Microb Ecol (2008) 56:696–703 DOI 10.1007/s00248-008-9389-4

ORIGINAL ARTICLE

Interactions of Chaperonin with a Weakly Active Anthranilate Synthase from the Aphid Endosymbiont Buchnera aphidicola Chia-Ying Huang & Chi-Ying Lee & Hsiao-Chen Wu & Mei-Hwa Kuo & Chi-Yung Lai

Received: 17 October 2007 / Accepted: 10 March 2008 / Published online: 14 May 2008 # Springer Science + Business Media, LLC 2008

Abstract The endosymbiotic bacterium Buchnera provides its aphid host with essential amino acids. Buchnera is typical of intracellular symbiotic and parasitic microorganisms in having a small effective population size, which is believed to accelerate genetic drift and reduce the stability of gene products. It is hypothesized that Buchnera mitigates protein instability with an increased production of the chaperonins GroESL. In this paper, we report the expression and functional analysis of trpE, a plasmid-borne fast-evolving gene encoding the tryptophan biosynthesis enzyme anthranilate synthase. We overcame the problem of low enzyme stability by using an anthranilate synthase-deficient mutant of E. coli as the expression host and the method of genetic complementation for detection of the enzyme activity. We showed that the Buchnera anthranilate synthase was only weakly active at the temperature of 26°C but became inactive at the higher temperatures of 32°C and 37°C and that the coexpression with chaperonin genes groESL of E. coli enhanced the function of the Buchnera enzyme. These findings are consistent with the proposed role of groESL in the Buchnera–aphid symbiosis.

C.-Y. Huang : C.-Y. Lee : H.-C. Wu : C.-Y. Lai (*) Department of Biology, National Changhua University of Education, 1 Jin Der Road, Changhua 50007 Taiwan, Republic of China e-mail: [email protected] M.-H. Kuo Department of Entomology, National Chung Hsing University, Taichung 40227 Taiwan, Republic of China

Introduction Most aphids carry the endosymbiotic bacterium Buchnera [1, 5], which provides the host insects with essential amino acids [1, 10, 11]. The 660-kb genome of Buchnera sp. APS is about one seventh the size of the E. coli genome and yet contains almost all the genes needed for the synthesis of essential amino acids [29]. Most of these genes are also present in Buchnera genomes from three other species of aphids [25, 32, 33]. Of special interest are the genes involved in the synthesis of tryptophan: The first step in the metabolic pathway dedicated to tryptophan biosynthesis is catalyzed by the enzyme anthranilate synthase (AS), encoded by the gene pair trpE and trpG. These two genes are located on a plasmid in Buchnera associated with many species of aphids [18, 27]. Since, in most bacteria, AS is the rate-limiting enzyme in the pathway and is regulated by product inhibition [8], the plasmid-borne location of trpEG was initially thought as a means to amplify the gene number for the overproduction of tryptophan. A finding consistent with such view is that in many strains of Buchnera, the number of trpEG is further increased by having nearly identical tandem repeats of the genes on each plasmid [18, 27]. However, Buchnera cells contain many copies of chromosomes [17], and the plasmid-to-chromosome ratios is not always in favor of plasmid-borne genes [26]. Instead, Latorre et al. [21] suggest that plasmid localization provides greater plasticity in gene evolution, allowing Buchnera to adapt to changing nutritional needs of the host insect. Buchnera is typical of endosymbiotic microorganisms in having a small effective population size due to a bottleneck effect created by strict maternal transmission [22]. Such bottleneck effect combined with a lack of genetic recombination is predicted to accelerate genetic drift, increase the fixation of mildly deleterious mutations, and hence desta-

Coexpression of Buchnera trpE with groESL

bilize gene products [23]. Indeed, many genes of Buchnera show higher nucleotide substitution rates than their counterparts from free-living bacteria such as E. coli and Salmonella [7, 23, 34]. For trpEG genes, there is a further rate increase attributed to their plasmid-borne location [28, 36]. It is hypothesized that Buchnera mitigates the damage of reduced protein stability by maintaining a high cytoplasmic level of the chaperonin GroESL, which helps denatured protein molecules to refold [23]. However, this hypothesis has not been tested directly using biochemical means. Although the AS enzyme occupies a key role in the aphid–Buchnera symbiosis and extensive alterations in trpEG location and structure imply changes in the enzymatic and regulatory properties of AS, direct studies on this enzyme are still lacking likely due to its lack of stability. In this report, we used an E. coli trpE mutant as an in vivo detection system to overcome this obstacle. Using the method of genetic complementation, we detected a weak and temperature-sensitive activity of Buchnera AS. We also showed that this enzyme activity could be enhanced by the coexpression with the chaperonin genes groESL.

