Bacteriophages specific for Yersinia pestis are routinely used for plague diagnostics and could be an alternative to antibiotics in case of drug-resistant plague. A major concern of bacteriophage therapy is the emergence of phage-resistant mutants. The use of phage cocktails can overcome this problem but only if the phages exploit different receptors. Some phage-resistant mutants lose virulence and therefore should not complicate bacteriophage therapy.
The purpose of this work was to identify Y. pestis phage receptors using site-directed mutagenesis and trans-complementation and to determine potential attenuation of phage-resistant mutants for mice. Six receptors for eight phages were found in different parts of the lipopolysaccharide (LPS) inner and outer core. The receptor for R phage was localized beyond the LPS core. Most spontaneous and defined phage-resistant mutants of Y. pestis were attenuated, showing increase in LD50 and time to death. The loss of different LPS core biosynthesis enzymes resulted in the reduction of Y. pestis virulence and there was a correlation between the degree of core truncation and the impact on virulence. The yrbH and waaA mutants completely lost their virulence.
We identified Y. pestis receptors for eight bacteriophages. Nine phages together use at least seven different Y. pestis receptors that makes some of them promising for formulation of plague therapeutic cocktails. Most phage-resistant Y. pestis mutants become attenuated and thus should not pose a serious problem for bacteriophage therapy of plague. LPS is a critical virulence factor of Y. pestis.
Citation: Filippov AA, Sergueev KV, He Y, Huang X-Z, Gnade BT, Mueller AJ, et al. (2011) Bacteriophage-Resistant Mutants in Yersinia pestis: Identification of Phage Receptors and Attenuation for Mice. PLoS ONE 6(9): e25486. https://doi.org/10.1371/journal.pone.0025486
Editor: Deepak Kaushal, Tulane University, United States of America
Received: February 23, 2011; Accepted: September 6, 2011; Published: September 28, 2011
Copyright: © 2011 Filippov et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was supported by the Defense Threat Reduction Agency, Joint Science and Technology Office, Medical S&T Division (http://www.dtra.mil/Research.aspx), grant #2.10052_08_WR_B. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Yersinia pestis is the causative agent of plague circulating in natural foci among multiple species of wild animals. Humans usually become infected via fleabites. Plague has killed many millions of people during three pandemics and is now recognized as a re-emerging disease. Y. pestis is considered a biothreat agent belonging to the most dangerous group of pathogens, CDC Category A –. Multiple drug-resistant strains of Y. pestis have been isolated from patients with bubonic plague. One of them carried genes of high-level resistance to 12 antibiotics, including virtually all of the drugs commonly used for plague prophylaxis and treatment . The emergence of multidrug-resistant strains requires the immediate development of alternative methods of plague therapy including the possible use of bacteriophages . There are more and more publications on successful bacteriophage treatment of various experimental and veterinary infections and some promising clinical trials in humans (for reviews, see –).
Many bacteriophages (phages) are capable of lysing Y. pestis –, and some of them have a high potential to be used for antibacterial treatment of plague. Several phages have been shown to be highly lytic and specific for Y. pestis and are routinely used for diagnosis of plague , . They include the Pokrovskaya , , , , φA1122 , , , , L-413C , , , , and Yep-phi  phages. The φA1122 ,  and L-413C  phages have been recently used for the development of improved Y. pestis detection techniques using fluorescently labeled φA1122  or amplification of phage DNA monitored by qPCR . The genomes of φA1122 , L-413C , and Yep-phi  have been sequenced and no genes potentially detrimental for mammals were found in them.
A major concern regarding the use of phages in the treatment of infectious diseases still remains the emergence of phage-resistant mutants –. This resistance can arise due to the alteration or loss of the bacterial cell surface receptor, blocking the receptor by the bacterial extracellular matrix, inhibition of phage DNA penetration, production of modified restriction endonucleases degrading phage DNA, or inhibition of phage intracellular development . Mutations affecting phage receptors represent the most frequent cause of phage resistance –. Phage receptors are diverse bacterial surface-exposed molecules including many outer membrane proteins, sugar residues in the O antigen or lipopolysaccharide (LPS) core, teichoic acids, polysaccharides of the capsule or slime layer, or components of flagella and pili , . In pathogenic bacteria, many phage receptors are essential virulence factors, some of which were identified through the selection of phage-resistant mutations. The receptors include capsular polysaccharides –, adhesion and invasion factors –, a protein involved in intracellular growth , and, very commonly, different components of LPS –. If a phage receptor lies in a surface structure important for virulence, the phage-resistant mutants resulting from a loss or alteration of the receptor will be avirulent or attenuated. Such mutant bacterial clones will then be eliminated from the host by the immune system and therefore should not present a problem if they arise during bacteriophage treatment , , , . Another way to overcome the problem of phage resistance is to use therapeutic phage cocktails, which requires avoiding cross-resistance, i.e. bacterial mutants resistant to one phage must remain sensitive to the other components of the cocktail , , . Thus, ideally, bacteriophages comprising the cocktail should employ different bacterial cell surface receptors , .
The purpose of this work was to study spontaneous and site-directed (non-polar) mutants of Y. pestis resistant to various bacteriophages, to identify phage receptors and to determine if phage-resistant mutants are impaired in virulence for mice. Nine phages capable of lysing Y. pestis were tested: L-413C, P2 vir1, φJA1, φA1122, Pokrovskaya, T7Yp, Y, PST, and R. We identified six receptors for eight bacteriophages in different parts of the LPS core. Altogether, the phages used at least seven different receptors in Y. pestis. Therefore, one can formulate a plague therapeutic cocktail containing several highly lytic phages exploiting different Y. pestis cell surface receptors. Most of the Y. pestis phage-resistant mutants were attenuated and thus should not present a problem for phage therapy. A direct correlation was observed between the length of the LPS core and virulence. The yrbH and waaA mutants were completely avirulent and thus LPS was shown to be a critical virulence factor of Y. pestis.
Materials and Methods
Animal studies were conducted in compliance with the Animal Welfare Act and other federal, state and local laws and regulations regarding animal work, in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. All experiments with mice were performed under the animal use protocol approved by the Walter Reed Army Institute of Research Institutional Animal Care and Use Committee (Protocol #IB05-09).
