• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jcmPermissionsJournals.ASM.orgJournalJCM ArticleJournal InfoAuthorsReviewers
J Clin Microbiol. Nov 2006; 44(11): 3849–3854.
Published online Aug 30, 2006. doi:  10.1128/JCM.00469-06
PMCID: PMC1698354

Variable-Number Tandem Repeats That Are Useful in Genotyping Isolates of Salmonella enterica subsp. enterica Serovars Typhimurium and Newport[down-pointing small open triangle]

Abstract

The genome of Salmonella enterica subsp. enterica serovar Typhimurium strain LT2 was analyzed for direct repeats, and 54 sequences containing variable-number tandem repeat loci were identified. Ten primer pairs that anneal upstream and downstream of each selected locus were designed and used to amplify PCR targets in isolates of S. enterica serovars Typhimurium and Newport. Four of the 10 loci did not show polymorphism in the length of products. Six loci were selected for analysis. Isolates of S. enterica serovars Typhimurium and Newport that were related to specific outbreaks and showed identical pulsed-field gel electrophoresis patterns were indistinguishable by the length of the six variable-number tandem repeats. Isolates that differed in their pulsed-field gel electrophoresis patterns showed polymorphism in variable-number tandem repeat profiles. Length of the products was confirmed by DNA sequence analysis. Only 2 of the 10 loci contained exact integers of the direct repeat. Eight loci contained partial copies. The partial copies were maintained at the ends of the variable-number tandem repeat loci in all isolates. In spite of having partial copies that were maintained in all isolates, the number of direct repeats at a locus was polymorphic. Six variable-number tandem repeat loci were useful in distinguishing isolates of S. enterica serovars Typhimurium and Newport that had different pulsed-field gel electrophoresis patterns and in identifying outbreak-associated cases that shared a common pulsed-field gel pattern.

Serovars of Salmonella enterica include pathogens that infect a wide range of hosts. Among these are serovars Typhimurium and Newport which together account for approximately 30 to 35% of human cases of salmonellosis (1).

Recently developed genotypic techniques have proven to be efficient in identifying clonally related strains of bacteria (11). These techniques include plasmid profiling, small-fragment restriction enzyme analysis, restriction fragment length polymorphism, ribotyping, arbitrary primer PCR, and large-fragment restriction enzyme analysis by pulsed-field gel electrophoresis (PFGE). Analysis by PFGE has been particularly useful in strain typing and in helping to identify links among patients that were previously not recognized by routine epidemiology.

PFGE offers a reliable method for assessing the clonal relatedness of Salmonella strains. PulseNet is a national surveillance network for five food-borne pathogens—Escherichia coli O157:H7, nontyphoidal Salmonella serotypes, Listeria monocytogenes, Campylobacter jejuni, and Shigella spp.—for which PFGE patterns collected at local sites are submitted to a national database for comparison (3, 15). A limitation of PFGE is that it is somewhat slow and labor intensive. Additionally, the PFGE patterns require sophisticated image analysis software to compare new patterns with those that are in the database. In order to make comparisons between the images produced at different laboratories, rigid standardization of the methodology is necessary.

Complete sequencing of the genome of S. enterica serovar Typhimurium LT2 (12) enabled the discovery of many variable-number tandem repeat (VNTR) loci that led to the development of multiple-locus VNTR analysis (9). The number of base pairs in each repeat unit (RU) varied from three to several hundred, and the copy number at each repeat locus varied from 2 to 15. Polymorphisms in tandem repeated mini-satellite loci caused by unequal crossing over are the basis for human forensic DNA typing and paternity testing (4). Genotypes are based on the number of repeating units (RUs) in many tandem repeat loci, each of which behaves independently. In many species of bacteria, polymorphism in VNTR loci has provided a means for strain typing (2, 5-7) (16).

In the present study, several previously uninvestigated VNTR loci were compared by PFGE for their ability to discriminate among isolates of serotypes Typhimurium and Newport. Moreover, sequence analysis was employed to investigate the organization of RUs in VNTR loci having different numbers of RUs.

MATERIALS AND METHODS

Salmonella isolates.

