Staphylococcus aureus and Escherichia coli have disparate dependences on KsgA for growth and ribosome biogenesis
© O'Farrell and Rife; licensee BioMed Central Ltd. 2012
Received: 12 July 2012
Accepted: 11 October 2012
Published: 24 October 2012
The KsgA methyltransferase has been conserved throughout evolution, methylating two adenosines in the small subunit rRNA in all three domains of life as well as in eukaryotic organelles that contain ribosomes. Understanding of KsgA’s important role in ribosome biogenesis has been recently expanded in Escherichia coli; these studies help explain why KsgA is so highly conserved and also suggest KsgA’s potential as an antimicrobial drug target.
We have analyzed KsgA’s contribution to ribosome biogenesis and cell growth in Staphylococcus aureus. We found that deletion of ksgA in S. aureus led to a cold-sensitive growth phenotype, although KsgA was not as critical for ribosome biogenesis as it was shown to be in E. coli. Additionally, the ksgA knockout strain showed an increased sensitivity to aminoglycoside antibiotics. Overexpression of a catalytically inactive KsgA mutant was deleterious in the knockout strain but not the wild-type strain; this negative phenotype disappeared at low temperature.
This work extends the study of KsgA, allowing comparison of this aspect of ribosome biogenesis between a Gram-negative and a Gram-positive organism. Our results in S. aureus are in contrast to results previously described in E. coli, where the catalytically inactive protein showed a negative phenotype in the presence or absence of endogenous KsgA.
Ribosome biogenesis in bacteria involves a small number of extra-ribosomal biogenesis factors . Depletion or loss of many of these factors leads to impaired ribosome assembly, and in many cases leads to growth defects or even loss of virulence in pathogenic bacteria. Understanding ribosome biogenesis in bacteria is an active field of study; the bulk of this work has taken place in the model organism Escherichia coli, a Gram-negative γ-proteobacterium, while lesser study has occurred in other organisms, principally the Gram-positive organism Bacillus subtilis. One ribosome biogenesis factor in particular, KsgA, has been studied intensively for many years in E. coli. KsgA dimethylates each of two adenosines in the 3’-proximal helix (helix 44) of the small subunit rRNA  and serves as an important checkpoint in the assembly of the 30S subunit . Cells lacking functional KsgA are often disadvantaged for growth when compared to wild-type cells. Specifically, knockout or mutation of ksgA in the organisms E. coli, B. subtilis, Mycobacterium tuberculosis, Yersinia pseudotuberculosis, Chlamydia trachomatis and Erwinia amylovora is deleterious to cell growth, producing strains that either grow slower than or are unable to compete efficiently with wild-type strains. In addition, knockout of ksgA in Y. pseudotuberculosis confers an attenuated virulence phenotype on the knockout strain ; inactivating mutations of ksgA in the plant pathogen E. amylovora decrease virulence .
A key observation to come out of the body of work on KsgA is that overexpression of catalytically inactive KsgA produces a dominant negative phenotype, being deleterious to both ribosome biogenesis and cell growth, thus suggesting KsgA might serve as a potential antimicrobial drug target . In this context KsgA and its role in ribosome biogenesis and growth have been studied most extensively in E. coli. While ksgA gene knockouts have been tangentially studied in other organisms, no systematic study has been made of KsgA and its role in ribosome biogenesis and growth in another bacterial organism. In order to expand our knowledge of this system, we have extended studies of KsgA into the important Gram-positive human pathogen Staphylococcus aureus.
