Transcriptome analysis of the aphid bacteriocyte,
the symbiotic host cell that harbors an endocellular
mutualistic bacterium, Buchnera
Atsushi Nakabachi*†‡, Shuji Shigenobu§, Naoko Sakazume¶, Toshiyuki Shiraki¶, Yoshihide Hayashizaki¶, Piero Carninci¶,
Hajime Ishikawa?, Toshiaki Kudo*, and Takema Fukatsu‡
*Environmental Molecular Biology Laboratory and¶Genome Science Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan;‡Institute for
Biological Resources and Functions, National Institute of Advanced Industrial Science and Technology, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan;
§Okazaki Institute for Integrative Bioscience, National Institute for Basic Biology, National Institutes of Natural Sciences, Higashiyama, Myodaiji,
Okazaki 444-8787, Japan; and?School of General Education, University of the Air, Wakaba, Mihama, Chiba 261-8586, Japan
Edited by Nancy A. Moran, University of Arizona, Tucson, AZ, and approved February 24, 2005 (received for review December 5, 2004)
Aphids possess bacteriocytes, cells specifically differentiated to
harbor obligatory mutualistic bacteria of the genus Buchnera,
which have lost many genes that are essential for common bac-
terial functions. To understand the host’s role in maintaining the
symbiotic relationship, bacteriocytes were isolated from the pea
aphid, Acyrthosiphon pisum, and the host transcriptome was
investigated by using EST analysis and real-time quantitative
RT-PCR. A number of genes were highly expressed specifically in
the bacteriocyte, including (i) genes for amino acid metabolism,
including those for biosynthesis of amino acids that Buchnera
cannot produce, and those for utilization of amino acids that
Buchnera can synthesize; (ii) genes related to transport, including
genes for mitochondrial transporters and a gene encoding Rab, a
G protein that regulates vesicular transport; and (iii) genes for
putative lysozymes that degrade bacterial cell walls. Significant
in the exchange of amino acids between the host aphid and
Buchnera, the key metabolic process in the symbiotic system.
Conspicuously high expression of ii and iii shed light on previously
unknown aspects of the host–Buchnera interactions in the symbi-
EST ? quantitative RT-PCR
makes them notorious agricultural pests. Nutritional aspect of
this fecundity is based on an intimate symbiotic association with
have dozens of bacteriocytes, specialized cells for harboring
intracellular symbiotic bacteria of the genus Buchnera that
belong to the ?-subdivision of the Proteobacteria (1). Physiolog-
essential amino acids (amino acids that metazoa cannot synthe-
size; tryptophan, lysine, methionine, phenylalanine, threonine,
valine, leucine, isoleucine, arginine, and histidine) (2–6) and
riboflavin (vitamin B2) (7) that aphids cannot synthesize and are
scarce in the phloem sap diet (8, 9). Since the initial infection
?100 million years ago (10), Buchnera have been subjected to
strict vertical transmission through host generations, and the
mutualism between the host and Buchnera has reached to
such an extent that neither can reproduce in the absence of the
other (2, 3).
Recent studies on the complete genome sequences of three
lineages of Buchnera (those in association with Acyrthosiphon
pisum, Schizaphis graminum, and Baizongia pistaceae) provided
comprehensive knowledge on potential functions of Buchnera in
the symbiotic system (11–13). The sequenced Buchnera genomes
were only 0.61–0.65 Mb in size and encoded 510–570 proteins.
The gene composition confirmed and extended the previous
physiological data that Buchnera are able to provide their hosts
lthough aphids feed only on a nutritionally poor diet,
phloem sap, they show explosive reproductivity, which
with essential amino acids and riboflavin. Whereas massive
genome reduction is generally found in endocellular bacteria of
endocellular parasites retain only a few, if any, genes for
of the Buchnera genomes clearly reflects the mutualistic nature
of Buchnera for the host aphids (11).
The genome studies of Buchnera have provided not only
genetic insights into how the symbiotic system operates but also
the genomic information of the host side. Whereas Buchnera
retain genes for the biosynthesis of nutrients that are required by
the hosts, they lack many genes that seem to be essential for their
own living. For example, most genes involved in biosyntheses of
nonessential amino acids (amino acids that metazoa can synthe-
size; alanine, asparagine, aspartic acid, cysteine, glutamic acid,
glutamine, glycine, proline, serine, and tyrosine) and phospho-
lipids are lost, implying that Buchnera can synthesize neither
nonessential amino acids nor even their own cell membrane.
Most genes encoding transcriptional regulators are also missing,
suggesting that Buchnera may scarcely be able to regulate their
own metabolic and cellular activities (11). It appears likely that
these and other incomplete aspects of Buchnera functions must
be compensated for by the activities of the host bacteriocyte.
