pH Sensitivity of the GTPase Toc33 as a Regulatory Circuit for
Protein Translocation into Chloroplasts
Tihana Bionda1,3, Patrick Koenig2,3, Mislav Oreb1, Ivo Tews2and Enrico Schleiff1,*
1JWGU Frankfurt am Main, Cluster of Excellence Macromolecular Complexes, Center of Membrane Proteomics,
Department of Biosciences, Molecular Cell Biology, Max-von-Laue Str. 9, D-60439 Frankfurt, Germany
2Heidelberg University Biochemistry Centre, Im Neuenheimer Feld 328, D-69120 Heidelberg, Germany
The properties of membrane-embedded GTPases are
investigated to understand translocation of preprotein across
the outer envelope of chloroplasts. The homo- and hetero-
dimerization events of the GTPases had been established
previously. We show that the hydrolytic activity of the
GTPase Toc33 is pH insensitive in the homodimeric
conformation but has a bell-shaped pH optimum in the
monomeric conformation. Further, Toc33 GTPase homodi-
merization and protein translocation into chloroplasts are pH
sensitive as well. pH sensitivity might serve to regulate
translocation; alternatively, the documented pH sensitivity
might reflect a mechanistic requirement for GTPase silencing
during translocation as the GTPase switches between homo-
and heterodimeric conformations.
Keywords: Arabidopsis thaliana and Pisum sativum —
Dimerization — GTPase — pH sensitivity — Protein
translocation — TOC.
Abbreviations: pSSU, precursor of the small subunit
of Rubisco; Toc, translocon at the outer envelope of
chloroplasts; Toc33N, soluble GTPase domain of Toc33.
Protein–protein interactions are the basis of many
cellular functions (Frieden 1971, Stelzl and Wanker 2006);
many regulatory events and developmental changes affect
the assembly of proteinaceous complexes. These actions are
dependent on the protein concentrations as well as their
mobility. Applied to the membrane-bound GTPases of the
Toc complex that are restricted to two dimensions
(Tserkovnyak and Nelson 2006) and to a slower rate of
diffusion (Lenaz 1987), dissociations occur at a different
time scale and in a different concentration range compared
with soluble proteins.
The Toc complex (translocon at the outer envelope of
chloroplasts) is present in the outer envelope of chloroplasts
(Gutensohn et al. 2006, Oreb et al. 2008) and consists of
three core components, namely the pore-forming Toc75 and
the two GTPases Toc33 and Toc159, as well as further,
dynamically associated proteins (Inaba and Schnell 2008).
Toc33 and Toc159 are GTPases which regulate precursor
protein perception and their transport to the translocation
pore, possibly also being involved in the translocation
process itself (Inaba and Schnell 2008, Oreb et al. 2008). It
has been discussed that the function of Toc33 (also named
Toc34 in Pisum sativum) and Toc159 is regulated by
dimerization of the GTPase domains (Jelic et al. 2003,
Becker et al. 2004, Ivanova et al. 2004). Toc33 is able to
homodimerize (Koenig et al. 2008a), but during the
translocation a heterodimeric complex has to be formed,
presumably with both GTPases in a GTP-loaded state
(Becker et al. 2004). However, the exact mechanism and
physiological relevance of these dimerization events are not
The activity of the GTPase Toc33 from Arabidopsis
and Toc34 from pea has been studied in great detail (Jelic
et al. 2002, Yeh et al. 2007, Reddick et al. 2007, Koenig
et al. 2008a, Koenig et al. 2008b). Toc33 has a moderate
hydrolysis rate even in the absence of other stimulating
factors. The optimal pH for the hydrolysis reaction was
reported to be above pH 8 (Reddick et al. 2007), despite the
fact that the cytoplasmic pH is about pH 7.0 in roots and
pH 7.5 in leaves (Kurkdjian and Guern 1989, Felle 2001).
However, cytosolic pH changes transiently in response to
light in a biphasic manner (Remis et al. 1988) with an
amplitude of pH 0.3 (Felle and Betl 1986), as well as in
response to temperature changes (Aducci et al. 1982) and
other environmental factors (Kurkdjian and Guern 1989,
Felle 2001). We explore here whether external pH changes
might have a regulatory effect on Toc33/34 function and
We first investigated the multiple turnover import of
precursor proteins into chloroplasts, a process that has
previously been described as pH sensitive (Grossman et al.
