**7. Discussion**

The purpose of the present study was to elucidate further the interaction of CP43 with the extrinsic cytochrome c550 protein of PSII by studying the effect of additional alterations of the arginine residue at position 320 of the CP43 large extrinsic loop. Two additional mutations, R320K and R320D, were introduced into the large extrinsic loop and the effects of

Mutations in the CP43 Protein of Photosystem II

under chloride-limiting conditions.

this site is involved.

Affect PSII Function and Cytochrome C550 Binding 71

is not the case, particularly under chloride-limiting conditions. The susceptibility of the PSII centers to photoinactivation was assessed by exposing the cells to a high light intensity. The process of photoinactivation leads to a number of structural changes within the PSII complex. These events include the rapid turnover of the D1 protein and proteolysis of the D2 protein and CP43 (Nedbal et al., 1990; Mori and Yamamoto, 1992; Aro et al., 1993), with a resultant drop in oxygen-evolving activity. The results for the control strain (Table 2) showed no appreciable differences between the t0.5 values for samples grown in complete or chloride-deficient media. When grown in complete media and subjected to photoinactivating light, the R320K strain exhibited a t0.5 value similar to the control value. The calculated t0.5 for photoinactivation for the R320D mutant was 1.4 times less than that of control cells, indicating that this mutant is more sensitive to photoinactivation than the control. This suggests that, in complete media, there is a defect in the stability of the R320D PSII complex, while no defect is apparent for the R320K strain under these conditions. Photoinactivation experiments were also performed on strains grown under conditions of chloride depletion (Table 2). The results of these experiments indicate that, while t0.5 value for the control strain remained essentially the same compared to its value in complete media, the PSII stability of both mutant strains was altered under these conditions. For the R320K strain grown in chloride-deficient media, the t0.5 value was 2.9 min compared to 5.7 min for the control, indicating that this mutant is photoinactivating at twice the rate of the control under these conditions. Photoinactivation was difficult to measure in the R320D mutant due to its low rate of oxygen evolution. Given this, R320D cells demonstrated an extreme sensitivity to photoinhibition, with a calculated t0.5 for photoinactivation 6 times less than that of the control. Thus, the PSII centers in all the mutants become destabilized

Comparison of the results of this photoinactivation study to those obtained for the R320S mutant (Table 2; Young et al., 2002) highlights some notable trends. The PSII stability of these strains seems to follow the following pattern in complete media: Control ≈ R320K > R320S > R320D. This trend is only slightly modified when the strains are grown in chloridedeficient media, and follows the pattern: Control > R320K ≈ R320S > R320D. When chloride is not limiting in the media, it appears that various amino acid substitutions can be made at this site with limited effect on the stability of PSII. However, the results also indicate that when chloride is limiting, even substitution of an amino acid with similar properties to arginine, such as lysine, is not sufficient to preserve the integrity of the interactions in which

In order to gain a better understanding of the effect of the mutations on possible interactions with cytochrome c550, PSII core particles were isolated from His-tagged control and mutant strains using cobalt metal affinity chromatography. The effectiveness of this method for isolating active PSII particles has been documented (Bricker et al., 1998). Fig. 4 shows the optical absorption spectra obtained from the PSII particles derived from the control and two mutant strains. There are no apparent differences in the spectra, each of which shows a characteristic absorption maximum at 674 nm. Reduced minus oxidized difference spectra were obtained from the isolated control, R320K, and R320D His-tagged particles (Fig. 5). These indicated that while the control particles were able to retain cytochrome c550 within the particle, the mutant strains were not (Fig. 5). This was true whether the cells were grown in complete (Fig. 5) or chloride-deficient media (not shown). These results are similar to those obtained previously with the R320S mutant, which was unable to retain the

each on PSII activity were characterized. First, the growth characteristics of the mutant strains were assessed. Under photoautotrophic growth conditions (Fig. 2A), when chloride levels in the media were not limiting, the control and mutant strains grew at the same rate. This indicates that under these growth conditions, there appear to be no discernible anomalies associated with PSII. When grown in chloride-deficient media, however, major differences arose. Under these conditions, the control strain grew at rates comparable to those in complete media, but R320D exhibited very little capacity for photoautotrophic growth (Fig. 2B). The R320K strain exhibited intermediate growth between that of the control and R320D. When grown in chloride-deficient media that had been supplemented with bromide, all strains grew at very similar rates, with the rates of the mutant strains being restored to those observed in complete media (Fig. 2C). As stated previously, bromide is well known to substitute for chloride at the active site of PSII, and these data clearly indicate that the effect on growth arises as a consequence of a defect associated with the chloride requirement of PSII.