Bacteria, bacteriophages and growth media
Bacterial strains and bacteriophages used in this work are listed in Tables 1 and 2, respectively. High-concentration stocks of the φA1122, Pokrovskaya, Y, and T7Yp phages were prepared using Y. pestis CO92 Pgm− cultured at 28°C by the low multiplicity of infection method . The same technique was used for large-scale isolation of the PST and R phages but they were grown on Y. pseudotuberculosis PB1. Finally, this method was used to propagate T7 on E. coli C600 at 37°C. The L-413C, φJA1, and P2 vir1 phage stocks were prepared on Y. pestis CO92 Pgm− as described previously  using different temperatures of incubation, 28°C for L-413C and φJA1 or 37°C for P2 vir1. Bacteria and bacteriophages were grown in liquid Brain Heart Infusion (BHI) medium (Becton-Dickinson, Franklin Lakes, New Jersey, USA) or on BHI plates containing 1.5% and 0.7% Bacto Agar (Becton-Dickinson) except E. coli strains, which were grown in LB broth (Becton-Dickinson) or on LB agar plates. The slants with tryptose blood agar (Becton-Dickinson) containing 5% defibrinated sheep blood (Lampire Biological Laboratories, Pipersville, Pennsylvania, USA) were used for streaking Y. pestis frozen stocks to prepare the challenge suspensions in animal experiments .
SM buffer  was used for bacteriophage storage and dilutions. Phage plaque assays were performed by the double-layer agar method as described earlier  with overnight incubation for L-413C, P2 vir1, φJA1, Y, PST, and R, or 5–6 h incubation for φA1122, Pokrovskaya, T7, and T7Yp. Plaquing (plating) efficiency tests were done as described previously  using an incubator temperature of 28°C. Phage adsorption experiments were performed according to .
Isolation of spontaneous phage-resistant mutants and determination of the mutation frequencies
Spontaneous Y. pestis mutants resistant to bacteriophages were isolated from avirulent strains CO92 Pgm−, KIM D27, and A1122, as well as from wild-type virulent CO92 by plating on double-layer agar with the phage, 109 PFU (plaque-forming unit) per plate. To determine the mutation frequencies per cell per generation, ten isolated colonies were picked out from each strain, inoculated into ten tubes with BHI, grown with agitation until approximate concentration of 109 CFU/ml (colony-forming unit) and diluted in ten-fold increments down to 10−7. 0.1-ml aliquots from the 100–10−3 dilutions were inoculated onto the double-layer plates and were incubated at 28°C for six days because some of the phage-resistant mutants grow very slowly. Simultaneously, 0.1-ml aliquots of 10−5, 10−6, and 10−7 dilutions were plated on BHI agar without phage (three plates for each dilution), for determination of CFU numbers. To calculate mutation frequencies, mean numbers of mutants in one ml of culture were divided by mean total numbers of live bacterial cells. The phage resistant colonies were picked up and purified by two passages on double-layer plates with the phage and were used in further experiments.
Site-directed mutagenesis of Y. pestis CO92 LPS genes was performed by one-step PCR-mediated in-frame replacement with the Tn903 kanamycin cassette as described earlier . This kanamycin resistance gene, aph(3′)-Ia , does not provide cross-resistance to streptomycin and gentamicin ; the procedure was approved by the WRAIR Institutional Biosafety Committee (the protocol MUA #125-a, June 14, 2004) and by CDC (Memorandum of February 25, 2008). Primers used for mutagenesis and verifying the correctness of replacement were designed using NetPrimer online program (Premier Biosoft International, http://www.premierbiosoft.com) and are listed in Table S1. The mutagenesis primers were designed so that the replacements were non-polar and preserved adjacent genes including ribosome-binding sites intact.
Testing potential phenotypic alterations in Y. pestis phage-resistant mutants
Spontaneous and induced Y. pestis phage-resistant mutants were tested for plasmid content by the method of Kado and Liu , for pigmentation by plating on Congo red agar , for low calcium response on magnesium-oxalate agar plates , and for possible loss of natural polymyxin resistance associated with rough type of LPS on BHI agar with polymyxin B (10 and 150 U/ml; Sigma-Aldrich, St. Louis, Missouri, USA) and without the antibiotic.
Molecular cloning and trans-complementation
Y. pestis LPS biosynthesis genes were cloned into a pBAD-TOPO vector using pBAD TOPO® TA Expression Kit (Invitrogen) according to recommendations of the supplier. Primers for PCR cloning are presented in Table S2. All genes were cloned under the control of pBAD arabinose promoter, except Y. pestis yrbH, which was cloned under the control of its own promoter, as described elsewhere . Two tandem stop codons (see Table S2) were placed immediately upstream of the yrbH promoter to prevent the expression of a truncated protein from the pBAD arabinose promoter. The knockout mutants were cured of a mutagenesis helper plasmid pKD46  by growing at increased temperature, at 38°C. The recombinant plasmids were introduced into corresponding mutants using electroporation. All mutants were also transformed with DNA of the intact pBAD (to exclude a possible impact of the vector plasmid on LPS expression and phage susceptibility). Transformants that acquired a recombinant plasmid or the vector without an insert were tested for plaquing efficiencies on the double-layer BHI plates supplemented with 10 mM of L-arabinose to induce the pBAD promoter, as well as for the LPS size (see below).
SDS-PAGE analysis of LPS
The presence and size of LPS core in Y. pestis strains was analyzed as described previously , with the following minor modifications. Y. pestis strains were grown overnight at 28°C in BHI. After the lysis and deproteinization, LPS was visualized by fluorescent staining using Pro-Q® Emerald 300 Lipopolysaccharide Gel Stain Kit (Invitrogen).
Infection of mice
Y. pestis strains for animal challenge experiments were isolated from single colonies, tested for the presence of three virulence plasmids , pigmentation , and low calcium response  and kept at −80°C in BHI with 10% dimethyl sulfoxide. The challenge culture suspensions were prepared from the frozen stocks according to  but using normal saline instead of HIB broth. Mouse lethality was determined following subcutaneous inoculation of 100 µl saline containing each of various 10-fold concentrations of Y. pestis into groups of ten 8-week-old female BALB/c mice. The mice were observed twice a day until 14 days postinfection. Median lethal doses (LD50) were calculated by the method of Reed and Muench . In addition to the number of deaths, the mean time to death (MTD) was recorded.
Spontaneous bacteriophage resistance mutations in Y. pestis
First, we determined the frequencies of spontaneous mutations resulting in resistance of Y. pestis towards four different phages, L-413C, φA1122, Pokrovskaya, and Y (Table 3). Y. pestis CO92 showed different mutation frequencies to the L-413C, Pokrovskaya, and Y phages. No φA1122-resistant mutations were observed. Six independent mutants resistant to L-413C (two for each of the three strains, CO92 Pgm−, KIM D27, and A1122) were isolated and their phenotypes tested. Five out of six clones lost the native resistance to polymyxin B associated with the rough form of Y. pestis LPS, suggesting LPS core damage , . Since some phage-resistant mutants of Y. pestis previously showed pleiotropic effects on several phenotypic traits , , we tested L-413C-resistant mutants for plasmid content, pigmentation, and low calcium response and observed no alterations in these traits in comparison with the parental strains. All six mutants were found to remain sensitive to φA1122.