697

was extracted from fresh aphids using Gene-Spin-V2 genomic DNA isolation kit (Protech, Taipei) following the manufacturer’s instructions. The concentration of DNA samples was determined based on absorption of 260 nm UV light using a Jasco V530 UV/VIS spectrophotometer. Bacterial Strains and Media E. coli strain Top10 (Invitrogen) was used for cloning and plasmid construction. E. coli BL21(DE3) was used for the construction of a ΔtrpE strain. The conditionally lethal groESL strain MGM100 was kindly provided by Dr. Peter A. Lund [31]. The media used were either Luria–Bertani (LB; 1% tryptone, 0.5% yeast extract, 0.5% NaCl) or M9 minimal medium containing 2% mannitol as the carbon source. For maintaining plasmids, ampicillin (50 μg/ml), chloramphenicol (25 μg/ml), or kanamycin (50 μg/ml) was added to the media. For growing tryptophan-auxotrophic strains on minimal media, tryptophan (40 μg/ml) was added. PCR and Cloning of Buchnera groESL and trpE

Materials and Methods Aphids and Aphid DNA The original pea aphid (Acyrthosiphon pisum) population was collected from Wufong Township, Taichung, Taiwan, and raised on potted pea (Pisum sativum) under laboratory conditions (15°C, a photoperiod 12:12-h light/dark). Host plants were renewed every 4–5 days by placing fresh pots of seedlings next to the old ones and allowing aphids to migrate to the new plants. Deoxyribonucleic acid (DNA)

Primers used in this study are listed in Table 1. The Buchnera groESL open reading frames (ORFs) were amplified using the forward primer BamAPGroESf and the reverse primer EcoAP21027r. The polymerase chain reaction (PCR) product was cloned into the pGEM-T Easy plasmid to obtain the plasmid pHCYGEMBG. The Buchnera trpE ORF was amplified using the forward primer XhoAPtrpEf and the reverse primer ptrpE1566r. The trpE gene from Buchnera of the pea aphid is located on a plasmid that also carries tandem repeats of

Table 1 Primers used in this study Primer

Sequence

Target gene

BamAPGroESf EcoAP21027r XhoAPtrpEf ptrpE1566r Ptrp1452f Ptrp1471r ParaBf ParaBr EctrpE1f

5′-TACGGATCCGAAGGAGATTATCATATGAAAATTCGTCCATTG-3′a 5′-TACGAATTCTTATTACTATGATATATTCTCAT-3′a 5′-TACCTCGAGAAGGAGTAGATGAAATGTTTTTGATTGAAAAG-3′a 5′-AAGCTTTTAAGAGGATCCCATTGTAAAATG-3′a 5′-AGCCGGTGTTGTTTTTAATT↓CAA-3′b 5′-TTCACTTCGTCTTCAGGTATTG↓A-3′b 5′-CATCGATTTATTATGACAACTTGACGGCTAC-3′a 5′-GTCGACTTTTTATAACCTCCTTAGAGCTC-3′a 5′-ATGCAAACACAAAAACCGACTCTCGAACTGCTAACCTGCGAAGGCG CTGGAGCTGCTTCGAAGTT-3′c 5′-GCCGTAGCTGCCGCGGCGACGACCTTCCGCCTCGGCAATTAACTGCAT ATGAATATCCTCCTTAG-3′c

Buchnera groESL Buchnera groESL Buchnera trpE Buchnera trpE Buchnera trpE Buchnera trpE E. coli PBAD E. coli PBAD E. coli trpE

EctrpE1303b a

Restriction sites are underlined. Putative initiation codons are in bold face Arrows indicate position of CA:TG insertions found in a trpE pseudogene [29] c Sequences homologous to the trpE of E. coli K12 are italicized b

