse class of polypeptides. o high-resolution mo ecu ar structure of an amyloid fibril has yet been determined experimentally because amyloid fibrils are noncrystalline solid materials and are therefore incompatible with xray crystallography and liquid state NMR. X—ray fiber diffractiOn shows that amyloid fibrils contain cross-,8 structural motifs, i.e., extended ,B-sheets in which the B—strand segments run approxi» mately perpendicular to, and the intermolecular hydrogen bonds run approximately parallel to, the long axis of the fibril (2, 3). Other molecular-level structural features of amyloid fibrils are not well established. In the case of fibrils formed by the full-length ,B-amyloid peptide associated with Alzheimer’s disease (All), which ranges from 39 to 43 residues in length in vivo (4, 5), several molecular models have been proposed (6—-10). These models exhibit many qualitative and quantitative differences, reflecting the paucity of experimental constraints. All of these models are inconsistent with recent mea- surements of l3C—13C nuclear magnetic dipole-dipole couplings (i.e., intermolecular distances) by solid state NMR (11—13), which imply an in-register parallel alignment of peptide chains within the cross—B motif in Ali—40 and A8142 fibrils (A13m_H denotes residues m to n of AB). Earlier solid state NMR measurements shewed the same in-register parallel alignment in A310,}; fibrils (14—16). 15742—16747 1 PNAS | December24,2002 1 vol. 99 | no.26 Here, we describe a molecular model for A6140 fibrils that is based on solid state NMR and other experimental data. The N MR data include: (i) 15N and 13C chemical shifts and NMR linewidths determined from 2D 13C—13C and 15N—"3C chemical shift correlation spectra of samples in which selected residues are uniformly 15N- and 13C—labelecl, which serve to identify B-strand segments, non-B- strand segments, and disordered segments; (ii) constraints on backbone (is and lit torsion angles obtained from measurements on doubly 13C—labeled fibril samples, which permit a quantitative characterization of non-Bastrand conformations at certain sites; and (iii) recent measurements of intermolecular 13C—13C distances (11, 12). Additional experimental constraints on fibril dimensions and mass-per-length (MPL) are obtained from electron microscopy (EM) measurements. The model, which is derived by incorporation of the experimental constraints into an energy minimization pro- cedure, reveals h0w amyloid fibrils with apparently favorable elec- trostatic and hydrophobic interactions can be constructed from the full-length AB peptide sequence. Materials and Methods Sample Preparation. Peptides with the human A3140 sequence DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLM— VGGVV were synthesized, purified, and fibrillized from 0.254 to 1.0-mM solutions at pH 7.4 as described (ll, 12). Fibrillized solutions were lyophilized for solid state NMR measurements. Typical solid state NMR samples were 10 mg. For EM, fibrillized solutions were diluted by a factor of 10—20 and negatively stained with uranyl acetate as described (11, 13). The following samples were synthesized with uniform 15N and 13C labeling of the specified residues: SU7 (F19, V24, (3125, A30,131, L34, M35), SU6 (A2, D7, 09, Y10, V12, M35), SUS (D23, K28, G29, 132, V36), and CU6 (K16, L17, V18, F19, F20, A21). The following samples were synthesized with 13C labels at the specified pairs of backbone carbonyl sites: DLl (D23, V24), DL2 (V24, G25), DL3 (G25, 826), DL4 (K28, 629), and DLS (629, A30). The notations SUn, CUn, and DLn indicate “scattered uniform” label- ing of n residues, “consecutive uniform” labeling of 11 residues, and the nth “double labeled” sample, respectively. Solid State NMR Measurements. For SUn and CUn samples, 2D '3C/ BC chemical shift correlation spectra were acquired at a protOn 7 NMR frequency of 400.9 MHZ and magicuangle spinning (MAS) frequencies of 21.7—23.3 kHz, using finite-pulse radio-frequency- driven recoupling (prFDR) mixing periods of 1.37—1.47 ms, as Abbreviations: EM, electron microscopy; MAS, magic-angle spinning; prFDR, finite-pulse radio-frequenEy-driven recoupling; DQCSA, double quantum chemical shift anisotropy: DLn, nth double labeled sample; MPL, mass-per-length. 