Oxygen evolution rates were obtained for the control and mutant strains (Fig. 3). When grown photoautotrophically, the control strain evolves oxygen at similar rates, regardless of the media in which the cells are grown. The R320K strain exhibits a modest reduction in oxygen evolution rate when grown in complete media, which is greatly magnified when chloride is limiting in the media. Under these conditions, the R320K strain appears to evolve oxygen at less than 40% of the control strain. This rate is restored when bromide is included in the chloride-deficient media. The R320D strain reveals the most dramatic phenotype. When grown in complete media or in chloride-deficient media supplemented with bromide, this strain evolves oxygen at a rate of approximately 60% of the control strain. When chloride is limiting, however, this rate drops to only about 5% of that of the control strain. Taken together, these results indicate that there is a defect in both mutant strains in their ability to evolve oxygen, regardless of the media in which they are grown. This defect is exacerbated by a deficiency of chloride in the media, which again points to a physical and/or mechanistic defect at the PSII active site involving some interaction with chloride.

The data from growth experiments and oxygen evolution assays indicated that both mutant strains, which are relatively functional under normal growth conditions, must assemble a fairly normal content of PSII centers. Confirmation of this was obtained from variable fluorescence yield assays. These assays allow for a qualitative estimate of the number of PSII centers present when using hydroxylamine as an artificial electron donor to measure the reduction of QA (Nixon & Diner, 1992). The results presented in Table 1 show that the mutant strains assemble a near normal complement of PSII centers within their thylakoid membranes compared to the control strain. Thus, the mutations do not prevent the assembly and incorporation of PSII centers into the thylakoids of these strains.

When water is used as the electron donor, the variable fluorescence yield assay provides information on the functionality of the PSII centers present within the membrane, with respect to their ability to conduct electron transport from water to QA. As shown in Table 1, the variable fluorescence yields for both mutant strains approximated those of the control strain, indicating that the mutant PSII centers are fully capable of oxidizing water. These results mirrored those previously obtained for the R320S strain (Table 1; Young, et al., 2002). It should be noted that the *psbC* deletion strain, which contains no functional PSII centers, did not give a fluorescence signal.

Although the R320K and R320D strains appear to assemble near control amounts of PSII centers that one might assume from the above data are functioning relatively normally, this

each on PSII activity were characterized. First, the growth characteristics of the mutant strains were assessed. Under photoautotrophic growth conditions (Fig. 2A), when chloride levels in the media were not limiting, the control and mutant strains grew at the same rate. This indicates that under these growth conditions, there appear to be no discernible anomalies associated with PSII. When grown in chloride-deficient media, however, major differences arose. Under these conditions, the control strain grew at rates comparable to those in complete media, but R320D exhibited very little capacity for photoautotrophic growth (Fig. 2B). The R320K strain exhibited intermediate growth between that of the control and R320D. When grown in chloride-deficient media that had been supplemented with bromide, all strains grew at very similar rates, with the rates of the mutant strains being restored to those observed in complete media (Fig. 2C). As stated previously, bromide is well known to substitute for chloride at the active site of PSII, and these data clearly indicate that the effect on growth arises as a consequence of a defect associated with the

Oxygen evolution rates were obtained for the control and mutant strains (Fig. 3). When grown photoautotrophically, the control strain evolves oxygen at similar rates, regardless of the media in which the cells are grown. The R320K strain exhibits a modest reduction in oxygen evolution rate when grown in complete media, which is greatly magnified when chloride is limiting in the media. Under these conditions, the R320K strain appears to evolve oxygen at less than 40% of the control strain. This rate is restored when bromide is included in the chloride-deficient media. The R320D strain reveals the most dramatic phenotype. When grown in complete media or in chloride-deficient media supplemented with bromide, this strain evolves oxygen at a rate of approximately 60% of the control strain. When chloride is limiting, however, this rate drops to only about 5% of that of the control strain. Taken together, these results indicate that there is a defect in both mutant strains in their ability to evolve oxygen, regardless of the media in which they are grown. This defect is exacerbated by a deficiency of chloride in the media, which again points to a physical and/or mechanistic defect at the PSII active site involving some interaction with chloride. The data from growth experiments and oxygen evolution assays indicated that both mutant strains, which are relatively functional under normal growth conditions, must assemble a fairly normal content of PSII centers. Confirmation of this was obtained from variable fluorescence yield assays. These assays allow for a qualitative estimate of the number of PSII centers present when using hydroxylamine as an artificial electron donor to measure the reduction of QA (Nixon & Diner, 1992). The results presented in Table 1 show that the mutant strains assemble a near normal complement of PSII centers within their thylakoid membranes compared to the control strain. Thus, the mutations do not prevent the assembly

and incorporation of PSII centers into the thylakoids of these strains.

