Structural investigations of redox regulation in ATFKBP13 4

4.2.1 AtFKBP13 is targeted to the thylakoid lumen of chloroplasts by the dependent pathway ∆pH-To determine whether AtFKBP13 is targeted to the chloroplast,and if so, what its suborgane

Chapter 4 Results and Discussion

CHAPTER 4 RESULTS AND DISCUSSION

BRIDGES

4.1.1 Overall Structure of AtFKBP13-S2

The AtFKBP13-S2 molecule reveals predominantly β-structures consisting of six

β-strands and two α-helices (Fig 4-1) The β-strands form an integral antiparallel β-sheet

than constitutes the core of the protein

Figure 4-1 Structure of AtFKBP13-S2 The redox active disulfides are shown in

ball-and-stick representation and the sulphur atoms are shown as yellow balls

Chapter 4 Results and Discussion

The secondary structures are arranged in the order: β1 β4 β5a α2 β5b α1 β2 β6a β6b β3

(Fig 4-2) The β5 strand of AtFKBP13 is split into two fragments, β5a and β5b, as in

MtFKB17 [Suzuki et al., 2003] and HsFKB12 [Meadows et al., 1993] However, the β5a

strand of hFKB12 is formed only when a ligand (ascomycin) is bound [Meadows et al.,

1993] Otherwise it is disordered [Michnick et al., 1991; Moore et al., 1991] Between β5a

and β5b is the inserted α2 helix, similar to that of MtFKB17 The β6 strand pair, which is

unique to AtFKBP13, is formed by strands β6a andβ6b connected by a short loop

5 15 25 35 45

CEFSVSPSGL AFCDKVVGYG PEAVKGQLIK AHYVGKLENG KVFDSSYNR

55 65 75 85 95 GKPLTFRIGVG EVIKGWDQGI LGSDGIPPML TGGKRTLRIP PELAYGDRG

105 115 125

AGCKGGSCLIP PASVLLFDIE YIGKA

Figure 4-2 Secondary structure elements of AtFKBP-13 The α-helix is

represented by a cylinder and the β-strand is represented by an arrow

Chapter 4 Results and Discussion

A notable feature in the tertiary structure of AtFKBP13-S2 is the presence of two intra-chain disulfide bonds These disulfide bonds, Cys5–Cys17 and Cys106–Cys111, form the two redox active motifs These disulfides have no counterparts in other animal

or yeast FKBPs Eukaryotic FKB protein sequences indicate that this region is rather a

unique feature of Arabidopsis FKBP13 Also, structural comparisons with other FKBPs

indicate that AtFKBP13 has a conserved PPIase domain with additional strands (β6a and

β6b) inserted at the C-terminus where Cys106 and Cys111 are located

5–129, blue), hFKBP12 (residues 1-107, magneta) and L pneumophila

FKBP25 (residues 496–612, green) The redox active-site disulfides of

AtFKBP13 are shown in ball-and-stick representation

N C

Chapter 4 Results and Discussion

Alignment between AtFKBP13 and representatives of other FKBPs, namely

hFKB12 [Wilson et al., 1995] and the Macrophage Infectivity Potentiator protein from Legionella pneumophila, LpMIP [Riboldi-Tunnicliffe et al., 2001] through the Dali

server [Holm, and Sander, 1993] gives the r.m.s deviation values of 1.3 Å (with a Z-value

of 17.5) and 1.4 Å (with a Z-value of 17.2), respectively The core regions for these structures are similar (Fig 4-3) The best matches are found in the regions that are directly involved in the prolyl isomerase activity A few residues are conserved among these structures These are the residues that form the substrate-binding pocket for the pipecolinyl ring of FK506 [Van Duyne et al., 1993], namely Tyr37, Phe47, Asp48, Phe59, Val66, Ile67, Trp72, Tyr99, Ile113, Leu119 and Phe121 These residues are essential for binding and maintaining the hydrophobic core of FK506 [Radzicka et al., 1992]

