- Open Access
Evaluation of Gram-negative bacterial infection by a stable and conjugative bioluminescence plasmid in a mouse model
© Huang et al.; licensee BioMed Central 2014
- Received: 17 June 2014
- Accepted: 11 August 2014
- Published: 19 August 2014
The green fluorescence protein (GFP)-associated fluorescence method and the luciferase-associated bioluminescence method are the two major methods for IVIS imaging system to investigate the bacterial infection in animal models. The aim of this study was to evaluate the infection route of Gram-negative bacteria carrying a stable and broad range of conjugative bioluminescence plasmid pSE-Lux1 in a mouse model.
Both encapsulated and non-encapsulated Gram-negative bacteria were used as hosts to evaluate conjugation efficiency and plasmid stability of pSE-Lux1, a recombinant of pSE34 and luxABCDE operon. The plasmid conjugation efficiencies of pSE-Lux1 ranged from 10−3 to 10−7 in various Gram-negative bacteria. Plasmid pSE-Lux1 maintained in Escherichia coli, Klebsiella pneumoniae, and Salmonella enterica serovars Choleraesues (abbreviated S. Choleraesuis) and Typhimurium (S. Typhimurium), than in Acinetobacter baumannii and Serratia marcescens, was shown to be of better stability for at least four days. To investigate systemic bacterial infections, K. pneumoniae strain CG354 was intravenously injected, and then was clearly observed to be non-pathogenic to Balb/c mice for a long-term bioluminescence monitoring for 6 days. For examining dynamic distributions of gastrointestinal tract infection, the invasion protein SipB-deficient mutant OU5045△sipB and OU5046△sipB of S. serovar Typhimurium constructed in this study, compared to wild-type strain OU5045 and its virulence plasmid-less strain OU5046, were of less virulence to mice.
This is the first study to evaluate the conjugative and stable bioluminescence vehicle system of pSE-Lux1 in a wide range of Gram-negative bacteria, a system that can provide a useful reporter approach to trace systemic and gastrointestinal bacterial infections in a mouse model.
- Conjugative plasmi
- Plasmid stability
- Gram-negative bacteria
Traditional in vivo animal models to investigate microbiologic infection require sacrifice for tissue smear and culture. Although numerous methods have been developed to image bacteria, the clinical applications are still limited due to inefficient in vivo detections for a long term of observation without additions of selective antibiotics and substrates, such as lucifirins . Recently, bioluminescence expression system has been developed to trace microorganisms in a living animal . Bioluminescence system in live bacteria is regulated by a luxABCDE operon, in which luxAB genes encodes for luciferases and luxCDE genes encode lipid acid reductases to catalyze the reduction of long chain lipid acids into aldehyde compounds that are the substrates specific for the catalysation of luciferases LuxA and LuxB .
Recently, many luxABCDE-carrying vectors have been constructed for bioluminescence assays; however, a steady, broad-host auto-bioluminescence vehicle is needed, especially a conjugatable plasmid is preferred to deliver exogenous DNA between bacteria and overcome the difficulty in transformation for some thick-capsule bacteria, such as Klebsiella pneumoniae,. Earlier, we constructed a conjugative vehicle pSE-Lux1, which is a chimeric vector in combination of a stable conjugative native pSE34 that include pilX1, pilX2, pilX4, pilX5, pilX6, pilX7, pilX8, pilX9, pilX10, pilX11, taxA, taxB, and tax of type IV secretion system, with a bioluminescence reporter p3ZLux4 that contains the luxABCDE-kan operon of pXen-5 (Bioware, Caliper Life Sciences, USA) and the portion of pGEM®-3Z (Promega Corporation, USA) with 300–400 copies per bacterial cell –. Although pSE34 carries pir, parGF and stbDE for plasmid stability and partition, plasmid stability of pSE34 has not been evaluated .
In this study, the bacterial conjugation efficiency and plasmid stability of pSE-Lux1 were determined in various Gram-negative bacteria, and distributions of the pSE-Lux1-carrying bacteria in Balb/c mice were evaluated by an in vivo imaging system.
Gram-negative bacterial species and plasmids used in this study # Ap r : ampicillin resistance; Cm r : chloramphenicol resistance; Gm r : gentamicin resistance; Km r : kanamycin resistance; Str r : streptomycin resistance
Strains and plasmids
E. coli TOP10
ΔlacX74 ara Δ139 Δ(ara-leu)
Wild type with a virulence plasmid
Ou and Chu 
A virulence plasmidless strains derived from OU5045
Ou and Chu 
OU5045 with sipB deletion; Cmr
OU5046 with sipB deletion ; Cmr
Salmonella Choleraesuis OU7085
Clinical isolate; drug-susceptible
Ou and Chu 
Encapsulated clinical isolate; Cmr
Encapsulated clinical isolate; Cmr
Encapsulated clinical isolate; Cmr
Encapsulated clinical isolate; Cmr
Klebsiella pneumoniae CG354
Encapsulated clinical isolate; Cmr
ColE1-typed replicon and lux operon; Apr; Kmr;
A temperature-sensitive plasmid of λ
Red mutagenesis vector; Gmr
Doublet et al. 