Knockout of ksgAleads to a cold-sensitive phenotype
Antibiotic resistance of RN4220 and Δ ksgA strains
Doubling times of RN4220 and ΔksgA strains
Doubling time (min)
408.2 ± 18.2
473.0 ± 17.2
82.1 ± 4.1
93.4 ± 2.0
48.5 ± 0.6
50.2 ± 2.2
39.2 ± 1.8
39.4 ± 1.7
50.6 ± 1.5
54.3 ± 3.5
Our laboratory previously observed that knockout of ksgA in E. coli led to a difference in sensitivity to aminoglycoside antibiotics . Specifically, the ΔksgA strain was more sensitive to the 4,6 class of aminoglycosides and less sensitive to 4,5-aminoglycosides, with no change in sensitivity to the aminoglycoside streptomycin. We performed a similar experiment in S. aureus, growing the RN and ΔksgA strains on increasing amounts of the antibiotics kanamycin (a 4,6 aminoglycoside), paromomycin (a 4,5-aminoglycoside) and streptomycin (Table 1). The ΔksgA strain was more sensitive to both kanamycin and paromomycin, with no change in sensitivity to streptomycin.
Overexpression of catalytically inactive KsgA is deleterious
Doubling times of RN4220 and Δ ksgA strains containing pCN constructs
Doubling time (min)
95.5 ± 13.8
40.5 ± 2.7
94.9 ± 11.0
39.6 ± 2.4
92.6 ± 9.5
39.2 ± 4.7
106.1 ± 11.6
41.4 ± 2.7
100.0 ± 8.0
38.3 ± 2.5
111.3 ± 11.5
51.0 ± 2.3
Overexpression of wild-type KsgA did not affect cell growth under any of the conditions we tested. Overexpression of the E79A mutant in cells lacking ksgA had a negative impact on doubling time, but only in the absence of WT enzyme. This effect was seen at 37°C but not at 25°C. In the RN strain, which expresses endogenous KsgA, overexpression of mutant protein did not significantly affect cell growth.
The existence of the ksgA gene was established about forty years ago in E. coli. It was shown to be the sole methyltransferase that converts two adjacent 16S rRNA adenosines (A1518 and A1519, E. coli numbering) into N6,N6-dimethyladenosines , modifications that appeared to hold wide phylogenetic distribution. It is now known that those modifications and the responsible methyltransferase are all but universally conserved throughout life, thus making KsgA (known as Dim1 in eukaryotes and archaea) a genetic element of the last universal common ancestor. This level of conservation, coupled with the knowledge that KsgA can be dispensed with in several bacteria, albeit with obvious growth defects [3–8], formed the basis of a sharp paradox. If KsgA was not essential, why was it universally conserved? Since evolution is not sentimental, the cellular importance of KsgA and Dim1 was certain but remained to be discovered. In time the stated paradox has partially unraveled. In Saccharomyces cerevisiae (and most likely other eukaryotic organisms) Dim1 is an important member of the rRNA processome, and loss of Dim1 leads to the accumulation of aberrant rRNA species at the expense of functional ribosomes . In E. coli, KsgA serves as a gate-keeper to prevent improperly assembled pre-30S subunits from entering the translation cycle . Under normal conditions, KsgA only provides modest benefit to 30S maturation and function. However, KsgA’s importance becomes clear under stress conditions, such as growth at cold temperature.
In this work, we sought to define the importance of KsgA to the survivability of the human pathogen S. aureus and to compare our results to those in the model organism E. coli. Somewhat surprisingly, we found that S. aureus has a lesser reliance on KsgA under the conditions tested. In E. coli, overexpression of KsgA rescued the cold-sensitive phenotype of ΔksgA cells at low temperature but was deleterious for cell growth at 37°C in both knockout and parental cells. Overexpression of a catalytically inactive mutant of KsgA, E66A, was deleterious in both strains at both temperatures, even in the presence of endogenous WT protein . We showed that in S. aureus the ksgA knock-out strain displayed a slow growth phenotype at low temperature when compared to the parental strain, similar to results in E. coli. However, unlike in E. coli, catalytic inactivation of KsgA’s enzymatic function has only mild phenotypic effects, and these effects are not dominant in the presence of WT KsgA. It is noteworthy that the negative growth effect was seen at 37°C but not at 25°C. This result was unexpected, both because ksgA knockout led to cold sensitivity and because negative effects in E. coli were exacerbated at low temperature; however, it is possible that growth at the lower temperature results in lower expression of the mutant protein and therefore a smaller negative effect.