4.1.3 Catalytic domain

The catalytic domain of AtFKBP13 is composed of two sub-regions, the PPIase (residues 26-125) and Cys-Xn-Cys redox active motifs (residues 5-17 and 106-111) The PPIase active region follows the general extended PPIase fold, which consists of a four-stranded β-sheet and an α-helix inserted between them In AtFKBP13, the corresponding β-sheet comprises strands β1-β5, helix α1 and a short 310 helix, α2 The Cys-Xn-Cys motif regions are composed of two disulfides, found at the N and C-termini, respectively The active site disulfide bond between Cys5 and Cys17 at the N-terminus is located at the β1 strand Both Cys5 and Cys17 are partially solvent exposed The second disulfide bond

Chapter 4 Results and Discussion

These disulfide bonds form additional secondary structures that are located on either side of the central β-sheet and reflect the intrinsic versatility and flexibility of this region The two active sites that are involved in redox regulation exist in their oxidized state in the crystal structure as shown by their clearly defined electron density (Fig 4-4)

The theoretical dihedral energy for the N-terminal disulfide is 3.60 kcal mol-1 and for the C-terminal disulfide is 4.00 kcal mol-1, calculated with the program AMBER [Weiner et al., 1984], indicate the disulfide bonds are stable with less conformational strain The corresponding value varies from 0.5-4.7 for most protein disulfide bonds [Darby and Creighton, 1995] and rarely reaches values over 5.0 Because the active site disulfide bonds are extremely stable, they may also act as strong reductants

the C-terminal disulfide region The amino acid residues in the region are

numbered

Chapter 4 Results and Discussion

Despite this similarity, however, the two redox active disulfides show remarkable difference in their B-factors The average B-factor for the side chain atoms (Cβ and Sγ) of the two cysteines is 39.32 for the N-terminal disulfide versus 26.22 for the C-terminal disulfide Even though, all the residues in both redox active sites have very well defined electron density in the final 2Fo-Fc map, the side chain atoms of the two cysteines in the C- terminal disulfide have been better defined with more spherically shaped electron densities than those in the N-terminal disulfide

4.1.4 Surface of AtFKBP13

In AtFKBP13, the residues around the two redox active sites form two grooves on the protein surface, the active site Grooves N and C (Fig 4-5a) Groove C is built exclusively by residues from the C-terminus and is essentially hydrophobic

Figure 4-5 (A) Surface charge distributions (blue for positive and red for negative

charge) for the AtFKBP13 monomer The dark red region indicates

Chapter 4 Results and Discussion

a potential of less than -12 kT/e, while dark blue indicates greater than 12

kT/e The electrostatic potentials were calculated by GRASP (B)

Space-filling representation in which the sulfur atoms of Cys-106 and Cys-111 are

exposed on the surface of the molecule and (C) the S atom of Cys5 is fully

exposed on the surface, and the sulfur atom of Cys17 is buried

The residues in the vicinity of Cys106 include Gly105, Pro114, Ser110, Leu112, and Ile113 The Sγ atoms of Cys106 and Cys111 are exposed on the protein surface (Fig 4-5b) Unlike Groove C, the formation of Groove N involves residues both from the N and C-termini While atom Sγ of Cys5 is fully exposed on the surface, the sulphur atom

of Cys17 is buried (Fig 4-5c) The redox active site loops C and N adopt an open confirmation in the crystal structure so that the Sγ atom and the substrate binding site are exposed to solvent The two redox active grooves are adjacent to each other on the protein surface Separately located in the two grooves with different accessibility, the two redox active sites probably function independently to some extent The distance between the two active sites, measured between the Sγ atoms of Cys5 and Cys106 is 27.6 Å

Gupta et al (2001) identified the first chloroplast FKBP from Arabidopsis and the

Rieske protein, a component in the photosynthetic electron transport, as its putative target The interactionbetween AtFKBP13 and the Rieske protein probably occurs beforethey are imported into the thylakoid because the mature proteinsdo not interact with each

