Cadmium inhibitory action leads to changes in structure of ferredoxin:NADP+ oxidoreductase

This study deals with the influence of cadmium on the structure and function of ferredoxin:NADP+ oxidoreductase (FNR), one of the key photosynthetic enzymes. We describe changes in the secondary and tertiary structure of the enzyme upon the action of metal ions using circular dichroism measurements, Fourier transform infrared spectroscopy and fluorometry, both steady-state and time resolved. The decrease in FNR activity corresponds to a gentle unfolding of the protein, caused mostly by a nonspecific binding of metal ions to multiple sites all over the enzyme molecule. The final inhibition event is most probably related to a bond created between cadmium and cysteine in close proximity to the FNR active center. As a result, the flavin cofactor is released. The cadmium effect is compared to changes related to ionic strength and other ions known to interact with cysteine. The complete molecular mechanism of FNR inhibition by heavy metals is discussed. Electronic supplementary material The online version of this article (doi:10.1007/s10867-012-9262-z) contains supplementary material, which is available to authorized users.

Circular dichroism measurement is the technique which is usually applied with success for determination of secondary structure of proteins. However, in certain conditions a good quality spectrum, allowing precise analysis, is impossible. This is in our conditions, when phosphate buffer cannot be applied (as forming unsoluble cadmium phosphate). Application of cadmium-safe buffers, as Tris or Hepes, reduces a reliable range of measurement in the UV region up to [200][201][202][203][204][205] nm. For such range it is almost impossible to get reliable fit of secondary structure content. For accurate determination a positive dichroism peak of helix (190 nm), betasheet (maximum around 195 nm), but also negative dichroism peak of unstructured peptide should be measured. Without these details, fitting base on negative maxima at 209 and 222 nm (-helix), 215 nm (-sheet) and very week positive dichroism of unstructured chain is with very high error.
However, comparison of the spectra gives some indication if the secondary structure is impacted at all by cadmium treatment, both at pH 8.7 and 7.0 (Fig.S1). The observed change is not the effect of ionic strength increase, as confirmed by measurement with respective concentration of NaCl (Fig.S1).

S.2. Membrane-caused alteration in FNR inhibition by cadmium
presence of cadmium from the control spectrum. Arrows indicate the changes of the greatest magnitude As it has been already shown (Grzyb et al. 2008) FNR could bind to model membrane and such interaction influences significantly FNR structure. Such interaction is a probable model of one of FNR states in vivo. This is why we also attempted to check how the cadmium influence is impacted by presence of model membranes.
In this part of our research, first we checked if cadmium changed the binding of FNR to the membrane. In the experiment FNR was injected beneath monolayer (the Langmuirr-Bloggett type) of DGDG, and the significant increase in surface pressure was observed ( Fig. S2, solid line, 3 min), as it has been shown already (Grzyb et al. 2008).
CdCl 2 was introduced beneath the monolayer with FNR, after 50 min, when surface pressure stabilized. The introduced solution caused few minutes destabilization followed by 10 min increase in and stabilization. When CdCl 2 was injected beneath DGDG monolayer (Fig. S2, dashed line, 3 min), the surface pressure slowly decreased. Such a decrease might be an effect of interaction between cadmium and the hydrophilic lipid heads, increasing ordering. 30 min after CdCl 2 , the FNR solution was introduced beneath that monolayer, and caused the increase in surface pressure comparable to changes observed without cadmium in the solution beneath monolayer. together, allowed to conclude that cadmium presence increases slightly possibility of FNR attachment to membrane, most probably by changing the membrane propertiesincreasing amount of places available for protein to incorporate.
The incorporation of FNR into membrane cause increase of activity (as it has been also shown elsewhere (Grzyb et al. 2008) of about 25 and 29% (for MGDG-DGDG or DGDG liposomes, respectively, Fig.S3). Such activated enzyme is also more sensitive for inhibition -cadmium reduced the FNR activity to a greater extent than without liposomes. For the conditions applied, after incubation with CdCl 2 only, the enzyme activity was reduced by 57%, while in the presence of both CdCl 2 and DGDG (or MGDG-DGDG) liposomes the activity was reduced by 78% and 76%, respectively. This might suggest that liposomes did not compete with cadmium for binding sites, but stabilize the FNR structure allowing metal ions to penetrate deeper parts of the enzyme.
That is of the great importance for extrapolation of our findings to in vivo situation, as most probably much lower concentrations of cadmium will be necessary to cause inhibitory effect.

S.4. Fluorescence
The figure S4 illustrates changes in tryptophane steady-state fluorescence, mentioned in main part of paper. Although greater effect of cadmium is observed in pH 8.7 than in pH 8.7, the Hepes buffering range does not cover such high pH. On the other hand, at pH 8.7 the effect of cadmium is still significant (Grzyb et al. 2010 introduced beneath monolayer by injection of small volume of water stock solution (50 mM), when the surface pressure stabilized after injection of FNR. In second variant of experiment, cadmium chloride was injected first, and FNR was introduced beneath monolayer after stabilization of surface pressure. The changes in the surface pressure was monitored and recorded every 3 s. The whole setthe dish and the tensiometerwas closed in a container filled with argon atmosphere. Relative humidity was 100%.