1982). We incubated chloroplasts with the precursor of the
*Corresponding author: E-mail, email@example.com; Fax, þ49-69-798-29286.
3These authors contributed equally to this work.
Plant Cell Physiol. 49(12): 1917–1921 (2008)
doi:10.1093/pcp/pcn171, available online at www.pcp.oxfordjournals.org
? The Author 2008. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.
All rights reserved. For permissions, please email: firstname.lastname@example.org
by guest on October 30, 2015
small subunit of Rubisco (pSSU) and quantified the binding
and import (Fig. 1A). In contrast to the previous report
(Grossman et al. 1982), we observed a reduction of import
capacity at pH 8.5 when compared with pH 7.9 (Fig. 1A).
This discrepancy might be caused by the experimental set-
up, in which Grossmann and co-workers analyzed the
translocation of proteins translated from high molecular
weight pea RNA by precipitation of soluble extracts after
chloroplast lysis, whereas we quantified the maturation of
the precursor protein pSSU. Considering both bound and
imported protein as total bound protein (because the
imported protein had to be bound prior to import and is
provided in substochiometric amounts when compared with
translocation sites; Fig. 1B), we observed that binding is
slightly pH dependent, with maximal binding at pH 8.2.
However, the binding was just 2-fold enhanced at pH 8.2
Import - backgr.
Import (fraction of bound)
total bound (% input)
7.0 7.3 7.6
buffer with the indicated pH under import conditions for the indicated time period. After re-isolation of organelles, the amount of processed
precursor was quantified and is presented as a percentage of the translation product used. (B) The sum of the quantities of bound and
imported protein was considered as total bound protein, which is given as percentage of the translation product used. (C) Radioactively
labeled pSSU was incubated with chloroplasts in buffer at pH 7.9 in the absence of ATP and at 48C. After re-isolation of organelles, import
at the indicated pH was initiated by addition of ATP. The import is shown as a fraction of the initially bound protein. For (A–C), values
show the average of at least three independent experiments. (D) The import rate for pSSU (gray bar) and pOE33 (white bar) was calculated
from experiments as shown in (C) at different environmental pH values. The standard deviation of at least three independent results is
shown. The lines represent a least-squares analysis using a Gaussian distribution (gray for pSSU, black for pOE33).
pH sensitivity of precursor binding and import into chloroplasts. (A) Radioactively labeled pSSU was incubated with chloroplasts in
1918 pH sensitivity of Toc33 action
by guest on October 30, 2015
when compared with pH 7.0. To investigate the direct
influence of the pH on the translocation event, precursor
proteins were incubated with chloroplasts at 48C at pH 7.9
in the absence of ATP. In this way, the pH sensitivity of the
association of the precursor with the translocon was
omitted. After re-isolation of plastids in buffer with
different pH, the translocation event was initiated at 258C
by addition of ATP. Normalization of the imported protein
to the initially bound protein (Fig. 1C) shows a pH
dependence of the import rate, with the highest rate at
approximately pH 8.0 (determined by single Gaussian
analysis, Fig. 1D, gray bars and gray line). The rate at
pH 8 is about 5-fold enhanced when compared with that at
pH 7.0. To generalize the observation, we analyzed the pH
dependence of the import rate of the precursor of the
oxygen-evolving complex subunit of 33kDa (pOE33)
translated in rabbit reticulocyte lysate. Again, analyzing
the translocation of initially bound precursor, we observed
an optimal import rate at about pH 8.1 (Fig. 1D, white bars
and black line). Therefore, binding and import of precursor
proteins appeared to be pH sensitive; however, binding was
affected to a lower extent.