E. coli trpE

698

trpE pseudogenes [2, 18, 29]. Since pseudogenes outnumber the functional genes by several folds and both types of gene contain the same primer-binding site, PCR amplification will result in predominant products of the pseudogene, which is indistinguishable from the functional copy by molecular size. We therefore designed a pair of wild-type allele-specific internal primers and used these primers in an overlap extension PCR to increase the proportion of the functional gene in the final PCR product. The published sequences of the trpE pseudogene from the Buchnera strain APS differs from the functional copy at three nucleotide positions, including a CA insertion between positions 1501 and 1502. We designed two internal primers that span the insertion site and are complementary to each other. The internal reverse primer Ptrp1471r was paired with the forward primer XhoAPtrpEf to amplify a fragment spanning nucleotide positions 1 to 1482 of trpE. The internal forward primer Ptrp1452f was paired with the reverse primer ptrpE1566r to amplify a fragment spanning nucleotide positions 1452 to 1566 of trpE. The two PCR products were purified using PCR-M kit (Viogene, Taipei, Taiwan) and mixed in an equal molar ratio in an extension cycle with fresh enzyme and deoxyribonucleotide triphosphates to obtain the full-length product, which was further amplified using the two flanking primers. All PCR cycles comprised an initial denaturation stage of 5 min at 94°C, 35 rounds of amplification, each composed of 30 s at 94°C, 90 s at 60°C, and 180 s at 72°C, and a final extension of 5 min at 72°C. A Pfu and Taq enzyme mixture (unit ratio 1:18, purchased separately from Fermentas) was used for all PCR reactions. The final PCR product was cloned using the pGEM-T Easy PCR cloning vector (Promega). Recombinant plasmids with the correct fragment inserted in the same direction as the lacZ gene were identified using restriction analysis of miniprep DNA and nucleotide sequencing. Nucleotide sequencing was done by a commercial sequencing service (Mission Biotech, Taipei). Sequence analysis and comparison were done using the program BioEdit 7.01 [14]. Construction of an E. coli trpE Mutant We used the method described by Datsenko and Wanner [9]. The primers EctrpE1f and EctrpE1303b, which contain 5′ extensions homologous to the 5′ and 3′ termini of E coli trpE, were used to amplify a chloramphenicol-resistant (cat) gene cassette from the template plasmid pKD3. The PCR product was transformed into the E. coli strain BL21 (DE3) carrying the recombination helper plasmid pKD46 to replace the chromosomal trpE through homologous recombination. Successful replacement was confirmed by PCR using trpE-flanking primers and by a tryptophan-dependent growth phenotype.

C-Y. Huang et al.

Spot test for trpE Function Overnight cultures of E. coli strains transformed with the empty vector pGEM-T Easy or with vectors carrying an insert of Buchnera trpE were grown with shaking at 26°C to early log phase (optical density at 590 nm [OD590]=0.1) in 3 ml of a mannitol minimal medium supplemented with ampicillin and tryptophan. To remove tryptophan from the suspension, cells were spun down and resuspended in sterile water to an OD590 of 0.1. Serial fivefold dilutions in sterile water were made for each suspension, and 5 μl of each dilution was spotted on the surface of a tryptophanfree mannitol minimal medium plate supplemented with ampicillin. Plates were then incubated at 26, 32, or 37°C, and growth was examined daily. As the growth rate differed between temperatures, incubation at each temperature was terminated when colonies of the control strain BL21(DE3)/ pGEM-T Easy reached the size of 1 mm in diameter. The amount of growth for mutant strains was estimated by the dilution endpoint. Images of plates were saved using a desktop scanner. groESL Expression Plasmids The plasmid pOFXbad-SL1 is a gift from Dr. Oliver Fayet [6]. A pACYC184-based plasmid carrying the Buchnera groESL operon under the control of the PBAD promoter of E. coli was constructed as follows. A 1,243-bp fragment containing the E. coli araC gene and the PBAD promoter was amplified from the plasmid pKD46 using primers ParaBf and ParaBr. Flanking ClaI and SalI restriction sites incorporated into the PCR primers were digested with the restriction endonucleases ClaI and SalI (Fermentas). The digested fragment was purified and ligated with the pACYC184 plasmid DNA digested with the same enzymes to produce the plasmid pHCYC6. A 288-bp fragment was cut off from the plasmid pHCYC6 using the enzymes SalI and EagI and replaced with a Buchnera groESL-containing fragment cut from the plasmid pHCYGEMBG using the same enzymes to produce the plasmid pHCYC6BG. The expression of groESL from this plasmid was confirmed by standard sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis [20]. Genetic Complementation of the E. coli groE Mutant with Buchnera groE The conditional groE strain MGM100 carries the chromosomal copy of the groESL operon under the control of the PBAD promoter [31]. The Buchnera groESL-bearing plasmid pHCYGEMBG or the empty vector pGEM-T Easy was transformed into strain MGM100. Transformants were then streaked on LB plates containing 0.5% D-glucose (nonper-