1Present address: Department of Chemistry, University of lliinois, 845 West Taylor Street, Chicago, IL 60607. lliTo whom correspondence should be addressed at: National Institutes of Health, Building 5, Room 112, Bethesda, MD ZGBBZOSZQ Eemailz tycko®helix.nih.gov. www.pnas.org,/cgi,r‘doi/10.1073/pnas.252653499 s . i i a i i s t i 5, t t Z .2 i i t a l a 60 5C} 4‘} 30 20 it} 60 50 4t) 30 20 to Mama) Fig. 1. (a) Transmission electron microscope images of negatively stained amyloid fibrils after 14-day incubation of a 0.5 mM A3140 solution. A 3X expansion (inset) shows fibrils with the smallest diameters observed. (in) 2D 13C-33C chemical shift correlation spectrum of A3140 fibril sample SU7, show- ing resonance assignment paths for the seven uniformly 15N- and ‘3C-labeled residues in this sample. (6) Expansion of the aliphatic region of the 20 spectrum of SU7. (d) Aliphatic region of the 2D tic—*SC chemical shift corre- lation spectrum of ABM" fibril sample SUE. described (17, 18). 2D ISN/13C chemical shift correlation spectra were acquired at an MAS frequency of 9.0 kHz, using a frequency- selective 15N—13C cross-polarization technique with mixing periods of 1.2—2.5 ms, as described by Baldus er a]. (19). For DLn samples, constant-time prFDR (prFDR-CT) data were acquired at proton NMR frequencies of 400.9 and 399.2 MHz, using an MAS frequency of 20.0 kHz and with the parameters K = 3 and M + 2N I 96, in the notation of lshii at at. (20) Double quantum chemical shift anisotropy (DQCSA) data were acquired as described (21), using an MAS frequency of 5.0 kHz and DO preparation and mixing periods of 6.4 ms. Sensitivity of prFDR-CT and DQCSA measurements was enhanced by pulsed spin locking (22). 2D MAS exchange data were acquired at a proton NMR frequency of 599.1 MHz as described (23), using an MAS frequency of 3.78 kHz and 500-1115 exchange periods. NMR Data Analysis. Backbone torsion angles were predicted from chemical shift data as follows: (i) the TALOS database (24) was used to generate best-tit functional dependences of 13C and 15N secondary shifts on d) and it: (i.e., chemical shift surfaces); and (ii) for each seven-residue segment of A3140, an exhaustive com- parison with predicted chemical shifts for seven-residue segments in a nonredundant subset of protein structures in the Protein Data Bank (PDB) was conducted. For each PDB segment, a score was calculated according to the TALOS formula (24). The 10 most closely matching PDB segments were retained. The average ()5 and ti} values for these 10 segments, and the ranges of these values, are reported in Table 2. Backbone torsion angles were determined from data on DLn samples by comparison with numerical simulations, using FORTRAN programs that simulate the dependence of the data on 4) and 111(20, 21, 23). Simulations of prFDR-CT data were carried out for a fouraspin system (two carbonyl labels on two parallel peptide Petkova et at. U1 4:. M "an 130 NMR linewidth (ppm) 0.) (D 71016182023252913436 2 9121?19212428303235 Residue Fig. 2. 13C NMR linewidths for CO, Ca, and CB sites in Afi1—4o fibrils, deter- mined from 2D solid state NMR spectra as in Fig. 1. Linewidths of 2.5 ppm or less indicate well—ordered conformations. Larger linewidths in the N-terminal segment indicate structural disorder. chains). Simulations of DQCSA and 2D MAS exchange data were carried out for a two-spin system (two carbonyl labels on one chain). CSA principal values in simulations were determined experimen— tally from 1D MAS spectra (25). Molecular Modeling. The model in Fig. 4 was generated by energy minimization, using the CHARMm force field and algorithms contained in QUANTA 97 (Molecular Simulations, Waltham, MA), starting from an initial peptide conformation generated manually within MOLMOL (26). Residues 1—8 were considered disordered and were omitted. For residues 9—23 and 31—40, initial ab and divalues were taken from chemical shift predictions (set 1 in Table 2). For residues 24—30, initial 4') and 11; values were taken from measure- ments on DLn samples. Where no experimentally derived values were available, (l) and tpwere set to —140° and 140°, respectively. All (9 angles were set to 180°. Signs of the (I; and 0'} values from DLn data (see below) were chosen to permit an approximate alignment of all backbone carbonyl C20 and amide N—H bonds along a single intermolecular hydrogen-bonding direction, as required by the cross—,8 structural motif. Five copies of the initial peptide confor- mation related by S—A displacements along the hydrogen~bonding direction were generated so that energy minimization would take place in the context of a fiveestranded cross-,8 structure. Energy minimization included bond, angle, dihedral, improper, and van der Waals energy terms, but not electrostatic energy. The inuregister parallel alignment within the cross—18 motif was enforced by distance constraints between backbone carbonyl oxygens of each residue k and the backbone