Chapter 4 Results and Discussion

other Both yeasttwo-hybrid and in vitro protein interaction assays demonstratethat the full-length precursor proteins (or the cytoplasmic forms) interact In addition, the intermediate (or stromal) forms of the two proteins also interact well These resultssuggest that AtFKBP13 associates with the Rieske protein both before and after the import of the proteins into the chloroplast stroma In addition, AtFKBP13 and Rieske intermediate forms also caninteract after they are imported into the thylakoid lumen butbefore the thylakoid signal peptides are cleaved After thecleavage of the peptide, the two matured proteins probably dissociate

4.2.1 AtFKBP13 is targeted to the thylakoid lumen of chloroplasts by the dependent pathway

∆pH-To determine whether AtFKBP13 is targeted to the chloroplast,and if so, what its suborganellar location is, protein import assays have been performed with isolated chloroplasts and the AtFKBP13 precursor as a substrate The translated AtFKBP13 precursorisabout 27 kDa (estimated by mobility in SDS/PAGE) When isolatedintact, pea chloroplasts were incubated with the precursor proteinin the presence of ATP, a 13 kDa protein was generated (Fig 4-6A, lane 2) After incubation with the protease thermolysin (which under the used conditions does not penetrate the chloroplast envelope), intact chloroplasts were reisolated and fractionated.The resistance of the 13 kDa polypeptide to degradation byexogenously added thermolysin (Fig 4-6A, lane 3) indicates thatit is located within the chloroplast and is a product of the precursor protein import Further analysis revealed that the AtFKBP13 protein was associated with the

Chapter 4 Results and Discussion

Figure 4-6 AtFKBP13 is targeted to the chloroplast thylakoid lumen by

the ∆pH-dependent pathway (A) Chloroplast import assay of AtFKBP13:

translation products (lane 1), chloroplasts (lane 2), thermolysin-treated

chloroplasts (lane 3), stromal fraction (lane 4), thylakoid fraction (lane 5),

thermolysin-treated thylakoid fraction (lane 6), sonicated thylakoid membrane fraction (lane 7), soluble contents of thylakoid lumen (lane 8)

(B) Western blot analysis of stromal (lane S) and thylakoid lumen (lane L)

fractions of Arabidopsis chloroplasts by Coomassie staining (left),

anti-AtFKBP13 (middle), and anti-Plastocyanin (PC) antibodies (right) Approximate molecular masses (kDa) are shown on the side.

Sonication of the thylakoidfraction liberated the 13 kDa polypeptide in a soluble form (Fig 4-6A, lane 8), indicating that the protein is located inthe thylakoid lumen From this experiment, it was concluded thatAtFKBP13 is a previously uncharacterized thylakoid lumen protein

Chapter 4 Results and Discussion

The N-terminal extension of AtFKBP13 has the characteristic features of a thylakoid lumen protein presequence (Fig.4-7).It is bipartite, with the first region being hydrophilic innature and enriched in basic and hydroxylated residues, whichare features

of a chloroplast envelope-transfer signal [Keegstra and Cline, 1999].The twin arginine

hydrophobicdomain ending with AXA These features suggest that AtFKBP13 may be translocated into the thylakoid lumen by a pH-dependentpathway

MSSLGFSVGT CSPPSEKRKC RFLVNNSLNK AEAINLRNKQ KVSSDPELSF 50

GYGPEAVKGQ LIKAHYVGKL ENGKVFDSSY NRGKPLTFRI GVGEVIKGWD 150 QGILGSDGIP PMLTGGKRTL RIPPELAYGD RGAGCKGGSC LIPPASVLLF 200 DIEYIGKA

