We have shown that with minor modification, this model also accounts well for the kinetics and thermodynamics of inhibitor binding at the Q_B-site. We use the same model as the basis for our analysis of kinetic and thermodynamic parameters determining the characteristics of electron transfer in the two-electron gate of susceptible and resistant A. hybridis biotypes.

Fig. 1.

       We assume that the biphasic kinetics in the 10 ms range reflect electron transfer in two distinct populations of photosystem II. Reaction centers after long dark adaption have either a secondary quinone bound at the Q_B-site, the Q_B state (Q_AQ_B) (blue), or a vacant Q_B-site (Q_A.vacant) (yellow), as shown in Fig. 1. An actinic flash produces a singly reduced primary quinone, to give the species Q_A^-.vacant (yellow -> green path), or Q_A^-Q_B (blue -> green path). Both Q_A^-.vacant and Q_A^-Q_B are closed centers, and give rise to a high fluorescence yield. At 50 µs after the flash, when reactions on the donor side have reached equilibrium, the high fluorescence yield reflects the electrons remaining on Q_A^-. In centers in the Q_B state, the electron is transferred to Q_B in a first-order process (blue -> green path), generating Q_AQ_B^-, and giving rise to a low fluorescence, with half time of approximately 150 to 300 µs depending on biotype. Since the electron on Q_A can be transferred to Q_B only when a plastoquinone is present at the Q_B-site, electron transfer in centers initially with a vacant site must first involve a binding of quinone in a second-order process, and then transfer of the electron (yellow -> green path); one therefore expects for this fraction of centers a slower kinetic, reflecting the convolution of the binding and electron transfer reactions. The biphasic decay kinetics following an actinic flash in the range out to 10 ms can be interpreted as reflecting these two populations. Both phases are well fitted by exponential decay curves. In our experience, the more rapid phase accounts for more than half those centers which transfer an electron within 10 ms. Since the quinone pool is oxidized under our conditions, the minor fraction of initially vacant centers will become occupied from a pool of plastoquinone in > 10-fold excess. This pool will therefore undergo a negligible change in concentration during the reaction, and the process will follow pseudo-first-order kinetics.

       At neutral pH, the proton binding reactions follow the green path, and at low pH (below pK₁) the pink -> green path. As the pH is raised above pK₂, the rate slows, suggesting that under these conditions the reactions follow the gray -> green path.

       In principle, one can calculate the rate constants for electron transfer and binding reactions directly from the kinetic curves under specific initial conditions, using appropriate rate equations. A simpler alternative makes use of the above model, and the following explicit assumptions:

• (i) The fast phase of Q_A oxidation following an actinic flash, is due to reduction of Q_B in centers with bound Q_B by a first-order reaction, and the slow phase to reduction in center where Q has to first bind from the pool. The ratio of the amplitude of the fast component (B_o) to that of the slow component (A_o) in the decay kinetics is equal to the ratio of the initial fraction of reaction centers in the Q_B state over that in the vacant state.
• (ii) Under our experimental conditions, the binding of a plastoquinone to the vacant site is a pseudo-first order reaction, because binding does not deplete the oxidized pool significantly.
• (iii) The dissociation constant of a plastoquinone from its binding site following an actinic flash (from Q_A^-Q_B) is the same as that in the dark (from Q_AQ_B).

where species in square brackets represent fractions of the reaction centers in a particular state (including protonated species) as indicated in the bracket. Q_A and Q_A^- represent centers with a vacant Q_B-site, with the oxidized and the reduced Q_A respectively. Rate constants are as defined in Table 1, except that in these equations, no account is taken of protonation state, and values will therefore be constant only at a defined pH. Variables r1 and r2 are the apparent rate constants which appear in the solution after solving the rate equations, and their values are obtained directly by applying a non-linear least-squares fitting procedure to the kinetic curves, after the latter have been corrected for the non-linear relation between fluorescence yield and concentration of Q_A^-. A_o and B_o are the amplitudes of the slow and fast components, respectively, of the initial biphasic decay kinetics of Q_A^- following an actinic flash, again obtained directly from kinetic analysis of the data.

       The four rate constants k_AV, k_VA, k_AB and k_BA can be calculated at any pH through eqs. 1) - 4) from the four measured parameters, e.g. the fast and slow apparent rate constants, the fraction of vacant (or Q_B^-bound) reaction centers (both determined from the corrected decay kinetics by a non-linear least-squares fitting procedure), and the measurement of the apparent equilibrium constant, K_app. K_app is measured from the ratio of the back-reactions in the absence (S₂⁺Q_B^-) or presence of DCMU (S₂⁺Q_A^-).

       The variability of values for rate and equilibrium constants with pH, and the relation to the dependence of K_app on pH, is given by two different effects:

• The stabilization of the semiquinone by protonation of a residue close to the Q_B pocket. We showed that K_app, the apparent equilibrium constant for electron transfer after a single flash, showed pH dependence over the range between ~6.0 and ~8.0, and that above and below this range the value was independent of pH. We ascribed this to a coulombic effect of the charge on the semiquinone inducing a pK shift (from ~6.2 (pK₁) to ~8.1 (pK₃ in Fig. 1)), and suggested from analogy with bacterial reaction centers, and sequence alignment that the residue affected was likely His-252 of D1.
  
  Fig. 2.

