Alignment of the E. coli CcmD protein with representative bacterial CcmD sequences. The amino acid sequences of CcmD proteins of E. coli K-12 strain MG1655 (accession no. POABM5) (Ec), Nitrospira multiformis ATCC 2516 (YP_411906) (Nm), Nitrosomonas eutropha C71 (YP_747848) (Ne), Polynucleobacter sp. strain QLW-PIDMWA-1 (Psp.), Ralstonia metallidurans CH34 (YP_584534) (Rm), Shigella boydii Sb227 (YP_408515) (Sb), Salmonella enterica subsp. enterica serovar Typhi CT1 (NP_456796) (Set), Erwinia carotovora subsp. atroseptica SCRI1043 (YP_049982) (Ec), Vibrio vulnificus YJ016 (NP_935245) (Vv), Vibrio parahaemolyticus RIMD 2210633 (NP_798599) (Vp), Pasteurella multocida subsp. multocida PM70 (NP_244945) (Pm), Haemophilus influenzae Rd KW20 (NP_439249) (Hi), Psychromonas ingrahamii 37 (ZP_01349302) (Pi), Shewanella putrefaciens CN-32 (ZP_00815699) (Sp), Shewanella oneidensis MR-1 (NP_715900) (So), Rhodobacter sphaeroides 2.4.1 (YP_351852) (Rs), Roseovarius sp. strain 214 (ZP_01034520) (Rsp.), and R. capsulatus SB1003 (P29963) (Rc) were aligned with ClustalW (25). Amino acid similarities are indicated as follows: asterisk and gray shading, identical residues in all sequences; colons, conserved substitution; and period, semiconserved substitution. The predicted domain structure of side chains, the net charges ([R + K] − [D + E]) of the NTD and CTD, and the location of the GXY motif are also indicated.

Protein-protein interactions with CcmC:His₆ determined by analysis of proteins brought down with a nickel column. E. coli RK103 cultures containing pRGK369 (pGEXCcmABC:6XHisD:FLAG) were grown, membrane proteins were isolated, and nickel affinity chromatography was performed as described in the text. Purified proteins were separated by SDS-PAGE and probed with (A) six-His antisera (α6xHis), (B) GST antisera (αGST), or (C) FLAG antisera (αFLAG). The positions of standard proteins with the indicated molecular masses (10³) are shown on the left, and the positions of CcmC:His₆, GST:CcmA, and CcmD:FLAG are indicated by arrows.

Heme is transferred to CcmE with or without CcmD. E. coli RK103 harboring pRGK345 and pRGK371 (lanes 1 and 4), pRGK370 and pRGK371 (lanes 2 and 5), or pRGK371 (lane 3) was grown and induced as described in the text. Ten micrograms of crude n-dodecyl-β-d-maltoside-solubilized membrane proteins was separated by SDS-PAGE, the presence of holo-CcmE was detected by heme staining, and total CcmE was detected with CcmE antibodies. A typical gel with heme staining (lanes 1 to 3) and a CcmE Western blot (lanes 4 and 5) are shown, and the percentages of holo-CcmE, corrected for the amount of total CcmE present in each lane, are indicated at the bottom. Abbreviations: Δccm, E. coli RK103; SysIΔE, pRGK345; SysIΔDE, pRGK370; CcmE:6XHis, pRGK371.

Holo-CcmE* is not released into the periplasm without CcmD. Periplasmic protein fractions were prepared from E. coli RK103 coexpressing pRGK345 and pRGK362 (lane 1) or pRGK370 and pRGK362 (lane 2). Twenty micrograms of periplasmic proteins was separated by SDS-PAGE and transferred to nitrocellulose. (Top panel) Heme staining showing the levels of holo-CcmE*. (Bottom panel) Immunoblot with CcmE antisera. Abbreviations: Δccm, E. coli RK103; SysIΔE, pRGK345; SysIΔDE, pRGK370; CcmE*:6XHis, pRGK362; αCcmE, CcmE antisera.

Heme staining of holocytochrome c₄:His₆ induced with increasing concentrations of arabinose shows that an N-terminal six-His tag on CcmD reduces its function. E. coli RK103 coexpressing pRGK353 and either pRGK373 (lanes 1 to 7) or pRGK372 (lanes 8 to 14) was grown and treated as described in the text. A typical blot to determine the levels of holocytochrome c₄ (indicated at the bottom) compared to 100% (lane 3) is shown. The concentrations of arabinose are indicated above the lanes. The molecular masses of the standards (lane Std) are 29.2 kDa (top band) and 20.2 kDa (bottom band). Abbreviations: ΔCCM, E. coli RK103; SysIΔD, pRGK353; pBADc₄-CcmD, pRGK373; pBADc₄-6XHis:CcmD, pRGK372; c₄, B. pertussis cytochrome c₄.