AMPK Methods and Protocols

transferred into the z-axis (alongB 0 ) with varying mixing times

(Tmix) at which point interconversion between states is allowed to

occur. Two auto-peaks are observed for statesAandB, and two

cross peaks are observed for the exchange between states (A!B

andB!A) (Fig.2). The integral of each peak can be taken and fit

to a binding model in order to obtain kinetic rates [5, 7]:

IAA IBA

IAB IBB



¼

1 yBA

yAB yBB



∙exp Tmix

k^0 onR 1 A koff

k^0 on koffR 1 B

!"#

∙

I 0 A 0

0 I 0 B

"#

ð 9 Þ

k^0 on¼kon½Š¼X koff

pB

1 pB

ð 10 Þ

whereIAAandIBBare the measured peak volumes of the auto-

peaks,IABandIBAof the cross peaks, andI 0 AandI 0 Bin the absence

of mixing,R 1 AandR 1 Bare the^15 N longitudinal relaxation times, and

yAB,yBA, andyBBaccount for differential relaxation after the mixing

time.

CPMG relaxation dispersion experiments contain a spin-echo

pulse that has two fixed time delay periods (τ) with a 180pulse

in-between (τ-180-τ) and is applied as a repeated train of pulses

(τ-180-τ)nto refocus transverse magnetization [11]. In a typical

experiment, a series of 2D spectra are obtained with differing

length of spin-echo pulses which are conveniently denoted as a

frequency (νCPMG¼ 1/4τ). AsνCPMG frequency increases, the

dephasing of spins is decreased, which leads to increased intensity

and reduced relaxation rates (R 2 obs). Usually a series of 2D spectra

are recorded with varying νCPMG frequency (between 50 and

Fig. 2Schematic of ZZ-exchange NMR spectroscopy—In an exchange spectroscopy experiment, two auto-
peaks are observed (AAandBB) corresponding to theAandBstates, and two exchange cross peaks are
observed (A!BandB!A). NoteAAandBBare resolved in both dimensions of the 2D spectrum. The
intensity of these peaks can be fit to a binding model to obtain rate constants

Carbohydrate Binding Kinetics of theβ-Subunit CBM 91