Identification of bacteriophage receptors
To identify bacteriophage receptors, a series of defined non-polar Y. pestis CO92 Pgm− mutants affected in genes encoding the synthesis of different parts of LPS (Table 1) were obtained by one-step knockout kanamycin cassette mutagenesis . Several LPS biosynthesis genes were cloned in a pBAD vector (Invitrogen) and used for trans-complementation tests. Both knockout mutants and trans-complemented clones were tested for susceptibility to nine bacteriophages using comparative plaquing efficiency assays (see Tables 4 and 5). Such approach has been recently successfully used for identification of T7 receptor in E. coli K-12 . We additionally tested the knockout mutants for the efficiency of phage adsorption in comparison with the parental strain CO92 Pgm− (Table 6). The results of mutagenesis were verified by PCR (data not shown) and fluorescent staining of LPS that demonstrated an increase in electrophoretic mobility or the loss of LPS bands in the mutants (Fig. 1).
20 µl of deproteinized lysates  were loaded on 14% SDS-PAGE, run and stained with Pro-Q® Emerald 300 Lipopolysaccharide Gel Stain Kit (Invitrogen). 1 and 12, CandyCane™ Glycoprotein Molecular Weight Standards (Invitrogen); 2, CO92 Pgm−; 3, CO92 Pgm− yrbH, 4, CO92 Pgm− lpxM; 5, CO92 Pgm− waaA, 6, CO92 Pgm− wabC; 7, CO92 Pgm− wabD; 8, CO92 Pgm− waaL; 9, CO92 Pgm− waaF; 10, CO92 Pgm− hldE; 11, CO92 Pgm− yrbH (pYrbH).
The fine structure of LPS in all kinds of Y. pestis knockout mutants used in our work has been previously characterized in details using mass spectroscopy analysis in strains KIM6 of bv. Medievalis  (for the waaA and yrbH mutants) and 231 of bv. Antiqua  (for the rest of mutants). We used BLAST analysis  for comparison of 11 LPS biosynthesis proteins (YrbH, WaaA, HldE, WaaF, WaaL, WabC, WabD, WecA, WaaQ, WaaE, and WaaC) including all those, the genes for which were knocked out in this work, in 23 published whole genome sequences of Y. pestis strains of various biovars deposited in the NCBI databases (http://www.ncbi.nlm.nih.gov/). They showed 100% identity, or, in rare cases, 99% identity of amino acid sequences. This suggested that the LPS core structure of diverse Y. pestis strains is highly conserved and thus we could use the structural data of other authors , .
In contrast to the data of another group based on multi-step suicide plasmid mutagenesis , , in our experiments lpxM (msbB) mutants with altered lipid A acylation were susceptible to L-413C, as well as to eight other phages (Table 4). Mutations in genes responsible for early steps of LPS core synthesis, yrbH and waaA, resulted in the loss of both outer and inner LPS core  (see Fig. 2). The coreless mutants had slower growth , especially in early generations, and displayed resistance to eight phages of the nine tested (Table 4), suggesting that the receptors for these eight bacteriophages lie in the LPS core. Mutants affected in carbohydrate residues of the main LPS core chain (hldE, waaF, and waaL) were completely resistant to the L-413C and P2 vir1 phages (Table 4 and Fig. 2). The waaL mutants are deficient in O antigen ligase, but in Y. pestis lacking O antigen the WaaL protein forms a glycosidic bond between l-glycero-d-manno-heptopyranose (HepII) and a terminal outer core residue, β-N-acetyl-d-glucosamine (GlcNAc) . Since waaL mutants carried the minimal LPS outer core defect (lack of GlcNAc) and still lost phage susceptibility, we concluded that a critical part of the receptor for L-413C and P2 vir1 is the terminal GlcNAc residue of the LPS outer core (Fig. 2). This was confirmed by trans-complementation of waaL mutation with the waaL gene cloned in a pBAD vector, which resulted in full restoration of susceptibility to L-413C and P2 vir1 (Fig. S1 and Table 5). Adsorption tests (Table 6) showed that not only waaL but also waaF mutant was affected in the phage binding, and hldE mutant completely lost the ability to adsorb to L-413C and P2 vir1. Thus HepII/HepIII and HepI/Glc residues are also involved in the formation of L-413C and P2 vir1 receptors (Fig. 2; the receptor is outlined with bold purple line).
The structure of LPS core is presented according to . Kdo, 2-keto-3-deoxy-octulosonic acid; Ko, d-glycero-d-talo-oct-2-ulosonic acid; Hep, heptulose (ketoheptose); Glc, glucose; Gal, galactose; GlcNAc, N-acetylglucosamine. Black dashed lines designate the sites where the gene product, a corresponding glycosyltransferase, forms a glycosidic bond. The yrbH mutations affecting Kdo synthesis have the same effect on the LPS structure as waaA mutations, and defect in the hldE gene involved in ADP-l,d-heptose synthesis causes the same phenotype as the waaC mutation . The critical sugar residues for certain phage receptors have solid color fill matching with that of phage designations and with the color of lines outlining the receptors. The phage receptors are outlined with bold purple (for the L-413C and P2 phages), dashed red (for φJA1), bold green (for PST), dashed green (for Pokrovskaya), dashed blue (for T7Yp and Y), and bold red (for φA1122) lines.
The waaL mutation resulted in reduction of the φJA1 phage efficiency of plating by more than two logs (Table 4) suggesting that GlcNAc is involved in φJA1 reception (Fig. 2). Both waaF and hldE mutants further decreased plaquing efficiency by two more orders of magnitude (Table 4) and thus HepII/HepIII residues also take part in the formation of φJA1 receptor. Finally, φJA1 did not form plaques on the coreless mutants yrbH and waaA providing at least five log difference in the efficiencies of plating in comparison with the waaF and hldE mutants. This suggests that the φJA1 phage receptor includes both LPS outer and inner core sugars and Kdo/Ko residues form a critical part of the receptor (Fig. 2).