Coexpression of Buchnera trpE with groESL

missive condition) or 0.5% L-arabinose (permissive condition) and incubated at 26°C and 37°C. Growth was examined after 1 day. Coexpression of trpE and groE The strain WSJ213/pHCYBuE1 was transformed with the plasmid pOFXbad-SL1, pHCYC6BG, or pHCYC6. Stable transformants were streaked on mannitol minimal medium plates supplemented with L-arabinose (0.01 mg/ml) and antibiotics. Colony size was compared after incubation at 26°C for 1 week. For growth rate determination, transformants were grown in a liquid mannitol minimal medium supplemented with L-arabinose (0.01 mg/ml) and antibiotics at 26°C. Cell density was measured as OD590 at time intervals and plotted using the program Excel 2003. Growth rates were estimated from the exponents of best-fit exponential curves.

Results Partial Complementation of E. coli trpE-null Mutation by Buchnera trpE The trpE gene of B. aphidicola APS was amplified from pea-aphid DNA with the ribosome-binding site sequence GAAGGAGA incorporated into the forward PCR primer. Figure 1 Effect of temperature on the complementation of the E. coli trpE mutant WSJ213 by Buchnera trpE. Cultures of the parental strain BL21(DE3)/ pGEM, the trpE mutant WSJ213/pGEM, and the mutant strain complemented by Buchnera trpE (WSJ213/pHCYBuE1) were serially diluted, spotted on a tryptophan-supplemented or tryptophan-deficient medium, and incubated at 26°C, 32°C, or 37°C

699

The amplified fragment was then inserted into the pGEM-T Easy vector in the same direction as the lacZ gene to obtain the plasmid pHCYBuE1. Such an arrangement allows the cloned gene to be expressed from the Plac promoter on the cloning vector. Compared with a published sequence of the same gene (27, GenBank accession number L43555), the insert in pHCYBuE1 contained four base changes (C597T, T867C, G1149A, C1248T). All of these changes are silent and do not alter the amino acid sequence of the translated product. Initial genetic complementation experiments using the AS-deficient mutant WSJ213 showed that Buchnera trpE was not functional in E. coli, as no colony of WSJ213/ pHCYBuE1 appeared on the tryptophan-free mannitol minimal medium after 24 h of incubation at 37°C. However, further tests using lower incubation temperatures and long incubation times showed that the gene was weakly active and could partially restore the growth of WSJ213 on tryptophan-deficient media. Figure 1 shows the result of a semiquantitative spot test for growth: On the tryptophansupplemented mannitol minimal medium, the growth of both the parental strain BL21(DE3)/pGEM and the Buchnera trpE-complemented mutant strain WSJ213/pHCYBuE1 was largely unaffected by temperature, with only a small reduction in plating efficiency at 26°C. On the tryptophandeficient medium, however, the growth of WSJ213/pHCYBuE1 became highly temperature dependent: Colonies of this strain formed only at 26°C and with a lower plating

700

C-Y. Huang et al.

efficiency than for BL21(DE3)/pGEM at the same temperature. In the tryptophan-deficient liquid mannitol minimal medium, the generation time of the Buchnera trpE-complemented WSJ213/pHCYBuE1 was 8.3±1.2 h (n=3) at 26°C, significantly longer than that of the parental strain BL21 (DE3)/pGEM (3.9±0.1 h, P<0.05, Student’s t test). Coexpression of groESL Enhances trpE Function Buchnera groESL was amplified from aphid DNA and cloned into the pGEM-T vector to obtain the plasmid pHCYGEMBG, which was able to complement the conditional lethal E. coli groESL mutant MGM100 (data not shown), indicating that Buchnera groESL was functional in E. coli. We then constructed a pGEM-compatible pACYC184-based expression plasmid pHCYC6BG, which carries the Buchnera groESL controlled by the PBAD promoter. For the expression of E. coli groESL, we used the pACYC184-based plasmid pOFXSL-1, which carries the E. coli groESL controlled by the PBAD promoter [6]. Each groE plasmid or the empty control vector pHCYC6 was transformed into WSJ213/pHCYBuE1 for coexpression with trpE. In these transformants, the expression of trpE from pHCYBuE1 depended on the uninduced background activity of the Plac promoter, which produced sufficient AS to support the growth of WSJ213 on mannitol minimal media (as shown in Fig. 1). However, the expression of groESL required the induction with Larabinose, which at the concentrations of 0.5% and higher noticeably reduced growth. Hence, we tested different concentrations of L-arabinose and found that 0.01 mg/ml was the highest concentration without growth inhibition. Therefore, this concentration was used in all liquid and solidified media in the coexpression experiments. When induced with 0.01 mg/ml of arabinose, cells harboring a groESL expression plasmid accumulated GroESL to a level easily detectable by SDS-PAGE (Fig. 2). Figure 3a shows a comparison of colony size on mannitol minimal medium plates: Colonies of the strain WSJ213/pHCYBuE1+pOFXbad-SL1 expressing E. coli groESL were consistently bigger than those of the control strain WSJ213/pHCYBuE1+ pHCYC6, but colonies of the strain WSJ213/pHCYBuE1+ pHCYC6BG expressing Buchnera groESL were similar in size to the control strain. In the liquid medium, the generation times of WSJ213/pHCYBuE1 + pHCYC6, WSJ213/pHCYBuE1+pHCYC6BG, and WSJ213/pHCYBuE1+pOFXbad-SL1 were 11.0±1.25, 10.6±0.45, and 5.24±0.06 h, respectively (Fig. 3b). The coexpression of E. coli groESL significantly shortened the generation time, while the coexpression of Buchnera groESL did not. The coexpression of groESL did not expand the range of active temperature for trpE. Both WSJ213/pHCYBuE1+pOFXbadSL1 and WSJ213/pHCYBuE1+pHCYC6BG failed to form a