Figure 4-7 Sequence of precursor AtFKBP13 protein The amino acid

sequence of the AtFKBP13 protein deduced from the cDNA The double

arginine motif is shown as bold and highlighted The thylakoidal

processing peptidase site (AxA) is shown in italics and the arrow indicates

the putative cleavage site The mature protein sequence is shown in black

Chapter 4 Results and Discussion

4.2.2 AtFKBP13 precursor interacts with the Rieske protein

To function as chaperone or other regulatory entities, immunophilins often interact with their target proteins [Pratt et al., 2001; Harrar et al., 2001].To identify a putative target for AtFKBP13, a yeast two-hybrid screening procedure was used The precursor of AtFKBP13 wasused as bait in fusion with the Gal4 DNA-binding domain to screen an Arabidopsis cDNA library in a Gal4 activation-domain vector. Among the AtFKBP13 interacting clones that were sequenced,a number of them encoded the Rieske protein of various lengthsfused in frame with the activation domain The longest onehad four amino acid truncations in the N terminus The Rieske proteinis an essential subunit

of the cytochrome bf complex in the photosynthetic electron transport chain [Hope, 1993] It is located in the thylakoidwith a transmembrane domain and a soluble region in the lumen[Hope, 1993] As AtFKBP13 is located in the lumen, the Rieske protein mayserve as a physiological target for AtFKBP13

A search of the Arabidopsis Expressed Sequence Tag (EST) database identified

an EST encoding the full-length precursor of the Rieske protein This EST39G11T7 (GenBankaccession no AJ243702) was sequenced on both strands to confirmits identity and was used in subsequent experiments to determine the interacting domains of the

domains of the AtFKBP13or Rieske protein were fused in frame with either the bindingor activation domain of the Gal4 protein in the vectors Differentcombinations of these constructs were co transformed into yeast strain Y190 The interactions were studied

DNA-Chapter 4 Results and Discussion

by the growth on selection medium and filter lift assay and quantified by the galactosidaseactivity The precursor of AtFKBP13 interacts with the precursor,mature, and lumen domain but not with the transit peptide of the Rieske protein When the chloroplast envelope signal peptide was deleted from AtFKBP13, the truncated protein retained interaction with the Rieske protein However, the mature form of AtFKBP13 lackingthe entire transit peptide did not interact with the full-lengthor any domain of the Rieske protein The transit peptideof AtFKBP13 interacted with the precursor and mature form of Rieske,although the interactions were significantly weaker comparedwith the interaction between the AtFKBP13 precursor and the Rieske protein The chloroplast envelope signal or thylakoid-targetingsignal alone did not interact with the full-length or any domainof the Rieske protein The transit peptideof AtFKBP13 is required and, to a certain degree, sufficientfor interaction with the Rieske protein The mature region,more specifically, the lumen domain, of the Rieske protein is sufficientfor interaction

β-Figure 4-8 Interaction between AtFKBP13 and the Rieske protein

determined by in vitro protein interaction assays GST (lanes 1 and 5),

GST-Rieske precursor (lanes 2 and 6), GST-Rieske mature form (lanes 3

Chapter 4 Results and Discussion

and 7), and GST-Rieske lumen domain (lanes 4 and 8) were immobilized

on glutathione beads and used to "pull-down" the purified precursor (lanes

1–4) or mature (lanes 5–8) form of AtFKBP13 The approximate

molecular masses (kDa) are shown on the right side (A) An immunoblot

of co purified products probed with anti-AtFKBP13 The arrows on the

left and right sides indicate the position of the precursor and mature form

of AtFKBP13, respectively (B) A Coomassie blue stained gel shows the

amount of each bait protein used in the pull-down experiment

The interaction between AtFKBP13 and the Rieske protein in vitro was also

tested by protein–protein interaction assay As shownin Fig 4-8, the precursor, mature,

or lumen domain of the Rieske proteinco purified with the precursor but not with the mature formof AtFKBP13 (Fig 4-8A) When GST was used as an affinity agent,neither the precursor nor the mature form of AtFKBP13 was co purified(Fig 4-8A, lanes 1 and 5) The AtFKBP13 antibodies also reactedwith an unknown 33 kDa protein, associated only with the preparationof the GST-Rieske mature protein (Fig 4-8A, lanes 3 and lane7) The amount of GST and other bait proteins used in the experimentare shown in Fig 4-8B Some GST was detected in the preparationsof Rieske precursor and mature protein fusions, suggestingthat a small portion of GST fusion proteins were degraded (Fig 4-8B, lanes 2, 3, 6, and 7)