Using the same approach (see Tables 4, 5, and 6), we localized five more receptors for six other phages (Table 2 and Fig. 2). Since WaaA forms both GlcN-Kdo and Kdo-Ko bonds and we did not mutagenize waaE and waaQ genes, we could not identify a single sugar residue most critical for phage binding in these cases but mapped most important receptor components to three different pairs of adjacent inner core residues, Kdo/Ko (for φA1122 and φJA1), HepI/Glc (for T7Yp and Y), and HepII/HepIII (for Pokrovskaya and PST). This was confirmed by trans-complementation tests with recombinant plasmids carrying the yrbH, waaA, hldE, and waaF genes (Table 5). The restoration of LPS core with normal mobility was shown in the complemented mutants (see Fig. 1).
The adsorption assays (Table 6) confirmed the data of plating efficiency tests on the most important receptor components and revealed some additional residues involved in the formation of phage receptors. For example, the experiments with the φA1122 phage confirmed that Kdo/Ko residues are a critical part of the receptor: the yrbH mutant lost 78% of its phage binding activity. Additionally, the hldE mutant showed some decreased φA1122 adsorption, suggesting that HepI/Glc residues are also involved in the phage reception. The adsorption experiments also allowed us to expand the receptors for the Pokrovskaya, T7Yp and Y phages (Table 6 and Fig. 2). The adsorption tests with the φJA1 and PST phages failed; we could not determine any binding even to Y. pestis CO92 Pgm− strain supposed to contain the intact bacteriophage receptors. The cause could be a low efficiency binding resulting in disruption of the phage-bacterium bonds when starting centrifugation of the bacterial suspensions infected with these two phages. Our data showed that the receptor for the R phage is located beyond the LPS core (see Tables 4 and 6 and Fig. 2).
Attenuation of Y. pestis phage-resistant mutants in mice
To determine if bacteriophage resistance mutations impact on Y. pestis virulence, we reproduced site-directed mutagenesis  of five genes (waaL, waaF, hldE, waaA, and yrbH) in the fully virulent wild type strain CO92 (Pgm+) (Table 1). Additionally, eight spontaneous mutants of CO92 resistant to the L-413C phage were isolated. All remained susceptible to φA1122. Seven of them became sensitive to polymyxin B. One Pmxs (S-2) and one Pmxr (S-7) spontaneous mutants were selected for further work. LD50 doses were determined for the two spontaneous and five site-directed phage-resistant mutants in comparison with the parental strain CO92 in BALB/c mice after subcutaneous injection (see Table 7). Six phage-resistant mutants of seven tested (86%) were shown to be attenuated, with significantly higher LD50 and longer mean times to death (MTD). Three of them (S-2, yrbH-1, and waaA-1) completely lost virulence. Only one spontaneous mutant resistant to L-413C (S-7) was fully virulent. Both yrbH and waaA mutants lack the LPS outer and inner core (; see also Fig. 1, lanes 3 and 5) and they both became fully avirulent. The same results were obtained in a repeated experiment with independently derived mutants, yrbH-2 and waaA-2 (Table 7).
The emergence of multidrug-resistant strains of Y. pestis  may result in outbreaks of untreatable bubonic and pneumonic plague with very high mortality and suggests the paramount importance of finding adequate alternatives to antibiotics including bacteriophages , which have already shown a high efficacy for the therapy of several other infections –. One of the major problems that can arise when using phages as therapeutics is the development of phage-resistant mutants –, most frequently resulting from the alteration or loss of the bacterial cell surface receptor –. To solve this problem, one can use therapeutic cocktails consisting of phages that exploit different bacterial receptors , , . Another way is to use the phages adsorbing on pathogenicity factors so that the receptor mutations resulting in phage resistance simultaneously attenuate the bacterium, which is subsequently eliminated from the host by the immune system , , , . Given the high potential of bacteriophages as alternative therapies against drug-resistant plague, here we addressed the problem of phage-resistant mutants in Y. pestis and both strategies of its overcoming: identified Y. pestis receptors for a number of bacteriophages and determined if the phage-resistant mutants become attenuated.
Nine phages able to lyse Y. pestis were tested. They included four plague diagnostic phages, L-413C , , , , , φA1122 , , , –, Pokrovskaya , , , , and Y , ; two pseudotuberculosis diagnostic phages also capable of lysing most Y. pestis strains, PST , – and R , , –; as well as P2 vir1, a virulent mutant of well-known coliphage P2  previously shown to lyse Y. pestis at 37°C , . Additionally, we isolated a new Y. pestis-specific phage from sewage, φJA1, and also selected a host range mutant of enterobacteriophage T7, T7Yp, which showed both a striking specificity and high lytic activity towards Y. pestis. First, we demonstrated a significant difference in the frequencies of spontaneous mutations of Y. pestis resistance to phages belonging to different morphologic and taxonomic groups. For example, resistance to L-413C was a rather frequent event (about 10−4 per cell per generation), whereas we could not isolate any spontaneous mutants resistant to φA1122, suggesting that these two phages use different bacterial surface receptors for their adsorption. Most L-413C-resistant clones became sensitive to polymyxin B, indicating that they have an LPS core defect , . All mutants resistant to L-413C remained sensitive to φA1122. This finding also suggested that L-413C and φA1122 exploit different surface structures as the receptors.
Site-directed mutagenesis of different LPS genes, mainly those encoding for glycosyltransferases involved in the synthesis of the LPS outer and inner core, and trans-complementation with the cloned genes followed by efficiency of plating and adsorption tests (Tables 4, 5, 6) allowed us to localize six Y. pestis cell surface receptors for bacteriophages in different parts of the LPS core (Fig. 2). It was demonstrated that the terminal residue of the outer core, N-acetylglucosamine, is a critical part of L-413C and P2 vir1 receptor. The most important parts of the rest of the receptors were mapped to three different pairs of adjacent Y. pestis LPS inner core residues, Kdo/Ko (for the φA1122 and φJA1 phages), HepI/Glc (for T7Yp and Y), and HepII/HepIII (for Pokrovskaya and PST). Apart from the critical receptor components, we found some additional LPS sugar residues involved in the phage adsorption that made up six different receptor structures for eight phages (Fig. 2). The seventh receptor, for the R phage, was shown to lie beyond the LPS core.
The location of phage receptors in different LPS core residues can at least partially explain the difference in the frequencies of phage resistance mutations. For example, L-413C provided the highest frequency of resistance, most probably due to the largest number of genes, mutations in which result in the receptor loss (waaL, wecA, waaF, waaC, hldE, waaA, yrbH and other genes responsible for Kdo biosynthesis; see Fig. 2, Refs. 51 and 80). At the same time, φA1122 binds to Kdo and HepI/Glc residues that can be affected by mutations in a relatively small number of genes.