Figure 2 SDS-PAGE analysis of expression of plasmid-borne, PBADregulated groESL. The control strain BL21(DE3)/pHCYC6 carrying an empty vector and the strain BL21(DE3)/pHCYC6BG carrying plasmid-borne Buchnera groESL under the control of the PBAD promoter were grown in LB and induced with 0.01 mg/ml of Larabinose for 0, 1, 2, and 3 h. M: molecular weight markers. Numbers indicate sizes in kilodaltons. Arrows indicate the expected positions of GroEL and GroES

colony at 32°C and 37°C after prolonged incubation (data not shown). Putative Initiation Codon of trpE The first published sequence of Buchnera trpE comprises 1,566 nucleotides and can be translated into a protein of 521 amino acids [27]; in another trpE sequence presented as part of the complete Buchnera genome [29], the first 30 bases from the first sequence are missing, and an ATT codon 30 bp further upstream is therefore chosen as the putative initiation codon (Fig. 4). This version of the gene can be translated into a protein of the same size but with the second to ninth amino acids different from the first one. In addition to the 30-bp deletion, the two sequences differ in three other nucleotide positions, one of which (A826G) causes an amino acid substitution (N276D, numbering based on [27]). We designed our PCR primers based on the second sequence but replaced the ATT codon with ATG and placed a ribosomal binding site (RBS) sequence 8 bases upstream from the engineered ATG. All resultant plasmids we sequenced were identical to the first sequence in that the first 30 bases were retained, yet one plasmid (designated pHCYBuEG43) contained a single-base deletion at the −23 position (the eighth position from the engineered ATG codon). This deletion would result in a frame shift and premature termination of the protein initiated from the engineered ATG. As the deleted position was part of the primer sequence, the deletion was likely to be artificially introduced during primer synthesis. However, this plasmid still supported the growth of WSJ213 on tryptophandeficient media, indicating that the downstream ATG codon was used as the initiation codon, as suggested by Rouhbaksh et al. [27]. A putative RBS of the sequence GAGA is located 9 bases upstream from this ATG.

Coexpression of Buchnera trpE with groESL

701

Figure 3 Effect of groESL overexpression on the growth of the strain WSJ213/pHCYBuE1. a Colonies of WSJ213/pHCYBuE1+pOFXbad-SL1 (E. coli groESL+trpE), WSJ213/pHCYBuE1+pHCYC6BG (Buchnera groESL+trpE), WSJ213/pHCYBuE1+pHCYC6 (trpE), and WSJ213 (none) on a mannitol minimal medium plate supple-

mented with L-arabinose (0.01 mg/ml) after 1 week of incubation at 26°C. b Growth rates of the same strains in a liquid mannitol minimal medium supplemented with L-arabinose (0.01 mg/ml). Asterisk indicates a significant difference in growth rate from the control strain WSJ213/pHCYBuE1+pHCYC6 (P<0.05, Student’s t test)