When water is used as the electron donor, the variable fluorescence yield assay provides information on the functionality of the PSII centers present within the membrane, with respect to their ability to conduct electron transport from water to QA. As shown in Table 1, the variable fluorescence yields for both mutant strains approximated those of the control strain, indicating that the mutant PSII centers are fully capable of oxidizing water. These results mirrored those previously obtained for the R320S strain (Table 1; Young, et al., 2002). It should be noted that the *psbC* deletion strain, which contains no functional PSII centers,

Although the R320K and R320D strains appear to assemble near control amounts of PSII centers that one might assume from the above data are functioning relatively normally, this

chloride requirement of PSII.

did not give a fluorescence signal.

is not the case, particularly under chloride-limiting conditions. The susceptibility of the PSII centers to photoinactivation was assessed by exposing the cells to a high light intensity. The process of photoinactivation leads to a number of structural changes within the PSII complex. These events include the rapid turnover of the D1 protein and proteolysis of the D2 protein and CP43 (Nedbal et al., 1990; Mori and Yamamoto, 1992; Aro et al., 1993), with a resultant drop in oxygen-evolving activity. The results for the control strain (Table 2) showed no appreciable differences between the t0.5 values for samples grown in complete or chloride-deficient media. When grown in complete media and subjected to photoinactivating light, the R320K strain exhibited a t0.5 value similar to the control value. The calculated t0.5 for photoinactivation for the R320D mutant was 1.4 times less than that of control cells, indicating that this mutant is more sensitive to photoinactivation than the control. This suggests that, in complete media, there is a defect in the stability of the R320D PSII complex, while no defect is apparent for the R320K strain under these conditions.

Photoinactivation experiments were also performed on strains grown under conditions of chloride depletion (Table 2). The results of these experiments indicate that, while t0.5 value for the control strain remained essentially the same compared to its value in complete media, the PSII stability of both mutant strains was altered under these conditions. For the R320K strain grown in chloride-deficient media, the t0.5 value was 2.9 min compared to 5.7 min for the control, indicating that this mutant is photoinactivating at twice the rate of the control under these conditions. Photoinactivation was difficult to measure in the R320D mutant due to its low rate of oxygen evolution. Given this, R320D cells demonstrated an extreme sensitivity to photoinhibition, with a calculated t0.5 for photoinactivation 6 times less than that of the control. Thus, the PSII centers in all the mutants become destabilized under chloride-limiting conditions.

Comparison of the results of this photoinactivation study to those obtained for the R320S mutant (Table 2; Young et al., 2002) highlights some notable trends. The PSII stability of these strains seems to follow the following pattern in complete media: Control ≈ R320K > R320S > R320D. This trend is only slightly modified when the strains are grown in chloridedeficient media, and follows the pattern: Control > R320K ≈ R320S > R320D. When chloride is not limiting in the media, it appears that various amino acid substitutions can be made at this site with limited effect on the stability of PSII. However, the results also indicate that when chloride is limiting, even substitution of an amino acid with similar properties to arginine, such as lysine, is not sufficient to preserve the integrity of the interactions in which this site is involved.

In order to gain a better understanding of the effect of the mutations on possible interactions with cytochrome c550, PSII core particles were isolated from His-tagged control and mutant strains using cobalt metal affinity chromatography. The effectiveness of this method for isolating active PSII particles has been documented (Bricker et al., 1998). Fig. 4 shows the optical absorption spectra obtained from the PSII particles derived from the control and two mutant strains. There are no apparent differences in the spectra, each of which shows a characteristic absorption maximum at 674 nm. Reduced minus oxidized difference spectra were obtained from the isolated control, R320K, and R320D His-tagged particles (Fig. 5). These indicated that while the control particles were able to retain cytochrome c550 within the particle, the mutant strains were not (Fig. 5). This was true whether the cells were grown in complete (Fig. 5) or chloride-deficient media (not shown). These results are similar to those obtained previously with the R320S mutant, which was unable to retain the

Mutations in the CP43 Protein of Photosystem II

*Photosynthesis. Research,* Vol. 24, pp. 1-13.

protein. *Biochemistry*, Vol. 31, pp. 4623-4628.