Therefore, nine bacteriophages tested in this work can use at least seven different receptors in Y. pestis. Based on the principle of formulating optimal therapeutic cocktails from the phages that employ different cell surface receptors , ,  and on our results, we recommend six bacteriophages as the best candidates for a plague therapeutic cocktail: φA1122, Pokrovskaya, Y, T7Yp, φJA1, and R. They use different receptors in Y. pestis. The best candidate in this list is the φA1122 phage previously shown to be rather specific , –, highly lytic  and active against virtually all known Y. pestis isolates , . We identified the Y. pestis receptor for φA1122 and demonstrated that the mutation of resistance to this phage is a very rare event. There are only two known Y. pestis strains resistant to φA1122 out of thousands tested . It is also important that φA1122 genome has been sequenced and no potentially detrimental genes were found in it . The Pokrovskaya , , , Y , , and R , ,  phages have been shown to be highly active against Y. pestis (especially, Pokrovskaya and Y) and to have broad ranges of susceptible strains. Our results provided important new information about the receptors for these bacteriophages, making them also promising as potential components of plague therapeutic cocktails. Finally, two new phage isolates, T7Yp (a coliphage T7 host range mutant) and φJA1 (obtained from sewage) demonstrate a high specificity and significant lytic potential towards Y. pestis, particularly, T7Yp (A.A. Filippov, Y. He, and K.V. Sergueev, unpublished data) and use known receptors on the bacterial cell surface, which also makes them good candidates for future plague therapeutic cocktails. For T7Yp, it is also important that its parental phage, T7, has been sequenced and no genes potentially harmful for warm-blooded animals were identified .
Determination of median lethal doses in BALB/c mice by subcutaneous route of administration for two spontaneous Y. pestis L-413C-resistant mutants and seven defined LPS-affected mutants resistant to different phages showed that eight of them (89%) were attenuated, displaying a significant increase in LD50 and MTD (Table 7). Five mutants (S-2, waaA-1, waaA-2, yrbH-1, and yrbH-2) became completely avirulent. Only one spontaneous mutant resistant to L-413C (S-7) that retained polymyxin resistance was fully virulent. This mutation seems to be unrelated to the phage receptor, instead affecting some steps of phage lytic cycle following its adsorption, e.g., DNA replication or phage assembly . The attenuation of most phage-resistant strains suggests that the emergence of such mutants in vivo should result in their elimination by the immune system and thus should not significantly decrease the efficiency of phage therapy.
Site-directed mutations in each of the three tested genes encoding for glycosyltransferases that incorporate different sugar residues into the LPS core (waaA, waaF, and waaL), as well as in two genes essential for Kdo (yrbH) and ADP-l,d-heptose (hldE) biosynthesis were shown to affect Y. pestis virulence (Table 7). It has been previously found that mutants of Yersinia enterocolitica , Salmonella enterica , Actinobacillus pleuropneumoniae , and Burkholderia cenocepacia  with different LPS core defects demonstrate reduced virulence. In Y. pestis, in vivo maintenance of a tetra-acylated structure of lipid A in LPS has been shown to be essential for mouse virulence after subcutaneous infection . Recently, five LPS core mutants of Y. pestis 231 (biovar Antiqua) have been tested for virulence in mice and guinea-pigs challenged subcutaneously, wabD, waaL, waaQ, waaE, and hldE . Of them, waaL and waaE mutants showed a slight increase in LD50 for mice and guinea pigs, whereas hldE was attenuated by four logs. Our results on the waaL and hldE mutants are in agreement with the previous data . Additionally, our data (Table 7) show that there is a clear correlation between the degree of LPS core truncation and attenuation. Some spontaneous undefined Y. pestis mutants resistant to the Pokrovskaya and/or L-413C phages have been observed to have reduced virulence or to become avirulent for mice or guinea pigs . Here we showed that the most likely cause of attenuation of such mutants is the LPS core truncation.
It is important that both yrbH and waaA mutants (two pairs of independently derived clones) lacking the LPS core ,  demonstrated full avirulence (Table 7). These two types of mutants have different mechanisms of the loss of LPS core. The waaA product is a transferase catalyzing Kdo glycosylation of lipid A , while the YrbH protein is arabinose 5-phosphate isomerase that converts ribulose 5-phosphate into arabinose 5-phosphate, which is the first committed step in the Kdo biosynthesis , . The complete loss of virulence in the two types of LPS coreless mutants suggests that LPS is a critical virulence factor of Y. pestis. These results are of great importance for the molecular pathogenesis of plague because previously only the loss of major components of type III secretion system encoded on the plasmid pCD1 –, a chromosomal gene cluster of siderophore yersiniabactin synthesis and reception – or a gene encoding for the NlpD lipoprotein  have been known to completely attenuate Y. pestis strains.
Trans-complementation of L-413C phage receptor defect. A, B, and C: L-413C plaquing on Y. pestis CO92 Pgm−, CO92 Pgm− waaL, and CO92 Pgm− waaL (pWaaL), respectively.
Primers for mutagenesis of Y. pestis LPS genes and verifying the sizes of amplicons. Notes: a50-bp flanking sequences of Y. pestis chromosomal DNA providing site-specific insertion of kanamycin cassette are shown in lower case type. bF1 and R1 primers targeted Y. pestis DNA on the gene flanks and provided amplification of a fragment with changed size due to the replacement with the Kmr gene. cInt (internal) primers amplified the corresponding intact Y. pestis gene and were used to exclude gene duplications and emergence of merodiploids. dThe kan-start and kan-stop primers targeted the Kmr gene and were used to amplify the DNA novel joints with corresponding F1 and R1 primers, respectively.
Primers for cloning of Y. pestis LPS genes. Notes: aTwo tandem stop codons in the yrbH forward primer are underscored. Lower case letters in the rest of forward primers designate modified ribosome binding site according to recommendations of the supplier of the TOPO cloning system (Invitrogen).
We wish to thank Mrs. Stacy M. Banko-Crawford and Ms. Dipali J. Patel for excellent technical assistance and Dr. Richard Calendar for critical reading of the manuscript and important suggestions. Drs. Richard Calendar, Martin E. Schriefer, Ian J. Molineux, and Bakyt B. Atshabar are gratefully acknowledged for providing bacteriophages. The findings and opinions expressed herein belong to the authors and do not necessarily reflect the official views of the WRAIR, the U.S. Army or the Department of Defense.