amide hydrogen of residue k + 1 of a neighboring chain, producing the 4.8-A backbone—backbone distance seen in diffraction data (3, 27, 28). Torsion angle constraints were included with the target values described above and with force constants that resulted in typical deviations of less than 10° from the target values in the energy-minimized structure. For N27 and K28, torsion angle force constants were reduced by a factor of 20, reflecting the absence of experimental constraints. An initial stage of energy minimization, using only the con- straints described above, resulted in a structure consisting of two Separate, parallel I(El—sheets, created by residues 9—24 and 30—40, with a net bend angle of ==60° between them due to nonrflwstrand conformations at G25, $26, and (329. Additional intramolecular distance constraints were then applied between Cy of D23 and N; of K28 (see below), between C; of F19 and Ca of G33, apd between N3 of 015 and Ca of G37, with target distances of 4.5 A. A second PNAS I December24.2002 I vol.99 I no. 26 I 16743 i Table 1. 13C and “N NMR chemical shift values (ppm) for 13C- and 15N-labeled sites in A31-“ fibrils, referenced to TMS (tetramethylsilane) or liquid NH; Residue C0 (3,, CB CT C5 1:. Cg, N); N Sample 42 173.7 49.9 13.2 ND 505 (176.1) (50.3) (17.4) 07 37173.0 51.5 40.4 177.9 120.6 506 (174.5) (52.5) (39.4) (173.3) (120.4) 6.9 169.3 42.9 107.2 506 (173.2) (43.4) (103.3) v10 172.0 55.0 39.5 126.5 130.7 116.5 155.2 122.4 SU6 (174.2) (55.2) (37.1) (123.9) (131.5) (116.5) (155.5) (120.3) v12 173.0 53.7 33.2 13.3, 13.3 127.0 506 (174.6) (60.5) (31.2) (19.4, 13.6) (119.2) K16 171.5 52.7 34.1 24.0 23.6 39.3 33.7 ND (:06 35.9 24.3 (174.9) (54.5) (31.4) (23.0) (27.3) (40.2) (327) L17 172.3 52.3 44.5 25.0 ~24.4, ~23.3 ND cue 173.0 51.5 «40.3 »~27.0 -»24.3, --»23.1 (175.9) (53.4) (40.7) (25.2) (23.2, 21.5) v13 170.3 53.9 33.5 19.2 121.7 c06 (174.6) (60.5) (31.2) (19.4, 13.6) (119.2) F19 170.2 55.3 41.0 135.7 129.6 129.5 125.3 130.5 (:06. 507 (174.1) (55.0) (37.9) (137.2) (130.2) (129.3) (123.2) (120.3) F20 170.2 54.6 41.0 135.7 129.2 129.2 125.3 ND c06 (174.1) (55.0) (37.9) (137.2) (130.2) (129.3) (123.2) 1421 172.7 43.2 21.0 130.9 ms 174.0 43.0 13.9 126.3 (176.1) (50.3) (17.4) (123.3) 023 173.1 51.0 41.9 130.2 113.5 505 174.1 52.4 39.4 173.0 123.3 (174.5) (52.5) (39.4) (173.3) (120.4) 024 173.3 53.6 31.3 19.9, 13.3 125.0 507 173.6 59.0 32.3 19.9, 19.9 125.0 (174.6) (50.5) (31.2) (19.4, 13.5) (119.2) 625 174.2 44.4 113.9 507 171.1 46.9 117.8 171.1 ~-44.2 113.9 (173.2) (43.4) (1033) 1:23 174.3 52.3 35.5 24.7 27.3 42.0 33.0 119.5 505 172.4 53.5 33.4 22.3 23.5 39.1 32.9 112.7 (174.9) (54.5) 131.4) (23.0) (27.3) (40.2) (32.7) (120.4) 629 172.4 47.2 117.0 505 153.6 42.4 104.1 (173.2) (43.4) (108.8) 430 173.2 43.4 20.5 122.1 507 127.4 171.3 49.5 20.5 119.2 (176.1) (50.8) (17.4) (123.8) 131 172.5 53.4 33.0 25.7, 13.3 13.3 120.6 507 (174.7) (59.4) (37.1) (25.5, 15.7) (11.2) (119.9) 132 173.3 55.7 40.2 25.2, 15.9 12.4 125.0 505 172.2 57.0 33.7 24.6. 15.3 12.1 125.0 (174.7) (59.4) (37.1) (25.5, 15.7) (11.2) (119.9) 1.34 171.0 52.1 44.3 27.0 ~24.0, 22.5 423.0 507 44.1 25.3 24.0, 22.9 ~123.0 (175.9) (53.4) (40.7) (25.2) (23.2, 21.5) (121.3) M35 171.2 52.1 34.6 30.3 15.7 125.4 507, 506 (174.6) (53.7) (31.2) (30.3) (15.2) (119.6) v35 171.3 53.3 31.9 13.9 125.5 505 (174.6) ‘ (60.5) (31.2) (19.4, 13.6) (119.2) Values preceded by m have an uncertainty of 0.6 ppm. Otherwise, the uncertainty is 0.3 ppm. Values that could not be determined are indicated by ND. Values in parentheses are randemrcoii shifts, taken from Wishart er al. {51) and adjusted to the TMS reference. stage of energy minimization resulted in a bend angle of #180“, i.e., a double-layered lB—sheet structure. In the final stage of energy minimization, distance constraints between F19 and G33 and between 015 and G37 were removed. The double-layered structure was maintained by van der Waals interactions between sidechains of the two fi-sheets. F inal torsion angles of the central peptide chain agree with target values to within an rms deviation of 4.1“ for 42 and 6.7° for 117, with the largest deviations being 11° for (b and 15° for 111 16744 | www.pnas.org7'cgi7'doi/10.1073/pnas.262663499 (except for a deviation of 28° for t); of N27). All 07 values are within 6“ of 180“. No unrealistic distortions of bond angles or steric clashes are observed. Results Fibril Dimensions and Morphologies. Fig. in shows representative EM images of ABM)" fibrils. Fibrils exhibit a diversity of morphol- ogies, often with a periodic modulation in diameter suggesting a Petkova et al. F l s i i a i t ; twist