1143, pp. 113-134.

Affect PSII Function and Cytochrome C550 Binding 73

Aro, E.-M., Virgin, I., & Andersson, B. (1993) Photoinhibition of photosystem II.

Bricker, T. (1990) The structure and function of CPa-1 and CPa-2 in photosystem II.

Bricker, T. (1992) Oxygen evolution in the absence of the 33 kDa manganese-stabilizing

Bricker, T., Morvant, J., Masri, N., Sutton, H., & Frankel, L. (1998) Isolation of a highly active

Bricker, T., A. Young, L.K. Frankel, and C. Putnam-Evans. (2002) Introduction of the

Bricker, T. & Frankel, L. (2002) The structure and function of CP47 and CP43 in

Bricker, T., Roose, J., Fagerlund, R., Frankel, L., & Eaton-Rye, J. (2011) The extrinsinc

Burnap, R., Qian, M., Shen, J.-R., Inoue, Y., & Sherman, L. (1994) Role of disulfide linkage

Burnap, R., Shen, J.-R., Jursinic, P., Inoue, Y., & Sherman, L. (1992) Oxygen yield and

stabilizing protein of Photosystem II. *Biochemistry*, Vol. 31, pp. 7404–7410. Carpenter, S., Charite, J., Eggars, B., & Vermaas, W. (1990) The psbC start codon in

Chu, H.-A., Nguyen, A. & Debus, R. (1994) Site-directed photosystem II mutants with

Clarke, S., & Eaton-Rye, J. (1999) Mutation of Phe-363 in the Photosystem II protein CP 47

Dau, H., & Haumann, M. (2008) The manganese complex of photosystem II in its

Debus, R., Barry, B., Sithole, I., Babcock, G., & McIntosh, L. (1988a) Directed mutagenesis

Debus, R., Barry, B., Babcock, G., & McIntosh, L. (1988b) Site-specific mutagenesis identifies

*Proceedings of the National Academy of Science* (USA), Vol. 85, pp. 427- 430.

mutant of CP 47. *Biochimica et Biophysica Acta*, Vol. 1409, pp. 50-7.

photosystem II. *Biochimica et Biophysica Acta,* Vol. 1556, pp. 92-96.

Photosystem II. *Photosynthesis Research*, Vol. 72, pp. 131-146.

doi:10.1016/j.bbabio.2011.07.006 [Epub ahead of print].

*Synechocystis* sp. PCC 6803. *FEBS Letters*, Vol. 1, pp. 135-137.

manganese cluster in vivo. *Biochemistry*, Vol. 33, pp. 6137-6149.

*Biochemistry,* Vol. *33*, pp. 13712-13718.

6803. *Biochemistry*, Vol. 38, pp. 2707-2715.

*Coordination Chemistry Reviews*, Vol. 252, pp. 273–295.

polypeptide. *Biochemistry*, Vol. 27, pp. 907-9074.

Inactivation, protein damage and turnover. *Biochimica et Biophysica Acta*, Vol.

photosystem II preparation from Synechocystis 6803 using a histidine-tagged

305Arg 305Ser mutation in the large extrinsic loop E of the CP43 protein of Synechocystis sp. PCC 6803 leads to the loss of cytochrome c550 binding to

proteins of photosystem II. *Biochimica et Biophysica Acta*,

and putative intermolecular binding residues in the stability and binding of the extrinsic manganese-stabilizing protein to the photosystem II reaction center.

thermoluminescence characteristics of a cyanobacterium lacking the manganese-

perturbed oxygen-evolving properties. 1. Instability or inefficient assembly of the

impairs photoautotrophic growth, alters the chloride requirement, and prevents photosynthesis in the absence of either PSII-O or PSII-V in *Synechocystis* sp. PCC

reactioncycle—basic framework and possible realization at the atomic level.

indicates that the donor to P680+ in photosystem II is tyr-161 of the D1

a tyrosine radical involved in the photosynthetic oxygen-evolving complex.

association between PSII and cytochrome c550 after purification of these particles, regardless of the media in which the cells were grown. Thus, even a conservative substitution (lysine) at this site, which results in a fairly functional mutant under normal growth conditions, prevents the strong association of cytochrome c550. It is possible that some cytochrome c550 associates with these mutants in vivo, but that the relatively harsh conditions used to isolate the particles disrupt this interaction. We are currently probing the kinetics of binding of cytochrome c550 to His-tagged PSII from these strains in order to examine this.
