MpeV is a lyase isomerase that ligates a doubly linked phycourobilin on the β-subunit of phycoerythrin I and II in marine Synechococcus

Abstract

Synechococcus cyanobacteria are widespread in the marine environment, as the extensive pigment diversity within their light-harvesting phycobilisomes enables them to utilize various wavelengths of light for photosynthesis. The phycobilisomes of Synechococcus sp. RS9916 contain two forms of the protein phycoerythrin (PEI and PEII), each binding two chromophores, green-light absorbing phycoerythrobilin and blue-light absorbing phycourobilin. These chromophores are ligated to specific cysteines via bilin lyases, and some of these enzymes, called lyase isomerases, attach phycoerythrobilin and simultaneously isomerize it to phycourobilin. MpeV is a putative lyase isomerase whose role in PEI and PEII biosynthesis is not clear. We examined MpeV in RS9916 using recombinant protein expression, absorbance spectroscopy, and tandem mass spectrometry. Our results show that MpeV is the lyase isomerase that covalently attaches a doubly linked phycourobilin to two cysteine residues (C50, C61) on the β-subunit of both PEI (CpeB) and PEII (MpeB). MpeV activity requires that CpeB or MpeB is first chromophorylated by the lyase CpeS (which adds phycoerythrobilin to C82). Its activity is further enhanced by CpeZ (a homolog of a chaperone-like protein first characterized in Fremyella diplosiphon). MpeV showed no detectable activity on the α-subunits of PEI or PEII. The mechanism by which MpeV links the A and D rings of phycourobilin to C50 and C61 of CpeB was also explored using site-directed mutants, revealing that linkage at the A ring to C50 is a critical step in chromophore attachment, isomerization, and stability. These data provide novel insights into β-PE biosynthesis and advance our understanding of the mechanisms guiding lyase isomerases.

Keywords: bilin lyase; cyanobacteria; lyase isomerase; phycobilisome; phycoerythrobilin; phycourobilin; posttranslational modification.

Figure 1

Model of Synechococcus sp. RS9916 PBS rod assembly. Model of an RS9916 phycobilisome (PBS) containing phycoerythrin I (PEI, red), phycoerythrin II (PEII, orange), phycocyanin (PC, light blue), and an allophycocyanin (AP, deep blue) core. A depiction of PBS rod assembly shows the addition of bilin via posttranslational modification of apo-α and apo-β monomers (white) forming holo-monomers (purple), which come together to form a heterodimer. Heterodimers are subsequently arranged in trimers (αβ)₃ followed by heterohexamers (αβ)₆ and with the help of linker polypeptides, the PBS rod is formed and bound onto the core (4, 49, 65, 66).

Figure 2

Chemical structures of phycoerythrobilin (PEB) and doubly linked phycourobilin (PUB). Posttranslational pigment attachment is catalyzed by bilin lyases or lyase isomerases with a thioether linkage at the 3¹ carbon of the bilin A ring during single attachment or additionally through the 18¹ carbon of the bilin D ring when doubly attached.

Figure 3

Phylogenetic tree of the MpeV enzyme family. Sequence names include abbreviation of the genus (see below), the strain name, the finest taxonomical level for each strain sensu (67), i.e., subcluster (e.g., 5.2), clade (e.g., VII), or subclade (e.g., IIIa), as well as the pigment type sensu (40). The pigment phenotype of each strain is indicated by a colored square and the CA4-island type (A or B) by a blue circle. Strains called “BGL specialists” (light orange square) correspond to strains that are genetically similar to pigment type 3 dB (a.k.a. CA4-B) but that have lost the ability to perform CA4 (i.e., natural CA4 mutants) and are stuck in some intermediate state between the green and blue light phenotype (35). Also worth noting, several strains with a low PUB:PEB ratio have acquired a complete or partial CA4-A island by lateral transfer (pigment type 3aA), but none is CA4-capable (35). Series of two numbers at nodes of the tree correspond to Bayesian posterior probabilities (PP, ranging between 0 and 1) and bootstrap values for Maximum Likelihood (ML), respectively. Only values higher than 50% for ML bootstrap values and 0.50 for PP are shown on the Bayesian tree. The Synechococcus sp. RS9916 strain used in the present study is indicated in bold. Abbreviations for genus names: Fre., Fremyella; Glo., Gloeobacter; Nos., Nostoc; Pro., Prochlorococcus; Syn., Synechococcus. Other abbreviations: BL, Blue light; BGL, Blue-green light; CA, Chromatic acclimaters; CA4, Chromatic acclimation type IV; GL, Green light; PEB, Phycoerythrobilin; PUB, Phycourobilin.