about the long axis. Similar morphologies have been reported previously for AB and other amyloid fibrils, in both EM and atomic force microscope images (29—31). The narr0west A3140 fibrils have diameters of 50 i 10 A. Characterization of Structural Order. Fig. 1 b and c shows the 2D 13C—13C chemical shift correlation spectrum of ABM“) fibril sample SU7 (see Materials and Methods for sample descriptions). Strong crosspeaks in the spectrum connect the chemical shifts of directly bonded, labeled carbon sites. The crosspeaks are readily assigned based on the chemical structures and known 13C chemical shift ranges of the amino acid sidechains. l3C NMR linewidths deter- mined from the full-width at half height of the crosspeaks are 1.5—2.5 ppm for fibrillized SU7. Fig. 1d shows a portion of the 2D 13C-13C correlation spectrum of A3140 fibril sample SU6. Line- widths for M35 are the same in SU6 and SU7, but significantly broader lines (3.0—5.5 ppm) are observed for labeled carbon sites of A2, D7, G9, and Y10 in SU6. 13C NMR linewidths for CO, Car, and CE sites in ABMD fibril samples SUS, SU6, SU7, and CU6 are plotted in Fig. 2. Linewidths in the 1.5-2.5 ppm range in solid state “C MAS NMR spectra are characteristic of well-structured peptides in rigid noncrystalline environments, whereas significantly larger linewidths are observed in disordered biopolymers (32, 33). Fig. 2 shows that the N—terminal segment of ABHD is disordered in the fibrils, with full structural order beginning after YlO. The N—terminal residues in A6149 and A1314; fibrils are not required for fibril formation (34) and have been shown to be susceptible to proteolysis in brain tissue (35, 36) and in vitro (37). Identification of IB-Strand Segments. NMR chemical shifts obtained from 2D 13C-‘3C correlation spectra as in Fig. 1 and from 2D l5N—"3C correlation spectra (data not shown) are summarized in Table 1. As shown previously by NMR experiments (38, 39) and ab initio calculations (40), secondary shifts A6 E 551,,“ — 5,0“ are streneg correlated with peptide or protein backbone conformation, where 551,,“ is the chemical shift in the folded or fibrillized state and 5,0“ is the random coil shift (i.e., the shift for the same residue type in an unstructured peptide in aqueous solution). In particular, A8 values for ,B-strand segments are characteristically negative for 13C0: and 13CO sites and positive for 13’C,G sites. Certain labeled residues exhibit more than one distinct set of chemical shifts, resulting in more than one set of 2D crosspeaks. The relative intensities of the crosspeaks in different sets vary with fibrillization conditions. We attribute the multiplicity of chemical shifts to differences in molecular structure associated with the differences in fibril morphology apparent in Fig. 1a and typically observed for AB fibrils (29) and other amyloid fibrils (30). Although the nature and degree of structural differences at the molecular level are not yet clear, all sets of‘3C chemical shifts for residues 9—21 and 30—36 in Table 1 consistently indicate ,B-strand conformations. In contrast, at least some 13C chemical shifts for D23, V24, G25, and G29 are inconsistent with expectations for a B-strand. Thus, the chemical shift data qualitatively suggest a conformation for the structurally ordered part of A3140 consisting of two B-strands that are separated by a bend or loop contained within residues 23—29. The conformation in the bend segment may vary with fibril morphology and fibrillization conditions. 13C chemical shifts for CO, Cor, and C3 sites and 15N chemical shifts for backbone amide sites were analyzed with an algorithm that predicts the backbone torsion angles d) and rpfor each uniformly labeled residue (see Materials and Methods). Predictions for two different choices of chemical shift values are given in Table 2. Both choices lead to q!) = —135° : 25° and if! = 140° 1' 20°, consistent with a B-strand conformation, for all residues in the 9—21 and 30—36 segments. Non-B—strand d) and lit values (and significant differences for different choices of chemical shift values) occur at D23, G25, and G29. Petkova et al. a Table 2. Residue—specific (l) and at: backbone torsion angles (degrees) for M31-“ fibrils, predicted from 13C and 15N chemical shifts in Table 1 or determined from measurements on the doubly 13C—lal2Ieled DLn samples «is, whom chemical «‘15, whom chemical (l), urtrom Residue shift set 1* shift set 21' DLn samples G9 —14S:11,151:15 —14s:11,151:15 Y10 -127i9,124:9 —127:9,124:9 V12 “119:8, 124:10 A119i8,124:10 K15 i149:12,15218 i149112,152:8 L17 —150:12,143:9 "15011214319 V18 —145:9,147: 1‘] —145:8,146:12 F19 —144:10,141:tz —144:10,139:15 F20 —147:9,151:11 —145:11,152i13 A21 -137:12,143:16 -127t11,141:19 D23 7145:1G,147:16 783i13,122:22 V24 *ID3I10,11?311 i100212,114122 mill-5,115 625 ""88 : 30, 124 t 33 ""53 1 48, 11 : 74 -70, —40 526 68, —55 K28 —134:12,152:14 451114156113 529 -59:SO,119:58 —150:18,156: 14 -120.