In an effort to understand the host’s role in this symbiotic
system, we assessed the mRNA population of the host bacterio-
cyte of the pea aphid, Acyrthosiphon pisum, by EST analysis and
real-time quantitative RT-PCR.
Materials and Methods
Aphids. Strain ISO, a parthenogenetic clone of the pea aphid
Acyrthosiphon pisum, was used in this study. Diagnostic PCR has
verified that this clone is free of secondary symbionts such as
R-type, T-type, U-type, Rickettsia, Wolbachia, Spiroplasma, or
faba at 15°C in a long-day regime of 16 h of light and 8 h of dark.
Parthenogenetic apterous adults (12–15 days old) were used for
This paper was submitted directly (Track II) to the PNAS office.
Freely available online through the PNAS open access option.
transporter 2; GCVT, glycine cleavage system T protein; GCDH, glutaryl-CoA dehydroge-
nase; PSAT, phosphoserine aminotransferase; ANT2, ADP?ATP translocase; PC, inorganic
phosphate cotransporter; OT, mitochondrial oxaloacetate transport protein; AS, ATP syn-
thase subunit c; GC, mitochondrial glutamate carrier; Rp n, ribosomal protein n; LSZ,
lysozyme; NCBI, National Center for Biotechnology Information.
Data deposition: The sequences reported in this paper have been deposited in DNA Data
Bank of Japan database (accession nos. BP535536–BP537955).
†To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
© 2005 by The National Academy of Sciences of the USA
April 12, 2005 ?
vol. 102 ?
no. 15 ?
RNA Preparation from Bacteriocytes. The aphids were dissected in
buffer A (20) on 1% agarose plate. Bacteriocytes freed from the
insect body were collected with a micropipette and immediately
lysed in TRIzol reagent (GIBCO?BRL). Total RNA was ex-
tracted according to the manufacturer’s instructions. In total,
60,000–80,000 bacteriocytes collected from ?2,000 insects were
used for the library construction.
cDNA Library Construction. Full-length cDNA library was gener-
ated by cap-trapper method (21, 22) with a slight modification.
Oligo(dT) primer for the first-strand cDNA synthesis:
TTTTTTTTVN-3?. Double-stranded linkers for the second-
strand cDNA synthesis: GN5 linker and N6 linker (molar ratio
of N6:GN5 ? 1:4) (23). Second-strand cDNA was digested with
BamHI and XhoI, and ligated to lambda FLC-I vector, which
carries two loxP sites (24). After amplification in C600 cells, the
phage DNA was converted into plasmid with Cre recombinase.
Sequence of Clones. Sequencing reactions were performed by
usingABI BIG DYE TERMINATOR V3.1(AppliedBiosystems)along
with the M13 forward primer: 5?-TGTAAAACGACGGC-
CAGT-3?. Reaction products were run on an ABI 3700 capillary
sequencer (Applied Biosystems).
Sequence Processing and Clustering. The raw chromatogram files
were base-called with PHRED (25, 26), and the vector sequence
was masked with CROSSMATCH. Only sequences that exhibited
?100 bases with a PHRED quality value of ?20 were used for the
subsequent analyses. Contaminants derived from the vector,
Escherichia coli genome, or Buchnera genome were detected by
using BLASTN (27) with high-stringency parameters (E value of
?1.0 ? 10?20, cost to open a gap ? 1, cost to extend a gap ? 3,
ESTs were deposited in the DNA Data Bank of Japan database
(accession nos. BP535536–BP537955). These sequences were
assembled by using the CAP3 program (28).
Sequence Analysis and Annotation. The combined set of the con-
sensus sequences was analyzed as a putative ‘‘unigene set.’’ The
against protein databases. First, each unigene was compared
against the fly protein database [GADFLY version 3.1, Berkeley
Drosophila Genome Project, which can be accessed at www.