Homodimerization has been considered as a regulatory
event of the Toc GTPases (Inaba and Schnell, 2008, Oreb
et al. 2008). We therefore investigated the pH dependence of
the dimerization of Toc33N, representing the cytosolically
exposed G-domain excluding the transmembrane and the
intermembrane space region, by size exclusion chromato-
graphy (Koenig et al. 2008b). Comparing the dissociation
constants extracted from the size exclusion experiments by
Equation 1 (Fig. 2A) at pH 8 with that at pH 6, pH 7 or pH
9 reveals increases by a factor of 9.7, 7.8 and 2.5,
fraction dimer [%]
Hydrolysis [fold max]
chromatography. The dimeric portion at a given pH is shown and lines represent least-squares fit analysis to Equation 1. (B) Toc33N(white
circle) or the G-domain of Toc159 (gray circle) was incubated in a 10ml final reaction volume with [a-32P]GTP to initiate multiple turnover
GTP hydrolysis. After 20min nucleotides were separated on PEI-cellulose plates (Merck, Darmstadt, Germany) using 600mM NaH2PO4at
pH 3.4 as the developing solvent. Radioactivity was quantified and normalized to the highest hydrolysis observed. The lines represent a
least-squares analysis using a Gaussian distribution (black for Toc33Nand gray for the Toc159 G-domain). (C) The apparent single turnover
rate values was determined. Lines represent the least square fit analysis according to the equation described by Koenig et al. (2008a) and
(D) gives the obtained rate constants for the monomeric (black) or dimeric form (white) for the reactions at different pH values.
pH sensitivity of Toc33. (A) The affinity for homodimerization of atToc33Nwas determined at different pH values by size exclusion
pH sensitivity of Toc33 action1919
by guest on October 30, 2015
respectively. Similarly, GTPase multiple turnover measure-
ments show that Toc33Nexhibits optimal activity at pH 8.2
(Fig. 2B, white dots, black line) under experimental
conditions ensuring monomeric protein species (Reddick
et al. 2007). The rapid decay of activity below pH 7.5 is
somewhat atypical for GTPases, because the apparent pKa
of the ?-phosphate of protein-bound GTP was found to be
in the acidic range (Schweins et al. 1997) and usually
coincides with the highest activity range of the GTPase (Seo
et al. 2003). The same observation was made for the G-
domain of Toc159, which exhibits optimal activity at pH 8.0
(Fig. 2B, gray dots, gray line). To extend this observation,
we performed single turnover experiments with Toc33Nat
different pHs and using protein concentrations across the
stochiometric point of dimer formation. The sensitivity of
GTP hydrolysis to receptor concentration increases when
the pH is shifted away from the optimum (Fig. 2C), with
hydrolysis being different at low concentrations but very
similar at high concentrations. Determination of the
hydrolysis rate constants (Fig. 2D, according to the
equation given in Koenig et al. 2008a) reveals a difference
in hydrolysis by the monomeric (black) and dimeric (white)
receptor under physiological conditions (pH 7.5) of about
20-fold, whereas both rates are comparable at optimal pH.
This analysis further reveals that the hydrolysis rate of the
dimeric species is not pH sensitive in the range investigated
(Fig. 2D, white), whereas the rate of the monomeric
receptor is pH sensitive (Fig. 2D, black) in a manner
comparable with that observed while performing multiple
Toc33Ndimerizes in a pH-sensitive manner (Fig. 2A).
This is known to occur for a number of enzymes, e.g.
dihydrofolate reductase, b-galactosidase or pigeon liver
malic enzyme (Chang et al. 1988, Gallagher and Huber
1997, Me ´ jean et al. 2001). The monomeric, but not the
homodimeric receptor exhibits an atypical pH-dependent
GTP hydrolysis. There could be a mechanistic or a
regulatory explanation for our observation. The dual
regulation of dimerization and GTP hydrolysis of the Toc
receptor by the environmental pH might serve as a sensor
for functional switches of the complex altering translocation
efficiency (Fig. 1). This is likely since the pH at the cytosolic
face of the chloroplasts can range between 6.5 and 8
(Kurkdjian and Guern 1989, Felle 2001), with lower values
in non-green tissues, and pH changes are often linked to
other physiological signals. In this way, the pH sensitivity
observed would constitute a control loop to the Toc
complex and thereby represent a regulatory circuit. An
alternative explanation is based on the documented transi-
tion between homodimeric (Koenig et al. 2008a) and
heterodimeric (Becker et al. 2004) interaction of G-domains
of the Toc GTPases during protein translocation. The
reduced capacity of Toc33 GTP hydrolysis at physiological
pH in the monomeric state ensures that hydrolysis occurs
preferentially in the dimeric state of the receptor. The
dependence of GTP hydrolysis is documented for the
homodimer (Fig. 2), and whether this is also true for
heterodimerization remains to be established. The observed
sensitivity of dimerization does not contradict such a model
since the affinity [proportional to ln(1/KD)] is reduced 52-
fold when shifting the pH from the optimum pH 8 to pH 7.