In S. aureus, KsgA also appears to be less critical for the assembly of mature ribosomes. Experiments in E. coli showed that loss or inactivation of KsgA had obvious effects on ribosome biogenesis even under conditions where a growth phenotype was not apparent . In other words, ribosome biogenesis is sensitive to disruptions in KsgA function that don’t affect overall cell growth. We did not see this effect in S. aureus; knockout or inactivation of KsgA resulted in, at most, slight disruption of polysome profiles even under conditions where cell growth was slowed.
On the basis of the data presented here, it would appear that in S. aureus KsgA holds less promise as a drug target than in E. coli. However, we did observe that knockout of ksgA rendered S. aureus marginally more sensitive to clinically used aminoglycoside antibiotics, similar to results seen in E. coli. A1518 and A1519 are located distal to the aminoglycoside binding site on the small ribosomal subunit; we therefore hypothesize that effects on antibiotic sensitivity are indirect, likely caused either by conformational or dynamic changes that are propagated from the site of KsgA methylation to the aminoglycoside binding site. This experiment highlights an additional difference between E. coli and S. aureus ribosomes. While lack of methylation by KsgA leads to increased sensitivity to the 4,6 class of aminoglycosides in both organisms, we see opposite effects on 4,5 aminoglycoside sensitivity. Both the KsgA target site and the aminoglycoside binding site are among the most highly conserved rRNA sequences; it is thus intriguing that distinct effects are seen between the two organisms.
Although ribosome biogenesis has not been well-studied outside of the model organisms E. coli and, to a much lesser extent, B. subtilis, it is possible that reported differences in ribosome biogenesis between Gram-negative and Gram-positive organisms are representative of an evolutionary divergence between the two groups of bacteria. One such difference is the case of the ribonuclease RNase III. RNase III is an endonuclease that is involved in processing of the pre-rRNA transcript in both E. coli and B. subtilis. However, this enzyme is strictly essential in B. subtilis but not in E. coli. Additionally, inactivation of RNase III has different effects on the maturation of 16S rRNA in the two organisms . Further work is required to demonstrate whether these results are more broadly applicable in other bacterial species. Our work suggests differences in ribosome biogenesis between E. coli and S. aureus; it remains to be seen if the differing reliance on KsgA can be defined by a phylogenetic Gram-positive/Gram-negative split.
KsgA plays a key role in ribosome biogenesis in E. coli, which cannot be separated from its methyltransferase function . Further evidence of KsgA’s significance in Gram-negative organisms comes from virulence studies in pathogenic organisms. Disruption of ksgA in Y. pseudotuberculosis confers an attenuated virulence phenotype on the knockout strain , and this attenuated strain confers protection against subsequent challenge with the wild-type strain . Additionally, mutation of ksgA in the plant pathogen E. amylovora decreases virulence  and disruption of KsgA in S. Enteriditis reduces invasiveness . These studies affirm that KsgA may be a novel drug target in Gram-negative organisms.
Studies on KsgA’s role in virulence have not been done in Gram-positive organisms, although in addition to the modest growth defects seen in the S. aureus ΔksgA strain disruption of the ksgA gene in the Gram-negative Mycobacterium tuberculosis was shown to negatively affect bacterial growth on solid media . It should be noted that disruption of ksgA in Y. pseudotuberculosis produced only a slight growth defect and allowed the bacteria to survive in infected mice, even though the strain was not as virulent as the wild-type strain . Likewise, E. amylovora mutants showed reduced virulence despite only small growth defects in vitro and the ability to grow in infected tissue . Further studies will be required to show whether KsgA is similarly correlated with virulence in Gram-positive organisms.
Given the vital role that the ribosome plays in the cell, it is unsurprising that it is an important target for antibiotic drugs . Although current antibiotic strategies are directed at the functioning of the ribosome, it has been suggested that the ribosome assembly presents a target for novel drug discovery . In support of this hypothesis, knockout of the non-essential ribosome biogenesis factors KsgA and YjeQ, a small-subunit associated GTPase, has been shown to affect bacterial virulence [6, 8, 17]. Therefore, a full understanding of these and other ribosome biogenesis factors in a variety of organisms is critical.