—125 A30 -138 :14,157 : 14 —144 :12,145 i 13 -165. 133 B1 -113:16.127:12 -118:15,129i11 i32 7123 :10,146 :14 712? i 9.147 t 12 L34 71:13:53,145117 444:3,145: 16 M35 i14119,138:11 i14119,138:11 V35 "11813120211 —11818,120:11 *First chemical shift value for each labeled site in Table 1. TSecond chemical shift value for each labeled site in Table 1. where more than one value is observed. Determination of Backbone Torsion Angles at Non-B-Strand Sites. Measurements were carried out on the DLn samples to further constrain the backbone conformation in residues 23—30. Three solid state NMR techniques that place independent constraints on the (l: and all torsion angles of the second labeled residue in each sample (i.e., v24 in DL1, (325 in DL2, 526 in DL3, (329 in DL4, and A30 in DLS) were used (20, 21, 23). Fig. 3 a and b shows prFDR-CT and DQCSA data for three samples and simulations that illustrate the sensitivity of these data to backbone conformation. In prFDR—CT measurements, the decay of 13C NMR signals from the labeled carbonyl sites reflects the strength of l3C—li'C dipole» dipole couplings, which depends primarily on the intramolecular 13C~13C distance and hence the d; angle. In DQCSA measurements, the decay of 13C NMR signals from the labeled sites reflects the relative orientation of the labeled carbonyl groups, which depends on both ()5 and it. As is apparent in Fig. 3, prFDR—CT and DQCSA data for different samples are significantly different, indicating a: U‘” 120mm “MWWWW ‘ __ ._ ' a dame g 9'3! o .ezsrsza 4, .9! & vzaroas } {fl ' . of. 6W a r l O 304: 4i i I! . [ 0'5'1’91‘52'025 u ’19 2'0 3's 4?” Bipolar evolution lime (ms) CSA evolution time {is} Fig. 3. Solid state NMR data on DLn A5140 fibril samples with 13C labels at the indicated backbone carbonyl sites. These data constrain the ¢and dangles of the second labeled residue. (a) prFDR-CT data and simulations for it: = 40" (solid line), 80" (dashed line), 120" (dot-dashed line), and 160° (dotted line). Simulations are scaled and baseline-corrected to match the first and last experimental data points. (b) DQCSA data and simulations for (b, ll: 2 "70°, 740" (green); 70”, fi65“ (red); and 7165“, 135° (black). PNAS | December24,2002 l vol.99 E no. 26 | 15745 b 4 V36 ‘32 K28 its 14 K6 F29 Fig. 4. Structural model for 9451-40 fibrils, consistent with solid state NMR constraints on the molecular conformation and intermolecular distances and incorporating the cross-,3 motif common to all amyloid fibrils. Residues 1—8 are considered fully disordered and are omitted. (a) Schematic representation of a single molecular layer, or cross-B unit. The yellow arrow indicatesthe direction of the long axis of the fibril, which coincides with the direction of intermolecular backbone hydrogen bonds. The cross-t3 unit is a double-layered structure, with in—register parallel B—sheets formed by residues 12—24 (orange ribbons) and 30410 (blue ribbons). (b) Central A1314.) molecule from the energy—minimized. fivechain system, viewed down the long axis of the fibril. Residues are color— coded according to their sidechains as hydrophobic (green), polar (magenta), positive (blue), or negative (red). significant differences in qb and ill. Qualitatively, this result indicates the presence of non-B-strand conformations. Data from 2D MAS exchange measurements (not shown) are also significantly different for different DLn samples. Values of d) and it: were determined from the prFDR-CT, DQCSA, and 2D MAS exchange data by comparison with numer- ical simulations as follows: (i) a range of 4) values that gave an acceptable fit to the prFDR—CT data was determined; (ii) the best-fit values of qb and difor DQCSA and 2D MAS exchange data contained within this range were determined; and (iii) the averages of the two best-fit (l) and it: values were taken as the final values, which are reported in Table 2. Note that these measurements are invariant to the substitution (1), it; —> web, — (it because of symmetry considerations (23). Signs of d) and IllVElluES in Table 2 were chosen by molecular modeling to give an A3140 conformation consistent with the required cross-B structural motif. Agreement between torsion angles determined from DLn samples and predicted from chemical shifts is