Figure 4

Recombinant protein activity of RS9916 MpeV on CpeB and MpeB. Relative absorbance representing purified recombinant RS9916 CpeB (A–B) and MpeB (F–G) expressed in the presence/absence of putative lyases MpeV, CpeS, and/or CpeZ, as indicated in the legend inset. All proteins were induced in E. coli cells with bilin synthesis genes, purified and diluted to similar concentrations prior to analysis. Purified CpeB (C–E) and MpeB (H–J) were resolved via SDS-PAGE and imaged with zinc-enhanced fluorescence at 460 to 490 nm (C and H), which excites PUB and at 520 to 545 nm (D and I), which excites PEB. The same gels were then stained with Coomassie blue (E and J) to visualize proteins. Positions of target substrates CpeB and MpeB in gels are indicated by red arrows. This data is representative of three independent biological replicates.

Figure 5

Extracted ion chromatograms and LC-MS spectra for trypsin-digested peptides from RS9916 recombinant β-subunits.A, extracted ion chromatogram (EIC, inset left) and LC-MS for the peptide MAAC₈₂LR at m/z 417.5³⁺ and 625.8²⁺ of recombinant RS9916 CpeB C82-PEB (∼560 nm). B, EIC (inset left) and LC-MS for peptide LDAVNAITSNASC₅₀IVSDAVTGMIC₆₁ENTGLIQAGGNCYPNRR at m/z 1200.2⁴⁺ and 960.46⁵⁺ of recombinant RS9916 CpeB C50, 61-PUB (∼490 nm). C, EIC (inset left) and LC-MS for the peptide KMAAC₈₂LRU at m/z 417.5³⁺ and 689.5²⁺ of recombinant RS9916 MpeB C82-PEB (∼560 nm). D, EIC (inset left) and LC-MS for peptide LDAVNAIAGNAAC₅₀IVSDAVAGICC₆₁ENTGLTAPNGGVYTNR at m/z 1116.8⁴⁺ and 1489.7³⁺ of recombinant RS9916 MpeB C50, 61-PUB (∼490 nm). Inset right graphs are the UV–visible absorbance spectra for the peaks present in the EIC (inset left). The type of bilin is indicated per panel. All samples expressed in the presence of CpeS, MpeV, CpeZ, and bilin synthesis genes. These results are representative of two independent biological replicates.

Figure 6

Recombinant protein activity of MpeV on RS9916 CpeB mutant variants. Recombinant protein coexpression of RS9916 CpeB and mutant variants in the presence of CpeA, MpeV, CpeS, and CpeZ for maximum solubility and chromophorylation. All purified protein samples were expressed in the presence of bilin synthesis genes. A, relative absorbance of nonmutated CpeB, denoted as wild-type (WT; black line), shows addition of phycoerythrobilin (PEB) to C82 with an absorbance peak at 559 nm and addition of a doubly ligated phycourobilin (PUB) at C50, 61 with an absorbance peak at 492 nm. Mutant variants CpeB-C50A (C50A; orange line), CpeB-C61A (C61A; purple line), and CpeB double mutant C50A/C61A (DM; red line) are shown. These samples were resolved by SDS-PAGE and analyzed by zinc-enhanced fluorescence of bound PUB excited at 488 nm and bound PEB excited at 532 nm (B–C). Western blot analysis using anti-CpeB antibodies was used to detect the total amount of CpeB present (D). This data is representative of two independent biological replicates.

Figure 7

Extracted ion chromatograms and LC-MS from recombinant RS9916 CpeB mutant coexpressions showing PUB bilin addition along C50, 61 residues. Coexpressions and abbreviations are same as Figure 6. A, extracted ion chromatogram (EIC) of 1153.8063 (37–77) ⁴⁺ from C50A CpeB mutant. B, EIC for 1006.7336 (37–77 unmod) ⁴⁺ from C50A mutant. C, EIC for 1153.8063 from C61A mutant. D, EIC for 1006.7337 from C61A mutant. E, mass spectrum (MS) of 11.56 min peak in B. F, MS from 10.43 min peak in C. G, MS from 10.96 min peak in D. H, MS from 11.37 min peak in D. The observed m/z ratios for each of the labeled peaks are listed in Table 3. This data is representative of two independent biological replicates.