Strain S. enterica serovar Typhimurium LT2 was obtained from the American Type Culture Collection (ATCC 7000720 29946). Fifty isolates of S. enterica subsp. enterica including serovars Typhimurium (30 isolates) and Newport (20 isolates) underwent VNTR analysis. Each serovar Typhimurium and serovar Newport isolate was from a different patient and was collected from 1 July 1997 to 30 June 1998. The isolates were selected from 191 isolates of serovars Typhimurium and Newport that were submitted to the Arkansas Department of Health, where they were serotyped using polyvalent sera.

PFGE analysis.

All 191 isolates were grown overnight at 37°C in nutrient broth, sedimented by centrifugation, and washed in 100 mM Tris-HCl buffer (pH 7.5) containing 100 mM EDTA and 150 mM NaCl. Bacteria were embedded in 2.0% Incert FMC agarose plugs (Biowhittaker Laboratories, Rockland, ME). The plugs were digested for 15 h at 37°C with 1 mg/ml lysozyme and for an additional 24 h with proteinase K at 50°C. After extensive washing with CHEF-TE buffer (100 mM Tris-HCl buffer [pH 7.5] with 100 mM EDTA), the plugs were stored in CHEF-TE buffer at 4°C. Prior to restriction, the agarose plugs were washed for 3 h at 4°C in TM buffer (100 mM Tris-HCl [pH 8.0], 5 mM MgCl2) followed by 1 h at 4°C in XbaI buffer. The bacterial genomic DNA was restricted with genome-grade XbaI (100 units) (Promega, Madison, WI) for 16 h at 37°C. The restriction was terminated by adding EDTA (pH 8.0) to a final concentration of 1.0 mM. The plugs containing the restricted DNA were placed in a 1.0% agarose slab gel and electrophoresed in TBE buffer (100 mM Tris [pH 8.0], 100 mM boric acid, 20 mM EDTA) under pulsed-field conditions (24 h, 5.8 V/cm2 [200 V], 5/35 s, and 120° angle) on the CHEF-DRII system (Bio-Rad Laboratories, Richmond, CA). DNA molecular weight standards (48.5-kb bacteriophage lambda ladder; Bio-Rad) were included in each gel. The gels were stained with ethidium bromide, and the chromosomal fragments were visualized by UV transillumination.

An image of the gel was recorded in an Eagle Eye II gel documentation system (Stratagene, Cedar Creek, TX). Images of the individual lanes were analyzed with Molecular Analyst Fingerprinting software (Bio-Rad Laboratories). PFGE profiles of each isolate were stored in a database and compared with those of previously analyzed isolates. Patterns were compared using grouping analysis on the basis of their similarity index (percentage of isolates having fragments of identical size). When an indistinguishable pattern was observed, the isolates were compared directly by electrophoresis of XbaI-restricted DNA on the same pulsed-field gel. Isolates that displayed indistinguishable XbaI PFGE patterns were subjected to a secondary PFGE analysis after restriction with SpeI (50 units for 16 h at 37°C) (Promega) to confirm their identity.

Identification of tandem repeats and development of PCR primers.

The complete genome of S. enterica serovar Typhimurium strain LT2 was analyzed for repeats using DNAstar Genequest software (http://www.dnastar.com/). Tandem repeats were named according to their location on the S. enterica serovar Typhimurium LT2 genome. Sequences upstream and downstream of selected tandem repeat sequences were chosen for primer analysis. Primer sequences were chosen with the help of Primer Express oligonucleotide selection software (Applied Biosystems, Foster City, Calif.). The conditions for PCR amplification were chosen for each primer pair using S. enterica serovar Typhimurium strain LT2 DNA as the target.

DNA extraction.

Isolates were grown overnight at 37°C in nutrient broth, sedimented by centrifugation, and washed in 100 mM Tris-HCl buffer (pH 7.5) containing 100 mM EDTA and 150 mM NaCl. The bacteria were lysed with lysozyme (1 mg/ml at 37°C for 1 h) and sodium dodecyl sulfate (1.0%; at 60°C for 1 h). DNA was extracted by phenol-chloroform-isoamyl alcohol and ethanol precipitated.

PCR amplification.