reasonable for V24 and A30. Agreement for (325 and (329 is not quantitative, but predictions from chemical shifts for non-,B-strand glycines are considered unreliable. Intermolecular Distance Constraints. Recent measurements of inter- molecular 13C—13'C dipole-dipole couplings in A3140 fibrils (11, 12) indicate intermolecular distances of 4.8 : 0.5 A between backbone carbonyl carbons of V12, L17, F20, V24, L34, and V39, [ii-carbons of A21 and A30, and a-Cal‘bODS of (39. These distances imply an in-register parallel alignment in the cross—,8 motif, extending from G9 through V39. For A2 and F4, intermolecular distances are greater than 6 A, consistent with N-terminal disorder. Discussion Structural Model for A151-“ Fibrils. Fig. 4 presents a structural model for A8140 fibrils consistent with the conformational constraints and intermolecular distance constraints described above, and incorpo- rating the cross-B structural motif established by x—ray diffraction 16746 | WWW,pnaS.org/cgi/doi/10.1073/pnas.262663499 3 60A 40% b sea 86A Fig. 5. (a) Cross section of an A131-“ fibril with the minimal MPL indicated by scanning transmission electron microscopy (13, 29), formed by juxtaposing the hydrophobicfacesoftwo cross-13 units from Fig.4. Residues 1—8 areincluded with randomly assigned conformations. (b) Possible mode of lateral association to generate fibrils with greater MPL and greater cross-sectional dimensions. data (3, 27, 28). This model results from a constrained energy minimization procedure (see Materials and Methods). Significant features of the model are as follows: (1') residues 1—8 are omitted because of the N—terminal structural disorder; (if) the peptide conformation contains two ,B-strancls, separated by a 180° bend formed by residues 25—29; (iii) the ,G—strands form two in—register parallel ,B-sheets, which interact through sidechain-sidechain con- tacts; (iv) except for D23 and K28, sidechains in the core of the resulting double—layered structure (015, L17, F19, A21, 131, M35, and V39) are neutral and primarily hydrophobic; (v) sidechains of D23 and K28 form a salt bridge across the bend; (vi) sidechains of A30, 132, L34, V36, and V40 form a hydrophobic face; and (vii) other charged and polar sidechains are distributed on the opposite face, on the convex side of the bend, and in the N-terminal segment. The double—layered B-sheet structure in Fig. 4 a and b constitutes a single molecular layer, or “cross-,8 unit.” The MPL of a cross-[3 unit, given by the molecular mass divided by the 4.8 A spacing between hydrogen-bonded chains in a B-sheet, is 9.0 kDa/A. Measurements by scanning transmission electron microscopy indi- cate a minimal MPL equal to twice this value for A3140 fibrils (29), as well as for AB1_42 and AB]0_35 fibrils (13). Given these MPL data and the single hydrophobic face described above, fibrils with minimal MPL may be formed by juxtaposing two cross-13 units as in Fig. 5a. The structurally ordered part (3f the resglting fibril has cross-sectional dimensions of $40 A X 60 A, in good agreement with the dimensions of the narrowest fibrils in Fig. 1a. Note that residues 10—40 of ABHO would have a length of as100 A in a single IB—strand. Thus, the experimental fibril dimensions Petkova er al. i g i 3 46 3. i environment that may exist in the interior of an am loid fibril electrostatic re ulsrons Between like char es mi ht destabilize the ‘— i require non—,B—strand conformations and a large net bend angle between B-strands, consistent with the NMR data. Fibrils with minimal MPL, as in Fig. 5a, may be termed “pro- tofilaments.” Fibrils with greater MPL and greater diameters may be laterally associated protofilaments (30). One possible mode of lateral association is depicted in Fin. 5b. The spacings between B-sheet layers in Fig. 5 are 29.5 A, in good agreement with the equatorial spacing in experimental fiber diffraction data (3, 27, 28). Significance of Hydrophobic and Electrostatic Interactions. Amyloid fibrillization is generally considered to be driven b hydrophobic e an e ectrostatic interactions 34, 41—43). AB 1_40 contalns 1 centra residues 17—21) and C—terminal (residues 30—40) hydro- phobic segments. The in-re 'ster arallel ali nment of A1310_35, ABM“), and A8143 chains within the E-sheets in the fibrils, deter- mined experimenta y y so id state NMR (11—16), maximizes the hydrophobic contacts of these segments. However, an in—regtster r e rgnmen a so resu ts in s ort distances (#5 A) between like charges on neighboring peptide molecules. In the low dielectric para e -s eet structure b e=100 kcal mol overwhelmin the avorable