We have extended the study of KsgA into S. aureus and found that KsgA is not as critical for bacterial growth and ribosome biogenesis as was previously shown to be the case in E. coli, although the ΔksgA knockout does have some negative effects. Additionally, overexpression of the catalytically inactive mutant did not have a dominant effect on growth or ribosome biogenesis in the presence of wild-type protein. Although knockout and mutation of KsgA did not lead to severe growth defects, work in Y. pseudotuberculosis and E. amylovora suggests that small growth defects in vitro may correlate with larger effects on virulence. Many researchers have suggested that targeting virulence may be a better strategy for antimicrobial therapy than targeting cell growth or viability [18, 19]. We believe that further research on the role of KsgA in the virulence of S. aureus and other pathogens will prove instructive and may provide a viable drug development target.
Strains and plasmids
The RN4220 strain, the pCN51 expression vector, and genomic DNA from S. aureus strain 8325 were gifts from Dr. Gordon Archer, Virginia Commonwealth University. The pMAD shuttle vector for knockout of ksgA was a gift from Dr. Gail Christie, Virginia Commonwealth University.
We constructed a ksgA knockout of the S. aureus RN4220 strain according to the method of Arnaud et al. Allelic replacement was performed using the primers in Additional file 3; chromosomal knockout was confirmed by PCR.
The ksgA gene was amplified from genomic DNA from S. aureus strain 8325, adding a ribosome binding sequence to ensure translation; primers used for cloning are shown in Additional file 3. The resulting fragment was subcloned into the pCN51 expression vector to produce pCN-WT. Mutagenesis was performed on this plasmid according to the Stratagene Quikchange protocol to produce pCN-E79A. The pCN51 constructs were transformed into strain RN4220 (RN) and the ksgA knockout strain (ΔksgA) by electroporation. Expression of active protein from the pCN51-KsgA plasmid was confirmed in the ΔksgA strain by the kasugamycin resistance assay (Additional file 4), as well as by showing that 30S subunits purified from this strain were not able to be further methylated by KsgA (Additional file 5).
Antibiotic resistance assay
Cells were grown in tryptic soy broth (TSB) at 37°C overnight; saturated culture was subcultured to an OD600 of 0.02 in TSB and grown with shaking at 225 rpm to an OD600 of 0.6-0.8. The culture was then diluted 1:100 and plated onto varying concentrations of antibiotic. Plates were grown at 37°C overnight; the minimal inhibitory concentration (MIC) was read as the lowest concentration of antibiotic which prevented growth.
30S subunits were prepared from the S. aureus RN4220 and ΔksgA strains as well as from an E. coli wild-type strain. Cells were grown in TSB (S. aureus) or LB (E. coli) to mid-log phase. Cells were harvested and the cell pellet resuspended in Buffer I (50 mM Tris, pH 7.4, 100 mM NH4Cl, 10 mM MgOAc, and 6 mM β-mercaptoethanol). Glass beads (0.090-0.135 mm, Thomas Scientific) were added to a final concentration of 1 mg/μl and the suspension was vortexed for 10 minutes. The lysates were cleared by centrifugation at 4°C, layered onto 1.1 M sucrose in Buffer II (50 mM Tris, pH 7.4, 1 M NH4Cl, 10 mM MgOAc, and 6 mM β-mercaptoethanol), and spun in a 70Ti rotor at 35,000 rpm for 22 hours at 4°C. The pellet of ribosomal material was resuspended in Buffer III (50 mM Tris, pH 7.4, 500 mM NH4Cl, 2 mM MgOAc, and 6 mM β-mercaptoethanol) and loaded onto a 10-40% sucrose gradient in Buffer III. The gradients were spun in an SW-28 rotor at 19,000 rpm for 17 hours at 4°C and 30S fractions were collected, dialyzed into Buffer K (50 mM Tris, pH 7.4, 500 mM NH4Cl, 2 mM MgOAc, and 6 mM β-mercaptoethanol) and stored at -80°C. E. coli KsgA was purified as previously described; activity assays were performed as previously described .