Discussion In the past, we have had great difficulty in detecting activities of Buchnera enzymes. In the literature, there were also very few reports of active Buchnera enzymes produced in vitro or in a heterologous host. The only exception is the successful complementation of temperature-sensitive E. coli groEL and groES mutants by corresponding Buchnera genes [24]. Our result represents the first example of detecting in E. coli the function of a biosynthetic enzyme from Buchnera. We achieved this by using a low incubation temperature of 26°C and a long incubation time of at least 1 week. The trpE is typical of Buchnera genes in being fast evolving and highly polymorphic in the population [2, 29]; therefore, its gene product, the AS enzyme, is also likely to be representative of Buchnera proteins in their lack of stability. Consistent with this notion, our method of gene complementation has successfully detected the weak activity of several other biosynthetic enzymes from Buchnera (Lai, unpublished data). The lack of a robust activity and possibly stability of Buchnera AS was clearly demonstrated by comparing the growth rates between the Buchnera trpE-complemented E. coli mutant WSJ213/pHCYBuE1 and its wild-type parental

strain BL21(DE3)/pGEM on the tryptophan-deficient minimal medium. The 8.3-h generation time of the mutant strain more than doubled the 3.9-h generation time of the parental strain, suggesting that the synthesis of tryptophan and hence the activity of AS had become the limiting factor for growth in the mutant strain even at the low temperature of 26°C. However, the induction of trpE expression with isopropyl-β-D-thiogalactopyranoside inhibited the growth of WSJ213/pHCYBuE1 (data not shown), likely due to the metabolic burden imposed by high-level gene expression. Therefore, we sought to improve the activity of Buchnera trpE through the use of chaperonin. Initially, we wanted to test the dependence of Buchnera AS activity on GroESL in vitro; however, AS activity from the cell lysate of Buchnera trpE-bearing E. coli was too low to be detected by a standard AS enzyme assay. Therefore, we decided to use the strain WSJ213/pHCYBuE1 in a coexpression experiment to demonstrate the interaction between AS and GroESL. Intriguingly, we found that GroESL of E. coli was able to improve the function of Buchnera trpE, while the cognate Buchnera GroESL was not. In our genetic complementation experiment, the groESL of Buchnera performed comparably to the E. coli genes, enabling the mutant MGM100 cells to grow into colonies after 18 h of

Figure 4 Comparison of the beginning region of two published trpE sequences from Buchnera of pea aphid and two sequences obtained in this study. trpE1: The Buchnera trpE sequence reported by Rouhbakhsh et al. [27], trpE2: the trpE sequence published as part of the B.

aphidicola APS genome sequence [28]. Arrowhead indicates the single-base deletion found in the insert of the recombinant plasmid pHCYBuEG43. Putative initiation codons are underlined or doubleunderlined

702

incubation. This could be attributed to the strong purifying selection acting on Buchnera groESL, making these genes more conserved and less affected by genetic drift than other Buchnera genes [34]. However, small effective population sizes of both aphid and Buchnera still impact groESL and cause their nonsynonymous to synonymous nucleotide substitution ratio to be higher than those of the E. coli groESL [15]. Consequently, Buchnera chaperonins GroESL are likely to be less active than the E. coli GroESL, and for the Buchnera genes to achieve the same degree of enhancement as the E. coli genes would require a higher level of induction that could become inhibitory to cell growth. The role of chaperonin GroEL in mitigating the harmful effects of genetic drift is supported by simulated evolutionary experiments in which genetic drift was artificially introduced into E. coli or Salmonella cultures [13, 30]. The reduced growth rates of the resultant mutants can be reversed by the overexpression of groESL. However, in the E. coli experiment, this fitness-enhancing effect is evident only when the growth medium is enriched with amino acids to support the high-level expression of groESL. Similarly, in our experiment, the enhancement of trpEG function could be detected only when the induction level of groESL was optimally adjusted, suggesting the need for a subtle balance between AS activity, chaperonin-assisted protein folding, translation capacity, and cell growth. It is possible that the stability of AS can be further improved in their native environment, i.e., the Buchnera cytoplasm, through synergistic action of GroESL with other cognate chaperone proteins. This speculation is consistent with a report showing that a single-point mutation affecting the heat shock protein ibpA renders Buchnera and its aphid host heat sensitive [12]. Alternatively, our results could reflect a true weak activity of AS in Buchnera cells. A weakly active AS could be explained by two evolutionary scenarios: a relaxed selection for tryptophan synthesis or a relaxed selection for AS activity. A relaxed selection for tryptophan synthesis should affect every gene involved in the biosynthesis pathway, but the effect on the plasmid-borne multicopy trpEG is likely to be more pronounced. An informative example is Buchnera from the aphid Diuraphis noxia: Similar to Buchnera of the pea aphid, this species carries on each plasmid multiple trpE pseudogenes and functional genes with high evolutionary rates, which indicate relaxed selection [19, 35]. Measures of tryptophan concentration in the body fluid of D. noxia show low levels of the amino acid, suggesting that tryptophan availability was not the major fitness-determining factor in the recent evolutionary history of the species [35]. In the latter scenario, the selection on AS activity could also be relaxed if AS is no longer the rate-limiting enzyme in the tryptophan biosynthesis pathway. Although this scenario contradicts with the traditional view of metabolic regulation on tryptophan