Amplification was performed separately for each VNTR locus in a 25-μl mixture containing 10 ng of DNA, 1 U of Taq polymerase (Promega), 50 μM each of the pair of flanking primers for one of the VNTRs, a 1 mM concentration of each nucleotide triphosphate, and 1× PCR buffer with 10 mM MgCl2. Amplification was performed on a Gene Amp 9600 PCR thermocycler (Roche Diagnostics). An initial denaturation at 96°C for 3 min was followed by 25 cycles of a three-step cycling program (96°C for 1 min, 55°C for 1 min, and 72°C for 30 s) with a final elongation step of 72°C for 10 min. For two primer pairs (819457 and 2628542), the three-step cycling program was modified (96°C for 1 min, 65°C for 1 min, and 72°C for 30 s).

The PCR products were analyzed on 1% or 2% agarose gels in Tris-borate buffer at 10 V/cm. Gels were stained with 0.1% ethidium bromide, and DNA was detected by UV transillumination. Images were recorded on a gel documentation system. Each PCR gel analysis was used to detect a single VNTR locus and included a strain LT2 control, a clinical strain (688) that served as an additional control, a minus DNA control, and a size marker (Hyperladder IV; Bioline).

Estimating the number of RUs.

In order to estimate the number of RUs at each locus, the size of the PCR product was compared with that of the product from LT2 and with the size standards. The total length of the DNA flanking the repeat locus was subtracted from the length of the PCR product, and the difference between the sizes was divided by the length of the repeat unit in LT2.

DNA sequencing of VNTR loci.

To confirm the number of repeats in each variant at each locus and that the PCR products represented that locus, one PCR product representing each size variant at each locus in each of the two serovars was sequenced. DNA sequencing was performed on the amplified fragments by using an ABI dideoxy chain termination kit and an ABI model 3100 genetic analyzer (Applied Biosystems, Foster City, CA) in the DNA Sequencing Laboratory, Department of Microbiology and Immunology, at the University of Arkansas for Medical Sciences.

RESULTS

Fifty-four direct repeat loci were found in the genome of S. enterica serovar Typhimurium LT2. The size of the individual RUs ranged from 6 to 273. The number of RUs in each locus ranged from 2 to 15, and the size of the individual loci ranged from 25 to 753 bp. These loci were distributed around the genome from 323344 to 4810948 bp.

Ten direct repeat loci were selected, and primer pairs complementary to sequences in DNA flanking each direct repeat locus were selected (Table (Table1).1). The loci were selected in order to test a wide range of repeat unit sizes. The size of the repeat units ranged from 6 bp to 232 bp. The primers were tested in a PCR assay with 0.1, 1.0, 10, and 100 ng of S. enterica serovar Typhimurium LT2 DNA, and cycling times were varied to optimize the conditions. Optimal conditions were selected on the basis of the detection of a single PCR product of the expected size in reactions containing 1 to 10 ng of LT2 DNA. In some cases, optimization involved selecting an alternative primer in the DNA flanking the 3′ end of the direct repeat.

TABLE 1.
VNTR loci S. enterica serovar Typhimurium strain LT2 and primers selected for PCR

In order to determine whether the chosen direct repeat loci would be useful for VNTR analysis, the 10 loci were tested by PCR analysis of four clinical strains of serotype Typhimurium and four clinical strains of serotype Newport that had different PFGE patterns. Analysis demonstrated polymorphism in the size of PCR products at six loci (Table (Table2).2). Heterogeneity of PFGE patterns correlated with heterogeneity of PFGE pattern although not all VNTR loci contributed to these results. At four loci (2341937, 2531837, 4596419, and 4810628) all eight strains produced the same size PCR product.

TABLE 2.
Number of RUs at 10 VNTR loci of four clinical isolates of S. enterica serovars Typhimurium and Newport

The six loci that correlated with PFGE heterogeneity were analyzed further to see if they could be used to discriminate between the strains of serotype Typhimurium that were associated with an outbreak of salmonellosis that had shared a common PFGE pattern. Ten isolates of serotype Typhimurium that had different PFGE patterns that were not associated with the outbreak were also included (Fig. (Fig.1).1). The 16 clinical isolates demonstrated 11 different PFGE patterns (PFGE patterns B to L) (Table (Table3).3). The six epidemiologically linked isolates that shared the same PFGE pattern were not distinguishable by VNTR analysis at any of the six loci. The clinical isolates with different PFGE patterns were distinguished from one another and from the six epidemiologically linked isolates by differences in the sizes of the PCR products at loci 2730867, 3184543, and 3629542. Each of the isolates demonstrated a different combination of RUs (VNTR profile) at these three loci.

