The subunit composition, amino acid sequence and haem-binding characteristics of bacterioferritin (BFR) from Pseudomonas aeruginosa have been studied. Unlike other BFRs, P. aeruginosa BFR was found to contain two subunit types, designated alpha and beta, which differed considerably in their amino acid sequences. The N-terminal 69 and 55 amino acids of the alpha and beta subunits respectively were determined. The alpha subunit differed most from other BFRs. The two subunits were present in variable proportions in different preparations. The maximum stoichiometry of haem binding was found to be sample-dependent and to be different from the previously reported one per subunit [Kadir and Moore (1990) FEBS Lett. 271, 141-143]. This previous haem-binding study was shown to have been carried out with damaged protein, which contained both normal alpha and beta subunits and shorter versions of these that appeared to have been produced by cleavage of the normal subunits. The possibility that aging processes degrade ferritins and affect their haem-binding characteristics is discussed.

The e.p.r. and magnetic circular dichroism (m.c.d.) spectra of bacterioferritin (BFR) extracted from Pseudomonas aeruginosa and Azotobacter vinelandii have been studied over a wide temperature range down to liquid-helium temperature. The e.p.r. spectra show the presence of low-spin Fe3+ haem with g values of 2.86, 2.32, 1.48 (P. aeruginosa) and 2.88, 2.31, 1.46 (A. vinelandii), in both the presence and absence of the BFR core. Together with evidence from the porphyrin-to-Fe3+ charge-transfer band at 2240 and 2270 nm the axial haem ligands are identified as two methionines. The low-temperature m.c.d. spectra in the region 300-1000 nm of P. aeruginosa and A. vinelandii BFR are identical with one another and unaffected by removal of the iron core. Hence it can be concluded that the presence of the iron core has no detectable effect on the electronic states and on the stereochemistry of the haem group. This was unexpected, in view of the observations by Watt, Frankel, Papaefthymiou, Spartalian & Stiefel [(1986) Biochemistry 25, 4330-4336] that the redox potential of the haem group in A. vinelandii BFR shifts from -475 mV to -225 mV on removal of the core. The e.p.r. spectra of holoBFR show a broad symmetrical derivative-shaped band centred at g = 2.0 which decreases in bandwidth as the temperature is raised. This signal is assigned to the uncompensated electron spins of the iron core.

Reactions of reduced horse spleen ferritin with horse and Saccharomyces cerevisiae ferricytochromes c, cow ferricytochrome b5, sperm-whale metmyoglobin and Pseudomonas aeruginosa ferricytochrome c-551 were investigated by u.v.-visible spectrophotometry. In all cases the reduced ferritin reduced the ferrihaemoproteins. The rate of reduction varied from less than 0.2 M-1.s-1 for metmyoglobin to 1.1 x 10(3) M-1.s-1 for horse ferricytochrome c (0.1 M-phosphate buffer, pH 7.4, at 25 degrees C). We conclude that the mechanism of ferrihaemoprotein reduction involves long-range electron transfer through the coat of ferritin and that such electron transfer is rapid enough to account for the rates of iron release observed by other workers in reductive release assays.

Horse spleen ferritin is shown to bind haem to generate a haemoprotein, named herein haemoferritin. A total of 14-16 haem molecules are bound per 24 subunits of ferritin. The molecular mass of the non-haem-iron-free haemoferritin has been determined to be 420 + 40 kDa, indicating that haem binding does not lead to dissociation of the 24 subunits that comprise the ferritin molecule. The functional role of the bound haem has been investigated with respect to the release of iron from the non-haem iron core. The bound haem is shown to increase the rate of iron release in a reductive assay system. In the absence of haem the rate of iron release depends on the redox potential of the reductant, but in the presence of haem the rate is largely independent of the reductant and is faster than the rate for the haem-free ferritin. These data indicate that electron transfer across the protein coat of ferritin is the rate-limiting step of iron release in the absence of haem, but in the presence of haem electron transfer is not rate-limiting.