C-Y. Huang et al.

biosynthesis, it is consistent with the findings that the degree of trpEG amplification in the Buchnera APS genome shows little correlation with the rate of tryptophan production and the fitness of the host insect [3, 4]. In either scenario, the lack of importance of the AS activity to aphid nutrition should make this enzyme a poor candidate for testing the contribution of chaperonin to cellular and symbiotic functions. However, a proteome-wide analysis of chaperonin-dependent proteins showed that AS of E. coli does not normally interact with GroESL [16]. Our finding that Buchnera AS interacts with GroESL therefore demonstrated a diversification in the substrate range of GroESL between these two lineages of enteric bacteria. Most of the diversification is likely to be contributed by the Buchnera lineage. As the species evolves into an obligate endosymbiont, many members of its chaperonin-interacting proteome were inevitably lost due to genome reduction, while new substrate proteins could be added through adaptation of the remaining protein genes to a continuously high level of cytoplasmic chaperonin. Acknowledgments This research was supported by National Science and Technology Program for Agricultural Biotechnology grant 93-2317-B-018-001 from the National Science Council of Taiwan, Republic of China.

References 1. Baumann P, Moran NA, Baumann L (2006) Bacteriocyteassociated endosymbionts of insects. In: Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E (eds) The prokaryotes. Springer, New York, NY 2. Birkle LM, Douglas AE (1999) Low genetic diversity among pea aphid (Acyrthosiphon pisum) biotypes of different plant affiliations. Heredity 82:605–612 3. Birkle LM, Minto LB, Douglas AE (2002) Relating genotype and phenotype for tryptophan synthesis in an aphid–bacterial symbiosis. Physiol Entomol 27:302–306 4. Birkle LM, Minto LB, Walters KFA, Douglas AE (2004) Microbial genotype and insect fitness in an aphid–bacterial symbiosis. Funct Ecol 18:598–604 5. Buchner P (1965) Endosymbiosis of animals with plant microorganisms. Wiley & Sons, New York 6. Castanié MP, Berges H, Oreglia J, Prere MF, Fayet O (1997) A set of pBR322-compatible plasmids allowing the testing of chaperone-assisted folding of proteins overexpressed in Escherichia coli.. Anal Biochem 254:150–152 7. Clark MA, Moran NA, Baumann P (1999) Sequence evolution in bacterial endosymbionts having extreme base compositions. Mol Biol Evol 16:1586–1598 8. Crawford IP (1989) Evolution of a biosynthetic pathway: the tryptophan paradigm. Annu Rev Microbiol 43:567–600 9. Datsenko KA, Wanner BL (2000) One-step inactivation of chromosomal genes in Escherichia coli K12 using PCR products. Proc Natl Acad Sci USA 97:6640–6645 10. Douglas AE (1998) Nutritional interactions in insect–microbial symbiosis: aphids and their symbiotic bacteria Buchnera. Annu Rev Entomol 43:17–37