Because the precursor, but not the mature form, of AtFKBP13interacted with the Rieske protein, and the precursor of AtFKBP13 wasnot detectable by the western blot in

Chapter 4 Results and Discussion Arabidopsis plants, it was difficult to study the in vivo interaction of the two partner

proteinsby using immunoprecipitation Indeed, the AtFKBP13 mature protein was

precipitated by using an AtFKBP13 antibody but didnot detect any Rieske protein that was co purified In a similarmanner, a Rieske antibody was used to precipitate the Rieske proteinbut AtFKBP13 was not co purified This further supports the idea that the two mature proteins most probablydo not interact with each other

4.2.3 AtFKBP13 down-regulates accumulation of Rieske protein

AtFKBP13 is localized in the thylakoid lumen, and the Rieske proteinis localized

in the thylakoid membrane with a soluble domain in the lumen, making it physically possible for the two matureproteins to interact However, our results suggest that theseproteins probably interact before they are targeted to theirfinal destination

Figure 4-9 Rieske protein accumulation in AtFKBP13 silenced plants

Western blot analysis of AtFKBP13 RNAi (lanes 1 and 3) and control

plants (lanes 2 and 4) with anti-Rieske (lanes 1 and 2) or anti-PC (lanes 3

and 4) antibody

Chapter 4 Results and Discussion

What is the functional relevance of interactionbetween the two protein partners?

To address this question, a "reverse genetics" approach was undertaken to suppress the expressionof AtFKBP13 in transgenic plants and the consequenceof this manipulation to the Rieske protein was examined

If the AtFKBP13 interaction with the Rieske protein is functionallyrelevant, it is expected that AtFKBP13 silencing would alterthe level or function of the Rieske protein Because the interaction may occur before the proteins are imported to the final destination,we suspected that AtFKBP13 might regulate the accumulationof the Rieske protein As the Rieske protein is an essential subunitof the cytochrome bf complex in the

photosynthetic electron-transport chain, changes in the Rieske protein may affect the photosyntheticelectron transport It is possible that the Rieske protein is modified in a more subtle manner

The level of Rieske protein was examined in the control and RNAi plants by westernblot analysis and found a significant difference in the accumulationof the Rieske protein As shown in Fig 4-9, the RNAi plants produced higher levels of the Rieske protein (Fig 4-9, lane 1) as comparedwith the control plants (Fig 4-9, lane 2) Statistical analysis on eight RNAi lines, eight empty vector control lines, and wild-type plants indicated that the Rieske protein in the AtFKBP13-silencedplants produced about 70% more Rieske protein (Fig 4-9, lanes 1 and 2) As a control westernblot analysis with a plastocyanin antibody did not reveal any significantdifference in the accumulation of PC

in the RNAiand control plants (Fig 4-9, lanes 3 and 4) RNA gel blot analysisdid not detect any significant difference in the Rieske mRNAlevels among RNAi and control