fruitfly.org] by using BLASTX, and automatically assigned a gene
annotation of the top hit (E ? 1.0 ? 10?10), regarding it as a
putative fly homolog. Gene Ontology (GO) classifications of the
corresponding fly homologs were transferred from FlyBase
(which can be accessed at http:??flybase.org, FlyBase Consor-
tium, 2003). Second, similarity search was performed against the
nonredundant protein database [www.ncbi.nlm.nih.gov, Na-
tional Center for Biotechnology Information (NCBI), March
2004]. Unigenes with no apparent fly homolog were annotated
in this step if similar proteins were found. In addition, if the
BLAST bitscore of top hit was significantly higher than that of fly
homolog, the annotation was replaced. These annotation pro-
cedures were automated by the custom Ruby script. Finally, the
gene assignments were inspected manually and annotated by
using a specifically developed annotation interface. All of the
information on our unigenes with their annotations is available
Real-Time Quantitative RT-PCR. RNA was isolated from whole
bodies and bacteriocytes of 12- to 15-day-old parthenogenetic
apterous adults by using TRIzol reagent, followed by RNase-
free DNase I treatment. Each whole-body sample and bacte-
riocyte sample derived from one individual and a batch of
bacteriocytes that were collected from ?10 individuals, re-
spectively. First-strand cDNAs were synthesized by using
pd(N)6 primer and the First-Strand cDNA synthesis kit (Am-
ersham Biosciences, Piscataway, NJ). External standards were
constructed by PCR using whole-body cDNA and gene-specific
primer sets that were designed referring to CAP3 base confi-
dence scores of contigs generated by the EST analysis (Table
3, which is published as supporting information on the PNAS
web site). Quantification was performed with the LightCycler
instrument and FastStart DNA Master SYBR green I kit
(Roche Diagnostics, Mannheim, Germany). Running param-
eters were 95°C for 10 min, followed by 40 cycles of 95°C for
15 s, 55°C for 10 s, and 72°C for 4 s. Signal intensity was
measured at the end of each elongation phase unless otherwise
stated. When primer dimers were detected in preliminary
experiments, signal intensity was measured at an additional
step after elongation (Table 3). Results were analyzed by using
the LIGHTCYCLER software, version 3.0 (Roche Diagnostics),
and relative expression levels were normalized to mRNA for
ribosomal protein (Rp)L7. All quantitative RT-PCRs were
performed in triplicate. Figures thus show each value as the
mean ? SE of 10 independent experiments of 10 independent
samples (n ? 10), with each one of them represented by the
mean of three separate quantitative RT-PCRs. Statistical
analyses were performed by using the Mann–Whitney U test.
Levels of transcripts for neither ribosomal protein S3A nor
elongation factor 1? showed significant difference between the
bacteriocyte and the whole body (P ? 0.05), proving appro-
priateness of the quantification system (data not shown).
Results and Discussion
Construction of the Bacteriocyte cDNA Library. The aphid bacterio-
cyte contains a large number of Buchnera cells in the cytoplasm.
Thus, to construct a cDNA library of the host bacteriocyte,
contaminants of the bacterial symbionts must be removed. First,
we attempted to synthesize the host cDNAs selectively by using
an oligo(dT) primer that targets the 3? poly(A) tail of eukaryotic
mRNAs. However, this conventional method did not work,
because the genome of Buchnera is highly AT-rich (11–13), and
thus, allowed frequent annealing of the oligo(dT) primer to their
the conventional method, 75% of clones were of Buchnera origin
(data not shown). Therefore, we adopted an alternative method,
the cap trapper that can select cDNAs derived from eukaryotic
mRNAs with the 5? cap structure (21, 22). Application of this
to an acceptable level (9.9%, see below).
Generation and Assembly of Bacteriocyte ESTs.Intotal,2,870cDNA
clones were sequenced from the 5? end. After removal of
low-quality sequences and contaminants derived from the vector
and E. coli genome, 2,602 high-quality sequences were obtained.
Of these sequences, 257 ESTs matched to the genome of
Buchnera aphidicola str. APS (GenBank accession no.
NC?002528). After removal of the Buchnera sequences, 2,345
ESTs of 656.6 bp average length were clustered by using the CAP3
program. The ESTs were assembled into 246 contigs with 91
ESTs remaining as singlets. Further analyses were performed by
using these 337 nonredundant sequences as a putative unigene
Similarity Search and Annotation. The unigene set was subjected
to BLASTX similarity searches. Of 337 unigenes, 233 (183
contigs and 50 singlets, 69.1%) and 244 (190 contigs and 54
singlets, 72.4%) showed significant similarities (E ?1.0 ?
10?10) to protein-encoding genes in the fly database (GADFLY,
version 3.1) and the NCBI nonredundant database, respec-
tively. These BLAST hits were used to annotate the unigenes as
described in Materials and Methods. The unigenes were ranked
www.pnas.org?cgi?doi?10.1073?pnas.0409034102Nakabachi et al.
by the number of corresponding EST clones (Table 1 and Table
4, which is published as supporting information on the PNAS
Comparative Analysis of Aphid Transcriptomes. To obtain an over-
view of the aphid bacteriocyte transcriptome (ESTBC), we
compared EST population in ESTBCwith those of other aphid
transcriptomes. As of May 2004, two aphid EST sets were
available in the dbEST depository of GenBank; one is from the
pea aphid whole-body library (ESTWB) containing 1,071 se-
quences, and the other is from a whole-body library of the brown
citrus aphid, Toxoptera citricida (ESTTC), containing 4,267 se-
without using PCR, which might bias cDNA populations. The
data were extracted, analyzed, and annotated by the same
method. ESTWBand ESTTCwere assembled into 742 unigenes
(151 contigs plus 591 singlets) and 2,176 unigenes (465 contigs
plus 1,711 singlets), respectively. Sequence comparisons dem-
onstrated that 25.2% (70 contigs plus 15 singlets) and 36.5% (97
contigs plus 26 singlets) of unigenes in ESTBCwere similar to
those in ESTWBand ESTTC, respectively.