FIG. 1.
Isolates of S. enterica subsp. enterica serotypes Typhimurium (A) and Newport (B and C) having different PFGE patterns. (A) PFGE patterns of molecular weight marker (lanes 1 and 12) and 10 isolates (lanes 2 to 11). (B) Agarose gel electrophoresis of molecular ...
TABLE 3.
Number of RUs at six VNTR loci in 16 clinical isolates of S. enterica serovar Typhimurium

The same six loci were used to analyze 10 isolates of serotype Newport that had different PFGE patterns and 6 isolates that were associated with an outbreak of salmonellosis and shared a common PFGE pattern that was different from that of the others. The 16 clinical isolates demonstrated 11 different PFGE patterns (PFGE patterns M to W) (Table (Table4).4). The six epidemiologically linked isolates that shared the same PFGE pattern were not distinguishable by VNTR analysis at any of the six loci. Among the 10 clinical isolates with different PFGE patterns, five were distinguished from one another and from the six epidemiologically linked isolates by differences in the sizes of PCR products at loci 2730867, 3184543, and 3629542 (Fig. (Fig.1).1). Five of the isolates (with different PFGE patterns) shared two different VNTR profiles.

TABLE 4.
Number of RUs at six VNTR loci in 16 clinical isolates of S. enterica serovar Newport

In order to further test the ability of VNTR to discriminate among different isolates, Salmonella samples of serotype Typhimurium that demonstrated identical PFGE patterns isolated from five pairs of patients in which no epidemiologic links were detected were analyzed. Among these cases, members of two pairs were distinguished from one another by producing a different-sized fragment at one of the six loci (2730867 and 3629542), and members of three pairs could not be distinguished (Table (Table55).

TABLE 5.
Number of RUs at six VNTR loci in 10 clinical isolates of S. enterica serovar Typhimurium from five pairs of patients with identical PFGE patterns

Sequencing of the different size PCR products at each of the 10 loci from each serotype confirmed the sizes and sequences of the individual VNTR loci. At two of the VNTR loci (3184543 and 2628542) the number of copies was an exact integer of the RU size. Both of these loci were polymorphic for the number of RUs. (Fig. (Fig.22).

FIG. 2.
Diagram showing the arrangement of RUs and partial RUs at 10 VNTR loci among clinical isolates of S. enterica subsp. enterica serovars Typhimurium and Newport. The loci are named according to their locations on the S. enterica serovar Typhimurium LT2 ...

At eight loci there were partial copies of RUs. Partial copies were maintained in all of the isolates and were located at ends of the loci. Partial copies located at the 3′ end of the VNTR locus were copies of the 5′ end of the RU, and partial copies at the 3′ end of the repeat unit are copies of the 5′ end of the RU. At four loci (2341937, 2531837, 4596419, and 4810628) the partial copies were located at one end of the repeat locus. None of these were polymorphic for the number of RUs, and the partial copies at the end of the loci were the same in both serotypes. At three loci (2730867, 3414090, and 3629542) partial copies were located at both ends. In spite of this, the RU copy number was polymorphic among isolates, and the partial copies at each locus were identical and maintained in each isolate.

DISCUSSION

VNTR analysis, also referred to as multiple-locus VNTR analysis, has been utilized to genotype several species of bacteria including various serotypes of S. enterica subspecies enterica (8). Using eight VNTR loci, 78 isolates of serovar Typhimurium demonstrated 54 VNTR profiles (9). Compared to PFGE that was performed on the same samples, VNTR showed an improved ability to discriminate among different isolates. Among isolates of a particularly refractory phage type DT 104 strain, in which PFGE is not useful in distinguishing epidemiologically related strains from nonrelated ones, 28 profiles were discovered among 37 isolates that shared the same or a similar (differing by one restriction fragment) PFGE pattern (9). Four of the six loci found to be useful in the present study for genotyping serotypes Typhimurium and Newport correspond to those previously reported (9): 2341937 (STTR2), 3629542 (STTR3), 3184543 (STTR5), and 2730867 (STTR6).