Cells were grown in TSB at 37°C overnight; cultures of strains transformed with pCN constructs included erythromycin (10 μg/ml). Saturated culture was subcultured to an OD600 of 0.1 in TSB; media contained cadmium (2 μM) and erythromycin (10 μg/ml) for experiments with the pCN constructs. Cells were incubated with shaking (225 rpm) and the OD600 was monitored. Data were fit to an exponential growth model using the Graphpad Prism software and doubling times were calculated from the equation Y = Y0. × eK× X.
Cells were grown in TSB, containing cadmium (2 μM) and erythromycin (50 μg/ml) as appropriate, to mid-log phase. Cells were harvested and the cell pellet resuspended in Buffer PA μg/ml (20 mM Tris, pH 7.8, 100 mM NH4Cl, 10 mM MgCl2, and 6 mM β-mercaptoethanol). Glass beads (0.090-0.135 mm, Thomas Scientific) were added to a final concentration of 1 mg/μl and the suspension was vortexed for 10 minutes. The lysates were cleared by centrifugation at 4°C and loaded onto a 10-40% sucrose gradient in Buffer PA. The gradients were spun in an SW-28 rotor at 19,000 rpm for 17 hours at 4°C. Gradients were analyzed at 254 nm using a Biocomp Piston Gradient Fractionator with a BIORAD Econo UV Monitor with a Full Scale of 1.0. Data were recorded using DataQ DI-158-UP data acquisition software and the 70S peaks were then normalized to 1.
The authors would like to thank Dr. Gail Christie and Dr. Gordon Archer for providing strains and plasmids and Kristin Lane and Dr. Sam Boundy for assistance in gene knockout and expression in S. aureus.
- Kaczanowska M, Ryden-Aulin M: Ribosome biogenesis and the translation process in Escherichia coli. Microbiol Mol Biol Rev. 2007, 71 (3): 477-494. 10.1128/MMBR.00013-07.PubMedPubMed CentralView ArticleGoogle Scholar
- Helser TL, Davies JE, Dahlberg JE: Mechanism of kasugamycin resistance in Escherichia coli. Nat New Biol. 1972, 235 (53): 6-9.PubMedView ArticleGoogle Scholar
- Connolly K, Rife JP, Culver G: Mechanistic insight into the ribosome biogenesis functions of the ancient protein KsgA. Mol Microbiol. 2008, 70 (5): 1062-1075. 10.1111/j.1365-2958.2008.06485.x.PubMedPubMed CentralView ArticleGoogle Scholar
- Ochi K, Kim JY, Tanaka Y, Wang G, Masuda K, Nanamiya H, Okamoto S, Tokuyama S, Adachi Y, Kawamura F: Inactivation of KsgA, a 16S rRNA methyltransferase, causes vigorous emergence of mutants with high-level kasugamycin resistance. Antimicrob Agents Chemother. 2009, 53 (1): 193-201. 10.1128/AAC.00873-08.PubMedPubMed CentralView ArticleGoogle Scholar
- Tufariello JM, Jacobs WR, Chan J: Individual Mycobacterium tuberculosis resuscitation-promoting factor homologues are dispensable for growth in vitro and in vivo. Infect Immun. 2004, 72 (1): 515-526. 10.1128/IAI.72.1.515-526.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Mecsas J, Bilis I, Falkow S: Identification of attenuated Yersinia pseudotuberculosis strains and characterization of an orogastric infection in BALB/c mice on day 5 postinfection by signature-tagged mutagenesis. Infect Immun. 2001, 69 (5): 2779-2787. 10.1128/IAI.67.5.2779-2787.2001.PubMedPubMed CentralView ArticleGoogle Scholar
- Binet R, Maurelli AT: The chlamydial functional homolog of KsgA confers kasugamycin sensitivity to Chlamydia trachomatis and impacts bacterial fitness. BMC Microbiol. 2009, 9: 279-10.1186/1471-2180-9-279.PubMedPubMed CentralView ArticleGoogle Scholar