Selective Up-Regulation of Genes Related to Amino Acid Metabolism,
Transport, and Defense Response in the Bacteriocyte. The CAP3
program assembled 2,345 ESTs of ESTBCinto 337 unigenes,
indicating 85.6% redundancy. This value is much higher than
that of ESTWB(30.7%) and ESTTC(49.0%), suggesting that the
bacteriocyte expresses relatively small number of genes at
conspicuously high levels. The numbers of EST clones under
each of the GO categories were compared among ESTBC,
ESTWB, and ESTTC(Table 2; a complete list is available upon
request). Among these categories, ‘‘amino acid metabolism
(GO:0006520),’’ ‘‘transport (GO:0006810),’’ and ‘‘defense re-
sponse (GO:0006952)’’ contained significantly higher percent-
age of ESTs in ESTBCthan in the whole-body transcriptomes,
ESTWB and ESTTC (P ? 0.001, Fisher’s exact test with
Bonferroni correction for multiple comparisons).
Up-Regulated Genes Involved in Amino Acid Metabolism. ESTBC
contained 164 EST clones relevant to amino acid metabolism.
Despite the relative abundance of EST clones related to amino
acid metabolism in ESTBC(Table 2), these ESTs corresponded
to only 13 unigenes, which was less than those in ESTWB(18
unigenes) and ESTTC (25 unigenes). To further verify the
abundance of transcripts detected in ESTBC, real-time quanti-
tative RT-PCR was carried out. As representatives, unigenes for
glutamine synthetase 2 (GS2; 55 clones) (EC 184.108.40.206), cationic
amino acid transporter 2 (CAT2; 27 clones), glycine cleavage
system T protein (GCVT; 24 clones) [an aminomethyl trans-
ferase (EC 220.127.116.11), which is a part of the glycine cleavage
multienzyme complex catalyzing the degradation of glycine],
Henna (13 clones)[an enzyme with activities of phenylalanine
4-monooxygenase (EC 18.104.22.168) and tryptophan 5-monooxy-
genase (EC 22.214.171.124), which are involved in catabolisms of
L-phenylalanine and L-tryptophan, respectively], glutaryl-CoA
dehydrogenase (GCDH; 6 clones) (EC 126.96.36.199; an enzyme that
catalyzes the oxidative decarboxylation of glutaryl-CoA, which is
involved in L-tryptophan metabolism and degradative pathways
of L-lysine and L-hydroxylysine) and phosphoserine aminotrans-
ferase (PSAT; 4 clones) (EC 188.8.131.52; an enzyme that is involved
in serine biosynthesis) were selected. Unigenes for GS2, CAT2,
and GCVT were among those with the largest number of EST
clones in ESTBC(Table 1).
Quantitative RT-PCR confirmed significantly higher expres-
sion of these genes in the bacteriocyte. Genes for GS2, CAT2,
GCVT, Henna, GCDH, and PSAT were expressed 20.7-, 93.3-,
30.4-, 297-, 23.5- and 18.7-fold higher in the bacteriocyte than in
the whole body, respectively (Fig. 1A).
Role of the Bacteriocyte in Amino Acid Metabolism of Aphids. The
bacteriocyte is the specialized cell for harboring the essential
symbiont Buchnera, whose pivotal role is the synthesis of
essential amino acids that are scarce in the phloem sap diet (2,
3). The genome of Buchnera is specialized for production of
essential amino acids: Genes for synthesis of essential amino
acids are retained, whereas genes for synthesis of nonessential
amino acids are mostly lost (11). In the bacteriocyte, therefore,
it is expected that essential amino acids are supplied by the
symbiont whereas nonessential amino acids must be synthe-
sized in excess by the host cell. In agreement with the
Table 1. Highly expressed genes in the bacteriocyte
Local ID No. of ESTs Protein homologSource organism NCBI accession no.
?-Tubulin at 84B (CG1913; ? Tub84B)
Cytosolic malate dehydrogenase
Rp L15 (CG17420)
Phosphoenolpyruvate carboxykinase (CG17725; Pepck)
2.3 ? 10?35
1.1 ? 10?34
6.9 ? 10?103
2.2 ? 10?80
9.3 ? 10?116
1.8 ? 10?66
1.8 ? 10?95
1.9 ? 10?46
1.1 ? 10?92
4.5 ? 10?28
1.1 ? 10?88
1.7 ? 10?22
2.2 ? 10?73
1.4 ? 10?41
1.0 ? 10?46
2.5 ? 10?76
1.6 ? 10?83
1.6 ? 10?30
2.2 ? 10?47
Genes are listed with annotations based on BLASTX similarity searches.