Conceived and designed the experiments: AAF KVS X-ZH. Performed the experiments: AAF KVS YH X-ZH BTG AJM CMF-P. Analyzed the data: AAF KVS X-ZH. Contributed reagents/materials/analysis tools: AAF YH MPN. Wrote the paper: AAF MPN.
- 1. Perry RD, Fetherston JD (1997) Yersinia pestis – etiologic agent of plague. Clin Microbiol Rev 10: 35–66.
- 2. Gage KL, Kosoy MY (2005) Natural history of plague: perspectives from more than a century of research. Annu Rev Entomol 50: 505–528.
- 3. Inglesby TV, Dennis DT, Henderson DA, Bartlett JG, Ascher MS, et al. (2000) Plague as a biological weapon: medical and public health management. Working Group on Civilian Biodefense. JAMA 283: 2281–2290.
- 4. Welch TJ, Fricke WF, McDermott PF, White DG, Rosso ML, et al. (2007) Multiple antimicrobial resistance in plague: an emerging public health risk. PLoS One 2(3): e309.
- 5. Anisimov AP, Amoako KK (2006) Treatment of plague: promising alternatives to antibiotics. J Med Microbiol 55: 1461–1475.
- 6. Housby JN, Mann NH (2009) Phage therapy. Drug Discov Today 14: 536–540.
- 7. Deresinski S (2009) Bacteriophage therapy: exploiting smaller fleas. Clin Infect Dis 48: 1096–1101.
- 8. Kutter E, De Vos D, Gvasalia G, Alavidze Z, Gogokhlia L, et al. (2010) Phage therapy in clinical practice: treatment of human infections. Curr Pharm Biotechnol 11: 69–86.
- 9. Pokrovskaya MP (1929) A plague bacteriophage in dead susliks. Gigiena Epidemiol 12: 31–34.
- 10. Advier M (1933) Etude d'un bactériophage antipesteux. Bull Soc Pathol Exotiques 26: 94–99.
- 11. Burrows TW (1962) Phage and bacteriocin investigations with Pasteurella pestis and other bacteria. Nature 193: 397–398.
- 12. Knapp W (1963) Varying behavior of Pasteurella phages. Zentralbl Bakteriol 190: 39–46.
- 13. Hertman I (1964) Bacteriophage common to Pasteurella pestis and Escherichia coli. J Bacteriol 88: 1002–1005.
- 14. Larina VS, Anisimov PI, Adamov AK (1970) A novel strain of plague bacteriophage for identification of Pasteurella pestis. Probl Particularly Dangerous Infect 11: 132–136.
- 15. Arutyunov YI (1970) Biological characteristics of plague and pseudotuberculosis bacteriophages. Zh Mikrobiol Epidemiol Immunobiol 8: 106–111.
- 16. Plotnikov OP, Larina VS, Kondrashin YI, Vorob'yova TV (1982) Spectrum of lytic activity of bacteriophages isolated from soil of rodents' holes. In: Anisimov PI, Karaseva ZN, Konnov NP, editors. Selection and Genetics of Causative Agents of Particularly Dangerous Infections. Saratov: Institute “Microbe” Press. pp. 56–61.
- 17. Novoseltsev NN, Marchenkov VI (1990) Yersinia pestis phage of a new serovar. Zh Mikrobiol Epidemiol Immunobiol 11: 9–12.
- 18. Garcia E, Elliott JM, Ramanculov E, Chain PS, Chu MC, et al. (2003) The genome sequence of Yersinia pestis bacteriophage φA1122 reveals an intimate history with the coliphage T3 and T7 genomes. J Bacteriol 185: 5248–5262.
- 19. Garcia E, Chain P, Elliott JM, Bobrov AG, Motin VL, et al. (2008) Molecular characterization of L-413C, a P2-related plague diagnostic bacteriophage. Virology 372: 85–96.
- 20. Schwudke D, Ergin A, Michael K, Volkmar S, Appel B, et al. (2008) Broad-host-range Yersinia phage PY100: genome sequence, proteome analysis of virions, and DNA packaging strategy. J Bacteriol 190: 332–342.
- 21. Zhao X, Wu W, Qi Z, Cui Y, Yan Y, et al. (2010) The complete genome sequence and proteomics of Yersinia pestis phage Yep-phi. J Gen Virol 92: 216–221.
- 22. Chu MC (2000) Laboratory manual of plague diagnostic tests. Fort Collins: Centers for Disease Control and Prevention. 129 p.
- 23. Naumov AV, Samoilova LV, editors. (1992) Manual on prophylaxis of plague. Saratov: Russian Research Anti-Plague Institute “Microbe” Press. 278 p.
- 24. Bobrov AG, Kirillina OA, Filippov AA, Kutyrev VV (1999) Restriction mapping of DNA of plague diagnostic phages Pokrovskaya and L-413C. Probl Particularly Dangerous Infect 79: 138–144.
- 25. Gunnison JB, Larson A, Lazarus AS (1951) Rapid differentiation between Pasteurella pestis and Pasteurella pseudotuberculosis by action of bacteriophage. J Infect Dis 88: 254–255.
- 26. Schofield DA, Molineux IJ, Westwater C (2009) Diagnostic bioluminescent phage for detection of Yersinia pestis. J Clin Microbiol 47: 3887–3894.
- 27. Sergueev KV, He Y, Borschel RH, Nikolich MP, Filippov AA (2010) Rapid and sensitive detection of Yersinia pestis using amplification of plague diagnostic bacteriophages monitored by real-time PCR. PLoS ONE 5(6): e11337.
- 28. Smith HW, Huggins MB (1983) Effectiveness of phages in treating experimental Escherichia coli diarrhoea in calves, piglets and lambs. J Gen Microbiol 129: 2659–2675.
- 29. Levin BR, Bull JJ (2004) Population and evolutionary dynamics of phage therapy. Nat Rev Microbiol 2: 166–173.
- 30. Gill JJ, Hyman P (2010) Phage choice, isolation, and preparation for phage therapy. Curr Pharm Biotechnol 11: 2–14.
- 31. Labrie SJ, Samson JE, Moineau S (2010) Bacteriophage resistance mechanisms. Nat Rev Microbiol 8: 317–327.
- 32. Lindberg AA (1973) Bacteriophage receptors. Annu Rev Microbiol 27: 205–241.
- 33. Heller KJ (1992) Molecular interaction between bacteriophage and the gram-negative cell envelope. Arch Microbiol 158: 235–248.
- 34. Smith HW, Huggins MB (1982) Successful treatment of experimental Escherichia coli infections in mice using phage: its general superiority over antibiotics. J Gen Microbiol 128: 307–318.