Hence the pH sensitivity would reflect a physical property
of the GTPase required for its mechanistic action during
protein translocation. Future research will have to distin-
guish between both possibilities.
Materials and Methods
Isolation of chloroplasts, in vitro production of radioactively
labeled precursor proteins and in vitro import into chloroplasts
were previously described (Schleiff et al. 2002). Mutants of atToc33
were generated by conventional PCR using AtTOC33 (bp 1–753)
as template (Jelic et al. 2003). Constructs were cloned into pET21d
(Novagen, Madison, WI, USA) to generate protein products with a
C-terminal hexahistidine tag, and recombinant His-tagged proteins
were purified as described (Koenig et al. 2008a). The DNA
sequence coding for amino acids 740–1,048 of psToc159 was
amplified by standard PCR from full-length cDNA and inserted
between the NcoI and XhoI sites of pET21d. The construct was
expressed in Escherichia coli and purified as described for Toc33
(Jelic et al. 2003).
For multiple turnover, a MES/Tris buffer system was used.
For buffers up to pH 7.0, 20mM MES was adjusted with Tris base
to the desired pH, and vice versa for buffers above pH 7.0. The
reaction mixture contained 20mM of this buffer, 1mM MgCl2,
1mM dithiothreitol, 50mM NaCl and 0.03mg ml–1Toc33N.
Single-turnover GTP hydrolysis was performed and analyzed
according to established protocols (Koenig et al. 2008a). For
molecular weight determinations, 200ml of purified protein at
various concentrations was loaded on a Superdex75 HR 10/300
column and analyzed as described (Koenig et al. 2008b). The
dissociation constant was determined according to Equation 1
(D¼dimeric fraction, T¼total amount of protein)
KDþ 4 ? T
ð Þ ?
KD? KDþ 8 ? T
The Volkswagenstiftung (to E.S.); the DFG (SFB807-
P17 to E.S.).
Aducci, P., Federico, R., Carpinelli, G. and Podo, F. (1982) Temperature
dependence of intracellular pH in higher plant cells. Planta 156: 579–582.
Becker, T., Jelic, M., Vojta, A., Radunz, A., Soll, J. and Schleiff, E. (2004)
Preprotein recognition by the Toc complex. EMBO J. 23: 520–530.
Chang, G.G., Huang, T.M. and Chang, T.C. (1988) Reversible dissociation
of the catalytically active subunits of pigeon liver malic enzyme. Biochem.
J. 254: 123–130.
1920pH sensitivity of Toc33 action
by guest on October 30, 2015
Felle, H.H. (2001) pH: signal and messenger in plant cells. Plant Biol. 3: Download full-text
Felle, H. and Betl, A. (1986) Light-induced cytoplasmic pH changes and
their interrelation to the activity of the electrogenic proton pump in
Riccia fluitans. Biochim. Biophys. Acta 848: 176–182.
Frieden, C. (1997) Protein–protein interaction and enzymatic activity.
Annu. Rev. Biochem. 40: 653–696.
Gallagher, C.N. and Huber, R.E. (1997) Monomer–dimer equilibrium of
uncomplemented M15 b-galactoside from Escherichia coli. Biochemistry
Grossman, A.R., Bartlett, S.G., Schmidt, G.W., Mullet, J.E. and
Chua, N.H. (1982) Optimal conditions for post-translational uptake of
proteins by isolated chloroplasts. In vitro synthesis and transport of
6-bisphosphatase. J. Biol. Chem. 257: 1558–1563.