Nakabachi et al.
April 12, 2005 ?
vol. 102 ?
no. 15 ?
expectation, we found that genes relevant to utilization of
essential amino acids (i.e., CAT2, Henna, and GCDH) and
genes relevant to synthesis of nonessential amino acids (i.e.,
GS2 and PSAT) were highly expressed in the bacteriocyte. A
gene involved in catabolism of a nonessential amino acid,
glycine, (i.e., GCVT) was also highly expressed in the bacte-
riocyte. Notably, glycine is among the few nonessential amino
acids that Buchnera is able to synthesize (11). These results
unveiled an important aspect of the molecular basis of inter-
dependency between the host and symbiont.
Up-Regulated Genes Involved in Transport. The bacteriocyte tran-
scriptome ESTBCcontained 440 EST clones related to transport,
representing 38 unigenes. Within the category of ‘‘transport,’’ its
subcategory ‘‘amino acid transport (GO:0006865)’’ contained
significantly higher percentage of ESTs in ESTBCthan in the
whole-body transcriptomes ESTWB and ESTTC (P ? 0.001)
(Table 2). This subcategory was represented by 28 clones cor-
responding to two ESTBCunigenes, a contig R2C00038 and a
singlet BCA014030, which were similar to genes encoding cat-
respectively. R2C00038 (27 clones) was one of the most highly
expressed unigenes in ESTBC as already described (Table 1).
Moreover, the transport category contained several other uni-
genes with remarkably large numbers of EST clones: genes for
ADP?ATP translocase (ANT2; 70 clones, a translocator that
exchanges ADP and ATP across the mitochondrial inner mem-
brane), inorganic phosphate cotransporter (PC; 26 clones, an
integral membrane protein that belongs to the sodium?anion
cotransporter family), mitochondrial oxaloacetate transport pro-
tein (OT; 25 clones, a mitochondrial inner membrane protein
that transports oxaloacetate and sulfate), and ATP synthase
subunit c (AS; 23 clones, a component of mitochondrial ATP
synthase) (Table 1). Other transport-related unigenes were also
examined: genes for Ras-like Rab GTPase (10 clones, a GTP-
binding protein that regulates vesicle transport) and mitochon-
drial glutamate carrier (GC; 7 clones, an integral membrane
protein involved in the transport of glutamate across the inner
Quantitative RT-PCR confirmed significantly higher expres-
AS, Ras-like Rab GTPase, and GC were expressed 7.98-, 68.1-,
182-, 2.92-, 40.7-, and 24.5-fold higher in the bacteriocyte than
in the whole body, respectively (Fig. 1B).
Table 2. GO classification of ESTs in aphid transcriptomes
Percent of total fly-hit ESTs
GO:0008150: Biological process
[i] GO:0007610: behavior
[i] GO:0007582: physiological process
[i] GO:0008152: metabolism
[i] GO:0006520: amino acid metabolism
[i] GO:0009987: cellular process
[i] GO:0007154: cell communication
[i] GO:0030154: cell differentiation
[i] GO:0050875: cellular physiological process
[i] GO:0008151: cell growth and?or maintenance
[i] GO:0006810: transport
[i] GO:0006865: amino acid transport
[i] GO:0007275: development
[i] GO:0050896: response to stimulus
[i] GO:0006952: defense response
[i] GO:0042742: defense response to bacteria
[i] GO:0050789: regulation of biological process
*GO terms were assigned to the aphid ESTs based on fly homolog’s annotations (FlyBase).
PSAT. (B) Genes related to transport: ANT2, PC, OT, AS, Ras-like Rab GTPase, and GC. (C) Genes related to defense response: LSZ and HSC70. White columns,
expression levels in the whole body; black columns, expression levels in the bacteriocyte; bars, SE (n ? 10). The expression levels are shown in terms of mRNA
copies of target genes per copy of mRNA for RpL7. Asterisks indicate statistically significant differences (Mann–Whitney U test;*, P ? 0.05;**, P ? 0.01).
Quantitative RT-PCR of aphid genes expressed in the bacteriocyte. (A) Genes related to amino acid metabolism: GS2, CAT2, GCVT, Henna, GCDH, and
www.pnas.org?cgi?doi?10.1073?pnas.0409034102Nakabachi et al.
Importance of Transport in the Bacteriocyte. In the aphid symbiotic
system, metabolic integration and interdependency between the
host and symbiont are so intricate that the partners are regarded
as comprising an almost inseparable biological entity (2, 3).
Located at the symbiotic interface, the bacteriocyte is expected
to be involved in exchange of various metabolites and substrates
between the host and the symbiont. In agreement with the
expectation, we identified a number of transport-related genes
that were strikingly up-regulated in the bacteriocyte.