VNTR analysis of S. enterica subspecies enterica serovar Typhi was based on the discovery of VNTR loci in the genome of serovar Typhi strain CT18 (13). Utilizing PCR to amplify five VNTR loci, the assay provided 49 distinct VNTR subtypes for 59 serovar Typhi isolates (9). Length polymorphisms were found among the strains for three of the five loci. Two loci showed no length polymorphism. When the same PCR primers were used to genotype isolates of S. enterica subsp. enterica serovars Typhimurium, Enterica, and Paratyphi A, B, and C, one locus was not present, and two did not show any length polymorphism.

In an effort to devise a method that could be used on a wide variety of serotypes, the genomes of S. enterica subsp. enterica strains CT18 (serovar Typhi) and LT2 (serovar Typhimurium) were searched for repeats that were common to both genomes (14). Ten VNTRs that were shared by the two genomes were used to design PCR primers that were applied to 99 human Salmonella isolates including nine commonly encountered serotypes. Although seven VNTR loci were useful in discriminating 27 serotype Typhi isolates into 25 genotype subgroups, only one locus was useful in discriminating the 39 serotype Typhimurium isolates into 8 genotype subgroups. The informative locus was one of the seven that discriminated the serotype Typhi isolates and was one of the loci, 3184543, used in the present study as well as in a previous study on serotype Typhimurium (9), where it was referred to as STTR5.

Sequencing the products of the PCRs confirmed that most of the VNTR loci included partial copies of the RUs. As with those present in the genome of strain LT2, partial copies were found on the ends of the VNTR loci (5′ end, 3′ end, or both ends). The partial copies are maintained in the sequence, even though the number of RUs in the VNTR varies among isolates.

In Arkansas, serotypes Typhimurium and Newport account, respectively, for 26% and 28% of the isolates of S. enterica subsp. enterica that were submitted to the Arkansas Department of Health during the study period. The present investigation was undertaken to determine whether PCR primers designed using the S. enterica subsp. enterica serotype Typhimurium LT2 genome might be useful in discriminating isolates belonging to both of these highly prevalent serotypes. It is able to discriminate isolates that have different PFGE patterns and, in some cases, isolates that share a common PFGE pattern. It can be performed using equipment commonly found in public health laboratories as part of a surveillance program utilizing serotype as a screen and then applying VNTR analysis to the isolates that share the frequently encountered serotypes Typhimurium and Newport. The data are numeric so that isolates can readily be compared. Using standard equipment, 10 isolates can be compared at six VNTR loci in less than 1 day for the cost of reagents for PCR and agarose gel analysis. The application of multiplex PCR using primers labeled with different fluors and capillary electrophoresis on a gene analyzer enable a high-throughput version of the analysis (10).

Acknowledgments

This work was supported by the U.S. Department of Agriculture under the authority of the Food Safety Consortium at the University of Arkansas, Agricultural Experimental Station, Fayetteville, Arkansas (grant 04-34211-7563).

We thank L. Hiley, Queensland Health and Pathology, Queensland, Australia, for critical reading of the manuscript.

Footnotes

[down-pointing small open triangle]Published ahead of print on 30 August 2006.