Coexpression of Buchnera trpE with groESL 11. Douglas AE, Prosser WA (1992) Synthesis of the essential amino acid tryptophan in the pea aphid (Acyrthosiphon pisum) symbiosis. J Insect Physiol 38:565–568 12. Dunbar HE, Wilson ACC, Ferguson NR, Moran NA (2007) Aphid thermal tolerance is governed by a point mutation in bacterial symbionts. PLoS Biol 5:e96 13. Fares MA, Ruiz-González MX, Moya A, Elena SF, Barrio E (2002) Endosymbiotic bacteria: groEL buffers against deleterious mutations. Nature 417:398 14. Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symp Ser 41:95–98 15. Herbeck JT, Funk DJ, Degnan PH, Wernegreen JJ (2003) A conservative test of genetic drift in the endosymbiotic bacterium Buchnera: slightly deleterious mutations in the chaperonin groEL. Genetics 165:1651–1660 16. Kerner MJ, Naylor DJ, Ishihama Y, Maier T, Chang H-C, Stines AP, Georgopoulos C, Frishman D, Hayer-Hartl M, Mann M, Hartl FU (2005) Proteome-wide analysis of chaperonin-dependent protein folding in Escherichia coli. Cell 122:209–220 17. Komaki K, Ishikawa H (1999) Intracellular bacterial symbionts of aphids possess many genomic copies per bacterium. J Mol Evol 48:717–722 18. Lai C-Y, Baumann L, Baumann P (1994) Amplification of trpEG: adaptation of Buchnera aphidicola to an endosymbiotic association with aphids. Proc Natl Acad Sci USA 91:3819–3823 19. Lai C-Y, Baumann P, Moran N (1996) The endosymbiont (Buchnera sp.) of the aphid Diuraphis noxia contains plasmids consisting of trpEG and tandem repeats of trpEG pseudogenes. Appl Environ Microbiol 62:332–339 20. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 21. Latorre A, Gil R, Silva FJ, Moya A (2005) Chromosomal stasis versus plasmid plasticity in aphid endosymbiont Buchnera aphidicola. Heredity 95:339–347 22. Mira A, Moran NA (2002) Estimating population size and transmission bottlenecks in maternally transmitted endosymbiotic bacteria. Microb Ecol 44:137–143 23. Moran NA (1996) Accelerated evolution and Muller’s ratchet in endosymbiotic bacteria. Proc Natl Acad Sci USA 93:2873–2878 24. Ohtaka C, Nakamura H, Ishikawa H (1992) Structures of chaperonins from an intracellular symbiont and their functional expression in Escherichia coli groE mutants. J Bacteriol 174: 1869–1874

703 25. Pérez-Brocal V, Gil R, Ramos S, Lamelas A, Postigo M, Michelena JM, Silva FJ, Moya A, Latorre A (2006) A small microbial genome: the end of a long symbiotic relationship? Science 314:312–313 26. Plague GR, Dale C, Moran NA (2003) Low and homogeneous copy number of plasmid-borne symbiont genes affecting host nutrition in Buchnera aphidicola of the aphid Uroleucon ambrosiae. Mol Ecol 12:1095–1100 27. Rouhbakhsh D, Lai C-Y, von Dohlen CD, Clark MA, Baumann L, Baumann P, Moran NA, Voegtlin DJ (1996) The tryptophan biosynthetic pathway of aphid endosymbionts (Buchnera): genetics and evolution of plasmid-associated anthranilate synthase (trpEG) within the aphididae. J Mol Evol 42:414–421 28. Rouhbakhsh D, Clark MA, Baumann L, Moran NA, Baumann P (1997) Evolution of the tryptophan biosynthetic pathway in Buchnera (aphid endosymbionts): studies of plasmid-associated trpEG within the genus Uroleucon. Mol Phylog Evol 8:167–176 29. Shigenobu S, Watanabe H, Hattori M, Sakaki Y, Ishikawa H (2000) Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS. Nature 407:81–86 30. Maisnier-Patin S, Roth JR, Fredriksson Å, Nyström T, Berg OG, Andersson DI (2005) Genomic buffering mitigates the effects of deleterious mutations in bacteria. Nat Genet 37:1376–1379 31. Sun Z, Scott DJ, Lund PA (2003) Isolation and characterization of mutants of GroEL that are fully functional as single rings. J Mol Biol 332:715–728 32. Tamas I, Klasson L, Canbäck B, Näslund AK, Eriksson AS, Wernegreen JJ, Sandström JP, Moran NA, Andersson SGE (2002) 50 million years of genomic stasis in endosymbiotic bacteria. Science 296:2376–2379 33. van Ham RC, Kamerbeek J, Palacios C, Rausell C, Abascal F, Bastolla U, Fernández JM, Jiménez L, Postigo M, Silva FJ, Tamames J, Viguera E, Latorre A, Valencia A, Morán F, Moya A (2003) Reductive genome evolution in Buchnera aphidicola. Proc Natl Acad Sci USA 100:581–586 34. Wernegreen JJ, Moran NA (1999) Evidence for genetic drift in endosymbionts (Buchnera): analysis of protein-coding genes. Mol Biol Evol 16:83–97 35. Wernegreen JJ, Moran NA (2000) Decay of mutualistic potential in aphid endosymbionts through silencing of biosynthetic loci: Buchnera of Diuraphis. Proc R Soc Lond B 267:1423–1431 36. Wernegreen JJ, Moran NA (2001) Vertical transmission of biosynthetic plasmids in aphid endosymbionts (Buchnera). J Bacteriol 183:785–790