hydro hobic energy. The ionization states of sidechains 1-40 1 rils may be determined experimentally from NMR chemical shifts. Cy shifts for D7 and D23 in Table 1 indicate deprotonated sidechains (44), whereas N5 shifts for K16 and K28 indicate protonated sidechains (45). Figs. 4 and 5 shgm: both how favorable h dro hobic interactions can be maximized and how electrostatic destabilization can be avoided. The only charged sidechains in the core of the structure in Fig. 4 are those of D23 and K28, which form a salt bridge that may stabilize the structure. All other charged or potentially charged sidcchains are at positions where they could be solvated as the fibrils grow. If protofilaments associate laterally as in Fig. 5b, intermo- lecular salt bridges between K16 and E22 would also prevent electrostatic destabilization. In support of D23-K28 salt bridges, we have detected a I5N—“C dipole-dipole coupling between Cy of D23 and NC of K28 in SUS AEMO fibril samples, corresponding to an interatomic distance |. Sipe, J. D. (1992)Arrrttt. Rev. Biochem. 61. 9477975. 2. Sunde. M.. Serpell. L. C.. Bartlnm. M., Fraser. 1’. 13., Pepys. M. 13. & Blake. C. C. F. (1997} J. Mol. Biol. 273, 7297739. 3. lnouye. H.. Fraser. P. E. & Kirschner. D. A. (I993)Biopltys.1. 64, 502619. 4. Glenner, G. G. & Wong, C. W. (1984) Biochrrrt. Biophys. Res. Commrrrr. 120, 885—890. 5. lwarsuho, T., Odaka, A.. Suzuki, N.. Mizusawa. 1-1.. Nukina, N. & lhara. Y. (1994) Neuron 13, 45—53. 6. George, A. R. & Hewlett. D. R. (1999) Biopoh’mcrs 50, 7337741. 7. Li. L. P“ Damien, T. A,, Bartolotti. L., Kominos, D. & Pedersen, L. G. (1999) Biophyr. J. 76, 287172878. 8. Tjernberg, L, 0.. Callaway, D. J. 13.. chrnbcrg, A., Hahne, 5., Lilliehook. C., Terenius, L., Thy/berg, J. S: Nordslcdl, C. (1999).l. Biol. Chem. 274. 12619—12625. 9. Chaney. M. D.. Webster. S. D.. Kuo. Y. M. & Roher. A. E. (1998) Protein Eng. 11, 761—767. 10. L320, N. D. & Downing, D. T. (1998) Biochemistry 37. 1731—1735. 11. Antztttkin. O. N.. Balbach, J. J., Leaprnan, R. D., Rizzo. N. W., Reed, J. & Tycko. R. {2000) Frat. Null. Atari. Sci. USA 97, 13045—13050. 1". Balhach,J.J., Fetkova, A. T.. Oylcr. N. A... Antzulkin. 0. N.. Gordon. 1). CL. Meredith. S. C. & Tycku. R. (2002) BiOpln'r. .l. 33, 1205-1216. 13. Anlzutkin, O. N.. Lcapman. R. D.. Balbnch. J. J. & Tycko. R. (2002) Biochemistry, 41, 15436—154511. 14. Benzinger. T. L. 5.. Gregory. D. M.. Burkuth, T. 5., Miller-Auer. H., Lynn. D. 6., Retro. R. E. tilt. Meredith. S. C. (1998) Proc. Natl. Atari. Sci. USA 95, 134117413412. 15. Burkolh. T. S..13enzinger. T. L. 5.. Urban. V.. Morgan, D. M., GregOry. D. M.. Thiyngarnjan. P.. Botto. R. 13.. Meredith, S. C. & Lynn, D. (3. (2000) J. Am. Chem. Soc. 122, 7883—7889. 16. Benzinger. T. L. 5.. Gregory. D. M., Burkorh, T. S., Miller-Auer. H.. Lynn. D. CL. Botto. R. E. & Meredilh, S. C. (2000) Biochemistry 39. 3491-3499. 17. 151111. Y. (2001).]. Chem: Phys. 114, 8473—8483. 18. Balhach, J. J., Ishii, Y., Antzulkin, O. N.. Leapruan. R. D.. Rizzo. N. W.. Dyda, 17.. Reed. J. 3: Tycko, R. (2000) Biochemistry 39, {3274843759. 19. Bttldus. M., Pelkova, A. T., Herzt'eld, J. & Griffin. R. G. (1993) Mal. Phys. 95, 1197—1207. 21). lshii. Y., Balbach, J. J. & Tycho. R. (2001) Chem. Phys. 266, 2317236. 21. Blanco. F. J. 3:: Tycho, R. (2001) J. Mag». Resort. 149. 131—138. 22. Fetkova, A. T. 8r. Tycho. R. (2002)}. Mag”. Resort. 155, 293L299. 23. Tycko. R., Weliky. D. P. ISL Berger, A. E. (1996) J. Chem. Phys. 105, 7915—7930. 24. Corniiescn. (3., Dclaglio, F. & Bax, A. (1999) J. Biorrtol. NMR 13. 2897302. 25. Herzfeld. J. & Berger. A. E. (1930) J. Chen-r. Phys. 73, 61121—61131). 26. Koradi, R., Billetcr. M. tit. Wiithrich. K. (1996)}. Mai. Graphics 14, 31755. 27. Malinchik. S. B..1nouyc. H, Szumowski, K. E. & KirSChner. D. A. (1995) Biopin's. J. 74.537445. Petkova et al. of a4 A, using the heteronuclear recoupling technique of Jaroniec er a1. (46). Comparison with other AB Fibril Models. Models for full—length AB fibrils suggested previously were based on antiparallel B—sheet structures (6—10) and are at variance with experimental intermo- lecular distance constraints (11—13). Several models include a true B-hairpin centered in residues 24w29, with intramolecular hydrogen bonding between ,B—strands on either side of a B-turn (6, 7, 10, 34). Such intramolecular hydrogen bonding is incompatible with the in-register parallel intermolecular alignment determined experi- mentally (11—13). A model proposed by Tjernberg er of. (8) includes a bend in residues 23—26 but is otherwise quite different from Figs. 4 and 5. Lynn, Meredith, Botto, and coworkers have