REFERENCES

1. Centers for Disease Control and Prevention. 2003. Salmonella surveillance summary, 2002. [Online.] http://www.cdc.gov/.
2. Frothingham, R., and W. A. Meeker-O'Connell. 1998. Genetic diversity in the Mycobacterium tuberculosis complex based on variable numbers of tandem DNA repeats. Microbiology 144:1189-1196. [PubMed]
3. Gerner-Smidt, P., K. Hise, J. Kincaid, S. Hunter, S. Rolando, E. Hyytia-Trees, E. M. Ribot, and B. Swaminathan. 2006. PulseNet USA: a five-year update. Foodborne Pathog. Dis. 3:9-19. [PubMed]
4. Jeffreys, A. J., J. F. Brookfield, and R. Semeonoff. 1985. Positive identification of an immigration test case using human DNA fingerprints. Nature 317:818-819. [PubMed]
5. Johansson, A., I. Goransson, P. Larsson, and A. Sjostedt. 2001. Extensive allelic variation among Francisella tularensis strains in a short-sequence tandem repeat region. J. Clin. Microbiol. 39:3140-3146. [PMC free article] [PubMed]
6. Keim, P., L. B. Price, A. M. Klevytska, K. L. Smith, J. M. Schupp, R. Okinaka, P. J. Jackson, and M. E. Hugh-Jones. 2000. Multiple-locus variable-number tandem repeat analysis reveals genetic relationships within Bacillus anthracis. J. Bacteriol. 182:2928-2936. [PMC free article] [PubMed]
7. Klevytska, A. M., L. B. Price, J. M. Schupp, P. L. Worsham, J. Wong, and P. Keim. 2001. Identification and characterization of variable-number tandem repeats in the Yersinia pestis genome. J. Clin. Microbiol. 39:3179-3185. [PMC free article] [PubMed]
8. Lindstedt, B. A. 2005. Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis 26:2567-2582. [PubMed]
9. Lindstedt, B. A., E. Heir, E. Gjernes, and G. Kapperud. 2003. DNA fingerprinting of Salmonella enterica subsp. enterica serovar Typhimurium with emphasis on phage type DT104 based on variable number of tandem repeat loci. J. Clin. Microbiol. 41:1469-1479. [PMC free article] [PubMed]
10. Lindstedt, B. A., T. Vardund, L. Aas, and G. Kapperud. 2004. Multiple-locus variable-number tandem-repeat analysis of Salmonella enterica subsp. enterica serovar Typhimurium using PCR multiplexing and multicolor capillary electrophoresis. J. Microbiol. Methods 59:163-172. [PubMed]
11. Maslow, J. N., M. E. Mulligan, and R. D. Arbeit. 1993. Molecular epidemiology: application of contemporary techniques to the typing of microorganisms. Clin. Infect. Dis. 17:153-162. [PubMed]
12. McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L. Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S. Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan, H. Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston, and R. K. Wilson. 2001. Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413:852-856. [PubMed]
13. Parkhill, J., G. Dougan, K. D. James, N. R. Thomson, D. Pickard, J. Wain, C. Churcher, K. L. Mungall, S. D. Bentley, M. T. Holden, M. Sebaihia, S. Baker, D. Basham, K. Brooks, T. Chillingworth, P. Connerton, A. Cronin, P. Davis, R. M. Davies, L. Dowd, N. White, J. Farrar, T. Feltwell, N. Hamlin, A. Haque, T. T. Hien, S. Holroyd, K. Jagels, A. Krogh, T. S. Larsen, S. Leather, S. Moule, P. O'Gaora, C. Parry, M. Quail, K. Rutherford, M. Simmonds, J. Skelton, K. Stevens, S. Whitehead, and B. G. Barrell. 2001. Complete genome sequence of a multiple drug resistant Salmonella enterica serovar Typhi CT18. Nature 413:848-852. [PubMed]
14. Ramisse, V., P. Houssu, E. Hernandez, F. Denoeud, V. Hilaire, O. Lisanti, F. Ramisse, J. D. Cavallo, and G. Vergnaud. 2004. Variable number of tandem repeats in Salmonella enterica subsp. enterica for typing purposes. J. Clin. Microbiol. 42:5722-5730. [PMC free article] [PubMed]
15. Swaminathan, B., T. J. Barrett, S. B. Hunter, and R. V. Tauxe. 2001. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg. Infect. Dis. 7:382-389. [PMC free article] [PubMed]
16. van Belkum, A., W. J. Melchers, C. Ijsseldijk, L. Nohlmans, H. Verbrugh, and J. F. Meis. 1997. Outbreak of amoxicillin-resistant Haemophilus influenzae type b: variable number of tandem repeats as novel molecular markers. J. Clin. Microbiol. 35:1517-1520. [PMC free article] [PubMed]

Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...