Chapter 4 Results and Discussion

plants, suggesting that AtFKBP13 affects the level of the Rieske protein by a transcriptionalprocess

post-4.2.4 Mature FKBP13 is a target for reduction by thioredoxin

The possibility of thioredoxin-linked reduction of AtFKBP13 arose from the finding of two solvent exposed disulfides in its structure and from recent evidence that AtCYP20-3, a functionally related protein present in the chloroplast stroma, has been found to be a thioredoxin target [Motohashi et al., 2001, 2003] We tested this possibility

by applying the NADP/thioredoxin reducing system of E.coli combined with

fluorescence gel analysis using mBBr as a thiol-specific probe In this assay, AtFKBP13 was incubated with thioredoxin and NTR in the presence of NADPH, a source of reducing equivalents The newly formed –SH groups were labeled with mBBr, the proteins separated by SDS-PAGE and the fluorescence was recorded The reduction of disulfide(s) by the NADP/thioredoxin system was reflected by an increase in the fluorescence of the treated protein (Fig 4-10) Owing to the absence of additional free cysteines in the sequence of AtFKBP13, the protein alone did not react with mBBr (Fig 4-10, lane 1, control treatment)

Chapter 4 Results and Discussion

Figure 4-10 Reduction of AtFKBP13 by the NADP_thioredoxin system

from E coli Ctrl- control (FKBP13 alone); Cpl- complete thioredoxin

system (NADPH+NTR+Trx) plus FKBP13; Trx- thioredoxin; -Trx, -NTR,

and -NADPH lanes indicate the complete thioredoxin system in which

each of these components was individually omitted Protein refers to the

complete treatment in which the gel was stained with Coomassie blue

By contrast, incubation of AtFKPB13 with the NADP/thioredoxin system for 20 minutes resulted in marked fluorescent labeling indicating disulfide reduction (Fig 4-10, lane 2, complete treatment) Reduction was not observed when any one of the components of the system was omitted, i.e., thioredoxin, NTR or NADPH (Figure 4-10, lanes 3-5)

A similar experiment performed with mutant forms of AtFKBP13 in which either the N- or C-terminal disulfide (Cys5/17 or Cys106/111) was replaced by serine established that both S-S bonds are fully reduced by thioredoxin Additional assays using

Chapter 4 Results and Discussion

a low concentration of DTT (0.5 mM) as hydrogen donor and different types of

thioredoxin (E.coli, chloroplast f and m, extraplastidic h) indicated the chloroplast m-type

to be the most efficient DTT alone was a poor reducing agent under these conditions, especially for cleaving the C-terminal disulfide

4.2.5 Mature AtFKPB13 PPIase activity is regulated by redox status

Having structurally resolved the position of the disulfide bridges and established their reducibility by thioredoxin, we sought to investigate the effect of redox status on catalysis of the PPIase reaction by AtFKBP13 In contrast to AtCYP20-3, the isolated

oxidized form of AtFKBP13 efficiently catalyzed the peptidyl-prolyl cis-trans

isomerization of a chromogenic synthetic pentapeptide in a standard PPIase assay (Table 4-2)

Table 4-1 Requirement for both disulfide (S-S) groups for the PPIase

Chapter 4 Results and Discussion

To ascertain the importance of each disulfide bridge to total PPIase activity, mutant forms of AtFKBP13 were constructed which lacked either one or both cysteine pairs AtFKBP13 proteins containing mutations in the amino (C5S/C17S) or carboxy (C106S/C111S) terminal cysteine pairs showed, respectively, about 30 and 55%

reduction in catalytic efficiency (k cat /K m) This result suggests that the surface-exposed terminal disulfide bridge may be more important for catalysis and may constitute the major site of regulation by thioredoxin

C-Activity was reduced almost 80% in the quadruple cysteine mutant (C5S/C17S

and C106S/C111S) Following incubation and reduction of AtFKBP13 by Arabidopsis thioredoxin m and DTT, the PPIase activity was reduced by 55%, whereas DTT alone

had a minimal effect These results suggest that AtFKBP13 PPIase activity can be

modulated by the redox state in vitro

Interestingly, unlike certain enzymes of the Calvin cycle (e.g., fructose bisphosphatase), neither pH nor Mg2+ appeared to be a factor in determining the extent of activation by thioredoxin [Buchanan, 1980] AtFKBP13 was found to remain catalytically active down to pH 5, suggesting that this enzyme is functionally adapting to the acidic conditions of the thylakoid lumen (Table 4-2)