• The second pH dependent effect was on the value for K_o, interpreted acording to the following scheme:

  Eq. (5)

  K₁d and K₂d (not shown in Fig. 1)are the acid dissociation constants of Q_AQ_B(H⁺) and Q_A.vacant(H⁺), respectively, giving pK₁ and pK₂. We have assumed above that K₁d and K₂d are the same before and after an actinic flash (and equal to K₁ and K₂ respectively); if this is the case, eq. (5) gives A_o/B_o in eq. (3). The factor [Q_A]pK /[Q_AQ_B]pK gives K_o,pK, the dissociation constant above pK₂.

         The behavior over the pH range 5 - 8.5 was different in an atrazine resistant strain, and could be accounted for by changes in the pK values above. For the efect on K_E, pK₃ appeared to be changed from ~8.1 in the susceptible to ~7.0 in the resistant biotype. For K_o, _E could be fitted to eq. (5) with K_o,pK = 0.12, and two pKs, e.g. pK₁ = 6.2 and pK₂ = 6.9 in the susceptible biotype; on the other hand, the data for the resistant strain show no pH dependence, giving K_o,pK = 0.58, and two pKs of the same value, e.g. pK₁ = pK₂. The results show that in the physiological range, plastoquinone binds more tightly at the Q_B-site in the susceptible than in the resistant strain.

Table 1.

___________________________________________________________________________
Constant    Reaction        			susceptible	resistant  	notes
___________________________________________________________________________
A.  pK values       								(1)
pK1   QAQB (H+)	--> 	QAQB + H+ 		6.2       	6.2 
pK2   QA.vac (H+)	--> 	QA.vac + H+	6.9       	6.2 
pK3   QAQB- (H+)	--> 	QAQB- + H+ 	8.1       	7.0 
___________________________________________________________________________
B.  Equilibrium (K) or rate (k) constants.
Ko,pK  QA-QB	--> 	QA- (+ Qpool)		0.12 ± 0.02 	0.57 ± 0.06 	(2,3)
KE   	QA-QB 	--> 	QAQB-       		3.1 ± 0.7      	4.7 ± 0.8
____________________________________________________________________________
kVA  QA- (+ Qpool)--> 	QA-QB        		500      	650   		(2,4)
k'VA  QA-(H+) (+ Qpool)--> QA-QB(H+)    	500      	650   		(2,4)
kAV  QA-QB	 --> 	QA- (+ Qpool)		 60      	370   		(2,4)
k'AV  QA-QB(H+) --> 	QA-(H+)(+ Qpool)	300     	3,370   	(2,4)
kAB   QA-QB	--> 	QAQB-         		3,000    	10,000   	(5,6)
k'AB   QA-QB(H+) 	--> 	QAQB-(H+)      	3,000    	10,000   	(5,6)        
kBA   QAQB- 	--> 	QA-QB         		987     	2,200   	(5,6)
k'BA    QAQB-(H+) --> 	QA-QB(H+)      		12      	340   		(5,6)
__________________________________________________________________________
Notes.    
(1) It is assumed that values for pK1 and pK2, and the fraction of vacant reaction centers in the dark (determined by Ko), are the same after dark adaptation as those following an actinic flash. (2)  Although these reactions involve the quinone pool, constants shown are based on the pseudo-first-order process, with [Qpool] assumed to be 1. 
(3) The true value can be obtained from Ko = K'o / [PQ]pool; we have previously estimated [PQ]pool = 5 mM under oxidizing conditions [32].
(4)  Rate constants kIJ are for reaction at pH above pK2. Rate constants k'IJ are for reaction at pH below pK1. 
(5)  Rate constants kIJ are for reaction at pH above pK3. Rate constants k'IJ are for reaction at pH below pK1. 
(6) The difference between rate constants k and k' does not reflect a change in the rate constant for the electron transfer process (since KE is independent of pH), but a change in concentration of substrate for the backreaction, [QAQB-], associated with pK3.

Model of the Q_B-site. Stereo view, showing sites of mutations.

Table 2.

Parameters of the two-electron gate in wild type and herbicide-resistant mutants of Chlamydomonas reinhardtii. t_½(Q_A^-) is the half time of the back reaction from Q_A^- to S₂ in the presence of DCMU. t_½(Q_B^-) is the half time of the back reaction from Q_B^- to S₂, measured as the rephasing time of the binary oscillations of the two-electron gate. K_o, the dissociation constant for plastoquinone from the Q_B site, is given without units, as a relative value for comparison between wild type and mutants.

strain 	 t½	 t½  	Ko       Kapp	  KE	  kAB	  kBA	  kAV	  kVA
	(QA-)	(QB-)	(*)
	 (sec)	(sec)				(ms-1)	(ms-1)	(ms-1)	(ms-1)
w. type	2.7	25	0.40	8.3	11.6	4.29	0.37	0.34	0.86
S264A	2.4	20	2.50	7.3	25.6   15.0	0.56	0.60	0.24
G256D	2.7	25	0.67	8.3	13.8	4.66	0.34	0.47	0.70
V219I	2.0	40	0.32   19.0	25.1	6.25	0.25	0.19	0.60
A251V**	3.0	15	2.47   	4.0	13.9	n.d.	n.d.	n.d.	n.d.
F255Y	2.5	25	0.74	9.0	15.7	3.73	0.24	0.84	1.07
L275F	2.7	20	0.46	6.4	 9.36	3.46	0.37	0.39	0.84

* Ko = Ao / Bo = kAV / kVA

** Assuming 2 components to decay.