proposed a model for AB1U_35 fibrils in which the peptide chain forms a single, continuous B—strand (15, 16). An alternative model for Afimss fibrils similar to that in Fig. 4 has been proposed independently by Ma and Nussinov (47), based on the solid state NMR data for ASN-3 fibrils (14—16), molecular modeiing, and dynamics simula- tions. The simulations of Ma and Nussinov, which include solvent molecules, indicate the stability of structures similar to that in Fig. 4 in an aqueous environment. Several groups (10, 48, 49) have suggested that certain amyloid fibrils have B-helical molecular structures, similar to the B-helices observed in proteins Such as R69 pertactin (50). The model in Fig. 4 resembles a B-helix in that C-termina] and N—terminal residues of separate B-strand segments are brought into proximity by bend segments with non—B-strand conformations (especially at glycine residues). In a B-helical version of this model, the C terminus of one ABHD chain would contact a residue near the N terminus of the next chain in the cross~B unit. Such contacts have not been established or ruled out experimentally. Finally, the models of the cross-B unit and the A3140 protofila— ment in Figs. 4 and 5 are lilter to be refined as new experimental data become available, such as additional constraints on backbone and sidechain torsion angles and additional constraints on intramo- lecular and intermolecular sidechain—sidechain contacts. We be” lieve that these models represent substantial progress toward full elucidation of the molecular structure of amyloid fibrils. 28. Fraser. P, E,, Nguyen. J. T.. Inouye. H., Surewicz, W. K.. Selkoe. D. J., Podlisny, M. B. 8L Kirschner. D. A. (1992) Biochemistry 31, 10715—10723. 29. Guldshury, C. S., Wirtz. 5.. Muller, S. A., Sunderji, 8., Wicki. P., Aebi, U. & Frey, P. (2000) J. Struct. Biol. 130, 217—231. 30. Jimenez, J. L., Nettleton, E. J., Bouchard. M., Robinson, C. V., Dobson, C. M. iii Saibil, H. R. (2002) Proc. Natl. Acatl. Sci. USA 99, 9196—9201. ' 1. Harper. J. D.. Lieber, C. M. 8L Lansbury, P. T. (1997) Chem. Biol. 4, 951—959. 2. Weliky, D. F., Bennett. A. E., Zvi, A.,Anglisler, J., Steinbach, P. 1.8: Tycko. R. (1999) Nari. Srrttct. Biol. 6, 141-445. 33. Long. H. W. & Tycko, R. (1998) J. Am. Cherrt. Soc. [20, 703977048. 34. Hilbich. C.. Kisterswoike, 3.. Reed. J., Masters, C. L. ilk Beyrcuthcr. K. (1991)J. Mol. Biol. 213, 149—163. 35. Rotter, A. B, Luwensun. J. D., Clarke, 5., Wolkow. C.. Wang, R.. Cotter. R. J., Reartlon. 1. M., Zurchemeely, H. A., Heinrikson, R. L., Ball. M. J. & Greenberg. B. D. “993”. Biol. Chem. 263, 31172—30153. 36. Saido, T. C.. Yamao-Harigaya, W..1\\alsubo. T. & Kawushima. S. (1996) Nam-05d. Lari. 215, 173—176. 37. Kheterpal, l., Williams. A., Murphy. C.. Bledsoe, B. & Welzel, R. (2001) Biochemistry 40, 11757-11767. 33. Saito, 1-1. (1986) Magrt. Resort. Client. 24. 8357852. 39. lwadale. M.. Asakura. T. 8:. Williamson. M. F. (1999) J. Biomol. NMR 13, 1997211. 40. de Dios. A. C.. Pearson. J. G. & Oldfield, E. (1993) Science 260. 1491—1496. 41. [-1ilbich. C.. Kislerswoike. 3,, Reed, J.. Masters, C. L. & Beyrenther, K. (1992)}. Mol. Biol. 228, 4607473. 42. Harper, .1. D.. Wong. 5. 5.. Lieber, C. M. St Lansbury. P. T. (1999) Biochemistry 38, 897278980. 43. Halverson, K.. Fraser. P. 13.. Kirschner. D. A. 8: Lnrtsbury, F. T. (1990) Biochemistry 29, 2639—2644. 44. Qin. J., Clore. G. M. & Gronenborn, A. M. (1996) Biochemistry 35, 7-13. 45. Zhu, L. Y., Kempie, M. D.. Yuan, P. it: Prendergast, F. G. (1995) Biodrenrisrnfid, 1319643202. 46. Jaroniec, C. P., Tounge. 13. A.. Herzfeld. J. 3r. Griffin, R. G. (2001).]. Am. Chem. Soc. 123, 3507-3519. 47. Ma, B. 8: Nussinov. R. (2002) Proc. Natl. Amrl. Sci. USA 99, 14126—14131. 48. Perutz, M. F., Finch, J. T., Ben-imam. J. & Lcsk, A. (2002) Pros. Natl. Acad. Sci. USA 99, 5591—5595. 4‘). Wills, 1-1., Michelitsch. M. D.. Guenebaut, V., Supattapone. 5.. Serban, A.. Cohen. F. 13., Agard, D. A. (it Prusiner. S. B. (2002) Pros. Nari. Acarl. Sci. USA 99, 356373568. 51). Emslcy. P., Charles, I. G.. Fairwettlher, N. F. 8t. lsaacs, N. W. (1996) Nature 331. 90—92. 51. Wishart. D. S.. Bigam. C. (3., Hum, A., Hodges, R. S. Sr. Sykes, B. D. (1995) J'. Biomol. NMR 5, 67781. or m PNAS f December24.2002 } vol. 99 i no. 25 | 16747 ...
View Full Document

This note was uploaded on 02/20/2012 for the course BME 201 taught by Professor Schmidt during the Fall '11 term at Purdue University-West Lafayette.

Page1 / 20

BME 201 Practice Exam 2 - i 9 E E e 2 2' E S i E E i i BME...

This preview shows document pages 1 - 20. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online