- McGhee GC, Sundin GW: Evaluation of kasugamycin for fire blight management, effect on nontarget bacteria, and assessment of kasugamycin resistance potential in Erwinia amylovora. Phytopathology. 2011, 101 (2): 192-204. 10.1094/PHYTO-04-10-0128.PubMedView ArticleGoogle Scholar
- Zarubica T: Specificity determinants of ArmS, a ribosomal RNA methyltransferase that confers antibiotic resistance. PhD thesis. 2010, USA: Virginia Commonwealth University, Department of Biochemistry and Molecular BiologyGoogle Scholar
- Helser TL, Davies JE, Dahlberg JE: Change in methylation of 16S ribosomal RNA associated with mutation to kasugamycin resistance in Escherichia coli. Nat New Biol. 1971, 233 (35): 12-14.PubMedView ArticleGoogle Scholar
- Lafontaine D, Vandenhaute J, Tollervey D: The 18S rRNA dimethylase Dim1p is required for pre-ribosomal RNA processing in yeast. Genes Dev. 1995, 9 (20): 2470-2481. 10.1101/gad.9.20.2470.PubMedView ArticleGoogle Scholar
- Condon C: RNA processing and degradation in Bacillus subtilis. Microbiol Mol Biol Rev. 2003, 67 (2): 157-174. 10.1128/MMBR.67.2.157-174.2003.PubMedPubMed CentralView ArticleGoogle Scholar
- Bergman MA, Loomis WP, Mecsas J, Starnbach MN, Isberg RR: CD8(+) T cells restrict Yersinia pseudotuberculosis infection: bypass of anti-phagocytosis by targeting antigen-presenting cells. PLoS Pathog. 2009, 5 (9): e1000573-10.1371/journal.ppat.1000573.PubMedPubMed CentralView ArticleGoogle Scholar
- Shah DH, Zhou X, Kim HY, Call DR, Guard J: Transposon mutagenesis of Salmonella Enteritidis identifies genes that contribute to invasiveness in human and chicken cells and survival in egg albumen. Infect Immun. in pressGoogle Scholar
- McCoy LS, Xie Y, Tor Y: Antibiotics that target protein synthesis. Wiley Interdiscip Rev RNA. 2011, 2 (2): 209-232. 10.1002/wrna.60.PubMedView ArticleGoogle Scholar
- Comartin DJ, Brown ED: Non-ribosomal factors in ribosome subunit assembly are emerging targets for new antibacterial drugs. Curr Opin Pharmacol. 2006, 6 (5): 453-458. 10.1016/j.coph.2006.05.005.PubMedView ArticleGoogle Scholar
- Campbell TL, Henderson J, Heinrichs DE, Brown ED: The yjeQ gene is required for virulence of Staphylococcus aureus. Infect Immun. 2006, 74 (8): 4918-4921. 10.1128/IAI.00258-06.PubMedPubMed CentralView ArticleGoogle Scholar
- Clatworthy AE, Pierson E, Hung DT: Targeting virulence: a new paradigm for antimicrobial therapy. Nat Chem Biol. 2007, 3 (9): 541-548. 10.1038/nchembio.2007.24.PubMedView ArticleGoogle Scholar
- Barczak AK, Hung DT: Productive steps toward an antimicrobial targeting virulence. Curr Opin Microbiol. 2009, 12 (5): 490-496. 10.1016/j.mib.2009.06.012.PubMedPubMed CentralView ArticleGoogle Scholar
- Arnaud M, Chastanet A, Debarbouille M: New vector for efficient allelic replacement in naturally nontransformable, low-GC-content, gram-positive bacteria. Appl Environ Microbiol. 2004, 70 (11): 6887-6891. 10.1128/AEM.70.11.6887-6891.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- O’Farrell HC, Pulicherla N, Desai PM, Rife JP: Recognition of a complex substrate by the KsgA/Dim1 family of enzymes has been conserved throughout evolution. RNA. 2006, 12 (5): 725-733. 10.1261/rna.2310406.PubMedPubMed CentralView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.