- 35. Smith HW, Huggins MB, Shaw KM (1987) The control of experimental Escherichia coli diarrhoea in calves by means of bacteriophages. J Gen Microbiol 133: 1111–1126.
- 36. Pickard D, Toribio AL, Petty NK, van Tonder A, Yu L, et al. (2010) A conserved acetyl esterase domain targets diverse bacteriophages to the Vi capsular receptor of Salmonella enterica serovar Typhi. J Bacteriol 192: 5746–5754.
- 37. Pruzzo C, Valisena S, Satta G (1983) Laboratory and wild-type Klebsiella pneumoniae strains carrying mannose-inhibitable adhesins and receptors for coliphages T3 and T7 are more pathogenic for mice than are strains without such receptors. Infect Immun 39: 520–527.
- 38. Ricci V, Piddock LJ (2010) Exploiting the role of TolC in pathogenicity: identification of a bacteriophage for eradication of Salmonella serovars from poultry. Appl Environ Microbiol 76: 1704–1706.
- 39. Begum YA, Chakraborty S, Chowdhury A, Ghosh AN, Nair GB, et al. (2010) Isolation of a bacteriophage specific for CS7-expressing strains of enterotoxigenic Escherichia coli. J Med Microbiol 59: 266–272.
- 40. Spears PA, Suyemoto MM, Palermo AM, Horton JR, Hamrick TS, et al. (2008) A Listeria monocytogenes mutant defective in bacteriophage attachment is attenuated in orally inoculated mice and impaired in enterocyte intracellular growth. Infect Immun 76: 4046–4054.
- 41. Zinder ND (1958) Lysogenization and superinfection immunity in Salmonella. Virology 5: 291–326.
- 42. Wright A (1971) Mechanism of conversion of the salmonella O antigen by bacteriophage ε34. J Bacteriol 105: 927–936.
- 43. Schoonejans E, Expert D, Toussaint A (1987) Characterization and virulence properties of Erwinia chrysanthemi lipopolysaccharide-defective, φEC2-resistant mutants. J Bacteriol 169: 4011–4017.
- 44. Al-Hendy A, Toivanen P, Skurnik M (1991) Expression cloning of the Yersinia enterocolitica O:3 rfb gene cluster in Escherichia coli K12. Microb Pathog 10: 47–59.
- 45. Zhang L, Skurnik M (1994) Isolation of an R− M+ mutant of Yersinia enterocolitica serotype O:8 and its application in construction of rough mutants utilizing mini-Tn5 derivatives and lipopolysaccharide-specific phage. J Bacteriol 176: 1756–1760.
- 46. Petty NK, Foulds IJ, Pradel E, Ewbank JJ, Salmond GP (2006) A generalized transducing phage (φIF3) for the genomically sequenced Serratia marcescens strain Db11: a tool for functional genomics of an opportunistic human pathogen. Microbiology 152: 1701–1708.
- 47. Petty NK, Toribio AL, Goulding D, Foulds I, Thomson N, et al. (2007) A generalized transducing phage for the murine pathogen Citrobacter rodentium. Microbiology 153: 2984–2988.
- 48. Goodridge LD (2010) Designing Phage Therapeutics. Curr Pharm Biotechnol 11: 15–27.
- 49. Welkos S, Pitt ML, Martinez M, Friedlander A, Vogel P, et al. (2002) Determination of the virulence of the pigmentation-deficient and pigmentation/plasminogen activator-deficient strains of Yersinia pestis in non-human primate and mouse models of pneumonic plague. Vaccine 20: 2206–2214.
- 50. Parkhill J, Wren BW, Thomson NR, Titball RW, Holden MT, et al. (2001) Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413: 523–527.
- 51. Anisimov AP, Dentovskaya SV, Kondakova AN, Lindner B, Shaikhutdinova RZ, et al. (2010) Yersinia pestis lipopolysaccharide in host-pathogen interactions. In: Shafferman A, Ordentlich A, Velan B, editors. The Challenge of Highly Pathogenic Microorganisms: Mechanisms of Virulence and Novel Medical Countermeasures. London, , New York: Springer. pp. 77–87.
- 52. No authors listed (1992) Centers for Disease Control and Prevention. Pneumonic plague – Arizona. JAMA 268: 2146–2147.
- 53. Brubaker RR (1969) Mutation rate to nonpigmentation in Pasteurella pestis. J Bacteriol 98: 1404–1406.
- 54. Jawetz E, Meyer KF (1943) Avirulent strains of Pasteurella pestis. J Infect Dis 73: 124–143.
- 55. Burrows TW, Bacon GW (1960) V and W antigens in strains of Pasteurella pseudotuberculosis. Br J Exp Pathol 41: 38–44.
- 56. Appleyard RK (1954) Segregation of new lysogenic types during growth of a doubly lysogenic strain derived from Escherichia coli K-12. Genetics 39: 440–452.
- 57. Herrero M, de Lorenzo V, Timmis KN (1990) Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in gram-negative bacteria. J Bacteriol 172: 6557–6567.
- 58. Sunshine MG, Thorn M, Gibbs W, Calendar R, Kelly B (1971) P2 phage amber mutants: characterization by use of a polarity suppressor. Virology 46: 691–702.
- 59. Bertani LE (1957) The effect of the inhibition of protein synthesis on the establishment of lysogeny. Virology 4: 53–71.
- 60. Kerby GP, Gowdy RA, Dillon ES, Csaky TZ, Sharp DG, et al. (1949) Purification, pH stability and sedimentation properties of the T7 bacteriophage of Escherichia coli. J Immunol 63: 93–107.
- 61. Knapp W, Zwillenberg LO (1964) Morphological differences between Pasteurella bacteriophages. Arch Gesamte Virusforsch 14: 563–566.
- 62. Ackermann HW, Poty F (1969) Relationship between coliphages T2 and 3, and phage PST of Pasteurella pseudotuberculosis. Rev Can Biol 28: 201–204.
- 63. Gurleva GG, Arutyunov YI, Khaliapina EE (1981) Serologic relationship and specificity of action of pseudotuberculosis and coli-dysentery phages. Zh Mikrobiol Epidemiol Immunobiol 5: 65–68.
- 64. Kotlyarova RI (1956) Pseudotuberculosis bacteriophage and its characteristics. Proceedings of Research Anti-Plague Institute of Caucasus and Transcaucasia. Stavropol: pp. 234–241.