The aphid–Buchnera mutualism is principally based on the
provision of essential amino acids from the symbiont to the host
(2, 3). The high expression of the gene encoding CAT2, which is
involved in the import of cationic amino acids such as lysine and
arginine (essential amino acids) from the environment into the
eukaryotic cells (30), is intriguing in this context. In the bacte-
riocyte, Buchnera cells are encased in a membrane of host origin
(31, 32). Transporters of this type located on the host membrane
will enable the transport of amino acids synthesized by Buchnera
into the cytoplasm of the bacteriocyte.
We identified several genes for mitochondria-related trans-
porters (ANT2, OT, AS, and GC) that were significantly up-
regulated in the bacteriocyte. The abundance of their transcripts
probably reflects high mitochondrial activity in the bacteriocyte,
where active ATP synthesis and energy transfer are required for
energy-consuming amino acid metabolisms. Certainly, electron
microscopic studies have identified a dense population of mito-
chondria in aphid bacteriocytes (31, 32). The genome of Buch-
nera lacks most genes for tricarboxylic acid (TCA) cycle, whereas
complete gene sets for glycolysis and respiratory chain are
retained (11). Because TCA cycle operates in mitochondria,
although speculative, the up-regulated mitochondrial activity
and transport in the bacteriocyte might be relevant to
cooperative metabolic interactions between Buchnera and the
One of the up-regulated genes encoded Ras-like Rab GTPase,
which regulates vesicular transport of proteins and lipids be-
tween compartments in eukaryotic cells (33). Because Buchnera
cells are encased in a host membrane, intracellular trafficking
mechanism of this type may play important roles in the symbiotic
system. The genome of Buchnera lacks key genes for phospho-
lipid biosynthesis, implying that Buchnera is unable to synthesize
its own cell membrane (11). Phospholipids of host origin might
be delivered to Buchnera cells by using the vesicular transport
Most Abundant Transcripts in the Bacteriocyte Encoded Invertebrate-
Type Lysozymes. The unigenes corresponding to the most abun-
dant transcripts, R2C00037 (134 clones) and R2C00204 (71
clones), were similar to each other (97.7% nucleotide sequence
similarity), representing 8.7% of total ESTs in ESTBC(Table 1).
The top BLAST hit for these unigenes was fly CG6426
(R2C00037: E ? 1.4 ? 10?37and R2C00204: E ? 4.1 ? 10?37)
whose function was unknown, but GO annotation ‘‘defense
response to bacteria’’ (GO:0042742) was assigned to the gene at
the ‘‘inferred from electronic annotation (IEA)’’ level of evi-
dence. In addition, subordinate hits with significant similarity
(E ? 1.0 ? 10?10) included lysozyme (LSZ) i-1 from Anopheles
gambiae (R2C00037: E ? 2.3 ? 10?35and R2C00204: E ? 1.1 ?
10?34), destabilase 2 homolog from Drosophila melanogaster
(R2C00037: E ? 1.7 ? 10?27and R2C00204: E ? 8.7 ? 10?27),
destabilase 2 from medicinal leech, Hirudo medicinalis
(R2C00037: E ? 5.1 ? 10?11and R2C00204: E ? 7.9 ? 10?12)
and destabilase I from H. medicinalis (R2C00037: E ? 8.7 ?
10?11and R2C00204: E ? 1.3 ? 10?11).
Lysozymes (EC 184.108.40.206) are the enzymes that destroy bacte-
rial cell walls, and are structurally classified into several types;
i.e., chicken, goose, invertebrate, plant, and bacteria types (34).
Phylogenetic analysis indicated that lysozymes encoded by
R2C00037 and R2C00204 belong to the invertebrate type (data
We performed quantitative RT-PCR by using the primer set
for conserved regions between R2C00037 and R2C00204,
whereby their transcripts were quantified collectively (Table 3).
The expression level was 156 times higher in the bacteriocyte
than in the whole body (Fig. 1C). It is also notable that the level
of the transcripts was strikingly higher, 25.3-fold, than that of the
control gene encoding RpL7. These results confirmed that the
lysozyme-encoding genes represent the most highly expressed
genes in the aphid bacteriocyte.
is to harbor the symbiotic bacteria, highly expresses the antibac-
terial genes. This finding might be relevant to lysosomal break-
down of Buchnera (31, 32) or elimination of microbial intruders
(35). Future studies should focus on the biochemical properties,
substrate specificity, and antimicrobial spectrum of the bacte-
Other Up-Regulated Genes. Based on comparative analysis between
ESTBCand ESTWB, several other genes seemingly up-regulated in
RT-PCR (Fig. 1C; see also Fig. 2, which is published as supporting
information on the PNAS web site). Whereas the gene for a heat
shock protein cognate 4 (HSC70; 26 clones) was among the most
highly expressed unigenes in the defense response category (see
Table 1), quantitative RT-PCR revealed no significant up-
regulation of the gene in the bacteriocyte (P ? 0.05) (Fig. 1C).
