Plant Molecular Biology 53: 643–654, 2004.
© 2004 Kluwer Academic Publishers. Printed in the Netherlands.
Physcomitrella patens as a model for the study of chloroplast protein
transport: conserved machineries between vascular and non-vascular
Nancy Rosenbaum Hofmann and Steven M. Theg
Section of Plant Biology, University of California at Davis, One Shields Avenue, Davis, CA 95616, USA (
for correspondence; e-mail email@example.com)
Received 18 July 2003; accepted in revised form 23 October 2003
Key words: chloroplast biogenesis, Physcomitrella patens, protein targeting
A single general import pathway in vascular plants mediates the transport of precursor proteins across the two
membranes of the chloroplast envelope, and at least four pathways are responsible for thylakoid protein targeting.
While the transport systems in the thylakoid are related to bacterial secretion systems, the envelope machinery is
thought to have arisen with the endosymbiotic event and to be derived, at least in part, from proteins present in the
original endosymbiont. Recently the moss Physcomitrella patens has gained worldwide attention for its ability to
undergo homologous recombination in the nuclear genome at rates unseen in any other land plants. Because of this,
we were interested to know whether it would be a useful model system for studying chloroplast protein transport.
We searched the large database of P. patens expressed sequence tags for chloroplast transport components and
found many putative homologues. We obtained full-length sequences for homologues of three Toc components
from moss. To our knowledge, this is the ﬁrst sequence information for these proteins from non-vascular plants.
In addition to identifying components of the transport machinery from moss, we isolated plastids and tested their
activity in protein import assays. Our data indicate that moss and pea (Pisum sativum) plastid transport systems
are functionally similar. These ﬁndings identify P. patens as a potentially useful tool for combining genetic and
biochemical approaches for the study of chloroplast protein targeting.
Abbreviations: EST, expressed sequence tag; LHCP, light-harvesting chlorophyll-binding protein; NIBB, National
Institute for Basic Biology; OE17, 17 kDa subunit of the oxygen-evolving complex; PC, plastocyanin; PEP,
Physcomitrella EST Programme; SPP, stromal processing peptidase; SRP, signal recognition particle; Tat, twin-
arginine translocation; Tic, translocon at the inner membrane of the chloroplast envelope; Toc, translocon at the
outer membrane of the chloroplast envelope; TPP, thylakoid processing peptidase; TPR, tetratricopeptide repeat
Unlike other plants, the moss Physcomitrella patens
has been shown to undergo homologous recombina-
tion in the nuclear genome at high rates (Kammerer
and Cove, 1996; Schaefer and Zyrd, 1997). This has
made it possible to directly target genes for knock-
out (Girke et al., 1998; Stepp et al., 1998; Girod
et al., 1999; Hofmann et al., 1999; Imaizumi et al.,
2002; Meiri et al., 2002) and should allow for allele
replacement. The power of homologous recombina-
tion for functional genetic analysis has been amply
demonstrated in yeast and bacteria and is beginning
to be explored in P. patens. Historically, most stud-
ies of chloroplast protein transport have used Pisum
sativum and Spinacia oleracea as model organisms.
While pea and spinach can be exploited readily for
biochemical studies, they are not genetically tractable.
Arabidopsis thaliana is a good model for genetic ana-
lyses (although it is still not possible to perform direct
allele replacement), but it is not especially amenable
to biochemical study. Protocols have been developed