- 65. Shashaev MA (1966) Comparative characterization of plague and pseudotuberculosis phages. Production of bacterial preparations for prophylaxis and diagnostics of particularly dangerous infections. Saratov: pp. 350–356.
- 66. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor: Cold Spring Harbor Laboratory Press.
- 67. Anderson GW, Leary SEC, Williamson ED, Titball RW, Welkos SL, et al. (1996) Recombinant V antigen protects mice against pneumonic and bubonic plague caused by F1-capsule-positive and -negative strains of Yersinia pestis. Infect Immun 64: 4580–4585.
- 68. Datsenko KA, Wanner BL (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 97: 6640–6645.
- 69. Oka A, Sugisaki H, Takanami M (1981) Nucleotide sequence of the kanamycin resistance transposon Tn903. J Mol Biol 147: 217–226.
- 70. Mingeot-Leclercq M-P, Glupczynski Y, Tulkens PM (1999) Aminoglycosides: activity and resistance. Antimicrob Agents Chemother 43: 727–737.
- 71. Kado C, Liu S-T (1981) Rapid procedure for detection and isolation of large and small plasmids. J Bacteriol 145: 1365–1373.
- 72. Surgalla MJ, Beesley ED (1969) Congo red-agar plating medium for detecting pigmentation in Pasteurella pestis. Appl Microbiol 18: 834–837.
- 73. Higuchi K, Smith JL (1961) Studies on the nutrition and physiology of Pasteurella pestis. VI. A differential plating medium for the estimation of the mutation rate to avirulence. J Bacteriol 81: 605–608.
- 74. Tan L, Darby C (2006) Yersinia pestis YrbH is a multifunctional protein required for both 3-deoxy-D-manno-oct-2-ulosonic acid biosynthesis and biofilm formation. Mol Microbiol 61: 861–870.
- 75. Batson HC (1956) An introduction to statistics in the medical sciences. Minneapolis: Burgess Publishing Co.
- 76. Anisimov AP, Dentovskaya SV, Titareva GM, Bakhteeva IV, Shaikhutdinova RZ, et al. (2005) Intraspecies and temperature-dependent variations in susceptibility of Yersinia pestis to the bactericidal action of serum and to polymyxin B. Infect Immun 73: 7324–7331.
- 77. Lebedeva SA (2000) Yersinia pestis variants resistant to diagnostic bacteriophage and problems connected with them. Zh Mikrobiol Epidemiol Immunobiol 3: 99–104.
- 78. Feodorova VA, Pan'kina LN, Savostina EP, Kuznetsov OS, Konnov NP, et al. (2009) Pleiotropic effects of the lpxM mutation in Yersinia pestis resulting in modification of the biosynthesis of major immunoreactive antigens. Vaccine 27: 2240–2250.
- 79. Qimron U, Marintcheva B, Tabor S, Richardson CC (2006) Genomewide screens for Escherichia coli genes affecting growth of T7 bacteriophage. Proc Natl Acad Sci USA 103: 19039–19044.
- 80. Tan L, Darby C (2005) Yersinia pestis is viable with endotoxin composed of only lipid A. J Bacteriol 187: 6599–6600.
- 81. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215: 403–410.
- 82. Feodorova VA, Pan'kina LN, Savostina EP, Sayapina LV, Motin VL, et al. (2007) A Yersinia pestis lpxM-mutant live vaccine induces enhanced immunity against bubonic plague in mice and guinea pigs. Vaccine 25: 7620–7628.
- 83. Dunn JJ, Studier FW (1983) Complete nucleotide sequence of bacteriophage T7 DNA and the locations of T7 genetic elements. J Mol Biol 166: 477–535.
- 84. Skurnik M, Venho R, Bengoechea JA, Moriyón I (1999) The lipopolysaccharide outer core of Yersinia enterocolitica serotype O:3 is required for virulence and plays a role in outer membrane integrity. Mol Microbiol 31: 1443–1462.
- 85. Yethon JA, Gunn JS, Ernst RK, Miller SI, Laroche L, et al. (2000) Salmonella enterica serovar Typhimurium waaP mutants show increased susceptibility to polymyxin and loss of virulence in vivo. Infect Immun 68: 4485–4491.
- 86. Ramjeet M, Deslandes V, St Michael F, Cox AD, Kobisch M, et al. (2005) Truncation of the lipopolysaccharide outer core affects susceptibility to antimicrobial peptides and virulence of Actinobacillus pleuropneumoniae serotype 1. J Biol Chem 280: 39104–39114.
- 87. Loutet SA, Flannagan RS, Kooi C, Sokol PA, Valvano MA (2006) A complete lipopolysaccharide inner core oligosaccharide is required for resistance of Burkholderia cenocepacia to antimicrobial peptides and bacterial survival in vivo. J Bacteriol 188: 2073–2080.
- 88. Montminy SW, Khan N, McGrath S, Walkowicz MJ, Shar F, et al. (2006) Virulence factors of Yersinia pestis are overcome by a strong lipopolysaccharide response. Nat Immunol 7: 1066–1073.
- 89. Goguen JD, Yother J, Straley SC (1984) Genetic analysis of the low calcium response in Yersinia pestis Mu d1(Ap lac) insertion mutants. J Bacteriol 160: 842–848.
- 90. Straley SC, Bowmer WS (1986) Virulence genes regulated at the transcriptional level by Ca2+ in Yersinia pestis include structural genes for outer membrane proteins. Infect Immun 51: 445–454.
- 91. Kutyrev VV, Filippov AA, Shavina NY, Protsenko OA (1989) Genetic analysis and simulation of the virulence of Yersinia pestis. Mol Genet Mikrobiol Virusol 8: 42–47.
- 92. Jackson S, Burrows TW (1956) The pigmentation of Pasteurella pestis on a defined medium containing haemin. Brit J Exp Pathol 37: 570–576.
- 93. Iteman I, Guiyoule A, de Almeida AM, Guilvout I, Baranton G, et al. (1993) Relationship between loss of pigmentation and deletion of the chromosomal iron-regulated irp2 gene in Yersinia pestis: evidence for separate but related events. Infect Immun 61: 2717–2722.
- 94. Bearden SW, Fetherston JD, Perry RD (1997) Genetic organization of the yersiniabactin biosynthetic region and construction of avirulent mutants in Yersinia pestis. Infect Immun 65: 1659–1668.
- 95. Tidhar A, Flashner Y, Cohen S, Levi Y, Zauberman A, et al. (2009) The NlpD lipoprotein is a novel Yersinia pestis virulence factor essential for the development of plague. PLoS ONE 4(9): e7023.