Unigenes Encoding Transcription Factors. Unigenes BCA016027 (a
singlet), R2C00051 (two clones), R2C00028 (nine clones), and
R2C00237 (two clones) encoded putative transcription factors
such as Beadex, CG17870, CG1101, and CG5033, respectively
(Table 4), none of which showed significant similarity to Distal-
less, Ultrabithorax?Abdominal-A, or Engrailed that were de-
tected with antibodies in bacteriocytes of several species of
ESTs Similar to Only Prokaryotic Genes.Inthebacteriocytetranscrip-
tome, we identified two unigenes that showed significant similarity
only to prokaryotic genes, but not to those of Buchnera. R2C00193
(10 clones) and R2C00214 (4 clones) matched to RlpA (rare
lipoprotein A) precursor of Yersinia pestis (E ? 7.3 ? 10?11) and a
hypothetical protein of Wolbachia pipientis (E ? 7.1 ? 10?22),
respectively (Table 4). It is notable that ESTTCalso contained a
transcript similar to R2C00193 (CD450666: E ? 1.0 ? 10?44).
Southern blot analysis confirmed that these transcripts have cor-
responding loci in the aphid genome (data not shown). Whereas
genes with significant similarity to prokaryotic genes have been
found in the genomes of a bruchid beetle and a silkworm (37, 38),
high level of expression of these genes in the bacteriocyte is
intriguing in the context of its obligatory interdependency with the
symbiotic bacterium, Buchnera.
In this study, we demonstrated that a number of genes that are
related to amino acid metabolism, intra- and intercellular
transport, antibacterial activity and other biological processes,
are highly expressed in the bacteriocyte. The up-regulation of
host genes relevant to amino acid metabolism corroborated
our previous finding that Buchnera-mediated production of
essential amino acids from nonessential amino acids is among
the most important processes in the symbiotic system, and,
furthermore, profoundly enriched our understanding of the
complementary metabolic features that underpin the integrity
of the host–symbiont relationship. The up-regulated genes
related to transport highlighted an important aspect of the
bacteriocyte that mediates exchange of various metabolites
Nakabachi et al.
April 12, 2005 ?
vol. 102 ?
no. 15 ?
and substrates at the host–symbiont interface. The bacterio- Download full-text
cyte-specific lysozyme genes provided promising candidate
molecules that might be involved in the control and mainte-
nance of the bacterial flora in the host cell. The other
up-regulated host genes, although their roles are currently
obscure, would provide clues to understanding of previously
unrecognized aspects of the host–symbiont interactions. Of
course, these hypothetical processes that we suggest for the
bacteriocyte must be verified by functional analysis of the
up-regulated host genes.
We thank Dr. Satoru Kobayashi for supporting computer analysis. This
work was supported by the Special Postdoctoral Researchers Program of
RIKEN, Japan Society for the Promotion of Science Grant-in-Aid for
Young Scientists Grant (B) 14760031, and a Research Fellowship of the
Japan Society for the Promotion of Science for Young Scientists.
1. Munson, M. A., Baumann, P. & Kinsey, M. G. (1991) Int. J. Syst. Bacteriol. 41,
2. Baumann, P., Baumann, L., Lai, C. Y., Rouhbakhsh, D., Moran, N. A. & Clark,
M. A. (1995) Annu. Rev. Microbiol. 49, 55–94.
3. Douglas, A. E. (1998) Annu. Rev. Entomol. 43, 17–37.
4. Sasaki, T. & Ishikawa, H. (1995) J. Insect Physiol. 41, 41–46.
6. Nakabachi, A. & Ishikawa, H. (1997) Insect Biochem. Mol. Biol. 27, 1057–1062.
7. Nakabachi, A. & Ishikawa, H. (1999) J. Insect Physiol. 45, 1–6.
8. Ziegler, H. (1975) in Transport in Plants I, eds. Zimmermann, M. H. & Milburn,
J. A. (Springer, New York), Vol. 1, pp. 59–100.
9. Sandstrom, J. & Moran, N. (1999) Entomol. Exp. Appl. 91, 203–210.
10. Moran, N. A., Munson, M. A., Baumann, P. & Ishikawa, H. (1993) Proc. R. Soc.
London Ser. B 253, 167–171.
11. Shigenobu, S., Watanabe, H., Hattori, M., Sakaki, Y. & Ishikawa, H. (2000)
Nature 407, 81–86.