Gutensohn, M., Fan, E., Frielingsdorf, S., Hanner, P., Hou, B., Hust, B.,
et al. (2006) Toc, Tic, Tat et al.: structure and function of
protein transport machineries in chloroplasts. J. Plant. Physiol. 163:
Inaba, T. and Schnell, D.J. (2008) Protein trafficking to plastids: one theme,
many variations. Biochem. J. 413: 15–28.
Ivanova, Y., Smith, M.D., Chen, K. and Schnell, D.J. (2004) Members of
the Toc159 import receptor family represent distinct pathways for protein
targeting to plastids. Mol. Biol. Cell. 15: 3379–3392.
Jelic, M., Sveshnikova, N., Motzkus, M., Ho ¨ rth, P., Soll, J. and Schleiff, E.
(2002) The chloroplast import receptor Toc34 functions as preprotein-
regulated GTPase. Biol. Chem. 383: 1875–1883.
Jelic, M., Soll, J. and Schleiff, E. (2003) Two Toc34 homologues with
different properties. Biochemistry 42: 5906–5916.
Koenig, P., Oreb, M., Ho ¨ fle, A., Kaltofen, S., Rippe, K., Sinning, I., et al.
(2008a) The GTPase cycle of the chloroplast import receptors Toc33/
Toc34: implications from monomeric and dimeric structures. Structure
Koenig, P., Oreb, M., Rippe, K., Muhle-Goll, C., Sinning, I., Schleiff, E.,
et al. (2008b) On the significance of TOC–GTPase homodimers. J. Biol.
Chem. 283: 23104–23112.
Kurkdjian, A. and Guern, J. (1989) Intracellular pH: measurement and
importance in cell activity. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40:
Lenaz, G. (1987) Lipid fluidity and membrane protein dynamics. Biosci.
Rep. 7: 823–837.
Me ´ jean, A., Bodenreider, C., Schuerer, K. and Goldberg, M.E. (2001)
Kinetic characterization of the pH-dependent oligomerization of R67
dihydrofolate reductase. Biochemistry 40: 8169–8179.
Oreb, M., Tews, I. and Schleiff, E. (2008) Policing Tic ‘n’ Toc, the doorway
to chloroplasts. Trends Cell Biol. 18: 19–27.
Reddick, L.E., Vaughn, M.D., Wright, S.J., Campbell, I.M. and
Bruce, B.D. (2007) In vitro comparative kinetic analysis of the
chloroplast Toc GTPases. J. Biol. Chem. 282: 11410–11426.
Remis, D., Bulychev, A.A. and Kurella, G.A. (1988) Photo-induced pH
changes in the vicinity of isolated Peperomia metallica chloroplasts.
J. Exp. Bot. 39: 633–640.
Schleiff, E., Motzkus, M. and Soll, J. (2002) Chloroplast protein import
inhibition by a soluble factor from wheat germ lysate. Plant Mol. Biol. 50:
Schweins, T., Scheffzek, K., Assheuer, R. and Wittinghofer, A. (1997) The
role of the metal ion in the p21ras catalysed GTP-hydrolysis: Mn2þ
versus Mg2þ. J. Mol. Biol. 266: 847–856.
Seo, H.S., Jeong, J.Y., Nahm, M.Y., Kim, S.W., Lee, S.Y. and Bahk, J.D.
(2003) The effect of pH and various cations on the GTP hydrolysis of rice
heterotrimeric G-protein alpha subunit expressed in Escherichia coli.
J. Biochem. Mol. Biol. 36: 196–200.
Stelzl, U. and Wanker, E.E. (2006) The value of high quality protein–
protein interaction networks for systems biology. Curr. Opin. Chem. Biol.
Tserkovnyak, Y. and Nelson, D.R. (2006) Conditions for extreme
sensitivity of protein diffusion in membranes to cell environments.
Proc. Natl Acad. Sci. USA 103: 15002–15007.
Yeh, Y.H., Kesavulu, M.M., Li, H.M., Wu, S.Z., Sun, Y.J., Konozy, E.H.,
et al. (2007) Dimerization is important for the GTPase activity of
chloroplast translocon components atToc33 and psToc159. J. Biol.
Chem. 282: 13845–138453.
(Received October 20, 2008; Accepted October 31, 2008)
pH sensitivity of Toc33 action1921
by guest on October 30, 2015