12. Tamas, I., Klasson, L., Canback, B., Naslund, A. K., Eriksson, A. S.,
Wernegreen, J. J., Sandstrom, J. P., Moran, N. A. & Andersson, S. G. (2002)
Science 296, 2376–2379.
13. van Ham, R. C., Kamerbeek, J., Palacios, C., Rausell, C., Abascal, F., Bastolla,
U., Fernandez, J. M., Jimenez, L., Postigo, M., Silva, F. J., et al. (2003) Proc.
Natl. Acad. Sci. USA 100, 581–586.
14. Stephens, R. S., Kalman, S., Lammel, C., Fan, J., Marathe, R., Aravind, L.,
Mitchell, W., Olinger, L., Tatusov, R. L., Zhao, Q., et al. (1998) Science 282,
15. Andersson, S. G., Zomorodipour, A., Andersson, J. O., Sicheritz-Ponten, T.,
Alsmark, U. C., Podowski, R. M., Naslund, A. K., Eriksson, A. S., Winkler,
H. H. & Kurland, C. G. (1998) Nature 396, 133–140.
16. Akman, L., Yamashita, A., Watanabe, H., Oshima, K., Shiba, T., Hattori, M.
& Aksoy, S. (2002) Nat. Genet. 32, 402–407.
17. Gil, R., Silva, F. J., Zientz, E., Delmotte, F., Gonzalez-Candelas, F., Latorre,
A., Rausell, C., Kamerbeek, J., Gadau, J., Holldobler, B., et al. (2003) Proc.
Natl. Acad. Sci. USA 100, 9388–9393.
18. Oshima, K., Kakizawa, S., Nishigawa, H., Jung, H. Y., Wei, W., Suzuki, S.,
19. Wu, M., Sun, L. V., Vamathevan, J., Riegler, M., Deboy, R., Brownlie, J. C.,
20. Ishikawa, H. (1982) Comp. Biochem. Physiol. B Biochem. Mol. Biol. 72,
21. Carninci, P., Kvam, C., Kitamura, A., Ohsumi, T., Okazaki, Y., Itoh, M.,
Kamiya, M., Shibata, K., Sasaki, N., Izawa, M., et al. (1996) Genomics 37,
22. Carninci, P. & Hayashizaki, Y. (1999) Methods Enzymol. 303, 19–44.
23. Shibata, Y., Carninci, P., Watahiki, A., Shiraki, T., Konno, H., Muramatsu, M.
& Hayashizaki, Y. (2001) BioTechniques 30, 1250–1254.
24. Carninci, P., Shibata, Y., Hayatsu, N., Itoh, M., Shiraki, T., Hirozane, T.,
Watahiki, A., Shibata, K., Konno, H., Muramatsu, M. & Hayashizaki, Y. (2001)
Genomics 77, 79–90.
26. Ewing, B. & Green, P. (1998) Genome Res. 8, 186–194.
27. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W.
& Lipman, D. J. (1997) Nucleic Acids Res. 25, 3389–3402.
28. Huang, X. & Madan, A. (1999) Genome Res. 9, 868–877.
29. Hunter, W. B., Dang, P. M., Bausher, M. G., Chaparro, J. X., McKendree, W.,
Shatters, R. G., Jr., McKenzie, C. L. & Sinisterra, X. H. (2003) J. Insect Sci. 3,
30. Hoshide, R., Ikeda, Y., Karashima, S., Matsuura, T., Komaki, S., Kishino, T.,
Niikawa, N., Endo, F. & Matsuda, I. (1996) Genomics 38, 174–178.
31. Hinde, R. (1971) J. Insect Physiol. 17, 1791–1800.
32. Griffiths, G. W. & Beck, S. D. (1973) J. Insect Physiol. 19, 75–84.
33. Zerial, M. & McBride, H. (2001) Nat. Rev. Mol. Cell Biol. 2, 107–117.
34. Bachali, S., Jager, M., Hassanin, A., Schoentgen, F., Jolles, P., Fiala-Medioni,
A. & Deutsch, J. S. (2002) J. Mol. Evol. 54, 652–664.
35. Nakabachi, A., Ishikawa, H. & Kudo, T. (2003) J. Invertebr. Pathol. 82, 152–161.
36. Braendle, C., Miura, T., Bickel, R., Shingleton, A. W., Kambhampati, S. &
Stern, D. L. (2003) PLoS Biol. 1, E21.
37. Kondo, N., Nikoh, N., Ijichi, N., Shimada, M. & Fukatsu, T. (2002) Proc. Natl.
Acad. Sci. USA 99, 14280–14285.
38. Daimon, T., Hamada, K., Mita, K., Okano, K., Suzuki, M. G., Kobayashi, M.
& Shimada, T. (2003) Insect Biochem. Mol. Biol. 33, 749–759.
www.pnas.org?cgi?doi?10.1073?pnas.0409034102Nakabachi et al.