46.3-kb conjugative plasmid with IncX- and ColE1-typed replicon and lux operon; Apr, Kmr;
Chen et al.
Mutagenesis in S. Typhimurium
Invasion effector gene sipB gene of Salmonella pathogenicity island 1 (SPI-1) was chosen to construct less virulence mutants of S. Typhimurium via the phage Lambda Red recombinase expression system ,. The primer sequences were designed according to the sequences of sipB of S. Typhimurium LT2 (accession number NC_003197) and cat (a chloramphenicol acetyl transferase-encoding gene) of pKD3 . The forward sequence SipB-Cm-F (5’-TGGAGTCTCG TCTGGCGGTA TGGCAGGCGA TGATTGAGTC cgcctacctg tgacggaaga-3’) and the reverse sequence SipB-Cm-R (5’-GCTGCGGTAT TCGTGACTTC CATGCCCAAC GCCACTTTAT ccctgccact catcgcagta-3’) were generated a 932-bp PCR amplicon, where the sipB gene portion is presented in capital letter and the cat portion in lower case. S. Typhimurium strains OU5045 and OU5046 had been previously transformed with a Lambda Red recombinases-carrying pKD46Gm by electroporation ,. The S. Typhimurium sipB-deleted mutants were named OU5045△sipB and OU5046△sipB.
Plasmid conjugation and stability
The conjugation test was conducted with a slight modification of methods described elsewhere . E. coli TOP10 and S. Typhimurium OU5046 carrying pSE-Lux1 played as donors, the drug-susceptible recipient bacteria, including A. baumannii strains AB08 and AB23, E. coli TOP10, K. pneumoniae strain CG354, S. Typhimurium OU5045 and OU5046, and S. Choleraesuis OU7085, and Serratia marcescens strains CB40 and CB47, were electroporated with gentamicin-resistant plasmid the pKD46Gm . Overnight donor and recipient bacteria were 10-fold diluted with LB broth and then co-cultured at 37°C without agitation for 16 hrs. Transconjugants and recipient bacteria were counted by plating bacteria on LB agar containing appropriate antibiotics. The conjugation efficiency of each recipient bacterium was determined by dividing the number of transconjugants by the total of recipients.
Plasmid stability was determined by methods as described elsewhere . Bacteria were cultured in LB broth medium without antibiotic and subcultured twice a day for a period of four days (D0-D4) in a 1000-fold dilution (approximate 80 generations in four-day period). The number of bacteria (CFU) was counted on agar media with or without kanamycin and ampicillin for both pSE-Lux1 and p3ZLux4 (Table 1) to select the bacteria with or without the plasmid. The plasmid stability was determined by the ratio of the bacterial survival number on selective media to the total bacterial number on non-selective media ,.
Evaluation of bacterial infection in mice
Animal test of bacterial infection in Balb/c mice was approved by the Institutional Animal Care and Use Committee (CGU11-164) of Chang Gung University, Taoyuan, Taiwan. Overnight-grown bacteria were washed and re-suspended with phosphate buffered saline (PBS) and then was used to challenge the mice via intravenous (IV) injection and oral administration ,. The mice were anesthetized by gas anesthesia (3% isoflurane), and then imaged by the in vivo imaging system IVIS® 100 Series (Xenogen) in accordance with the guidelines of Guide for the Care and Use of Laboratory Animals . After animal experiments or once illness appearances, mice were euthanized according to the previous guidelines.
For a long-term bioluminescence test of a systemic bacterial infection using the pSE-Lux1-carrying bacteria in mice, the mice (n = 3) were intravenously injected with 107 CFU of K. pneumoniae CG354 (a strain non-pathogenic to mice). The mice were tested for 6 days without selective pressures, and then were euthanized by breaking the neck while still anesthetized.
For the evaluation of gastrointestinal infection of Salmonella in mouse, the mice were orally gavaged with 10% sodium bicarbonate to neutralize stomach acid for 30 minutes prior to Salmonella challenge. The mice were separated into four groups (n = 5 per group) and each mouse were then orally gavaged with 109 CFU S. Typhimurium. The strains were wild type OU5045, virulence plasmid-less OU5046, and sipB-deletion mutant strains OU5045△sipB and OU5046△sipB. The mice were tested without selective pressures for four hours to strains OU5045 and OU5046, and for two days to mutants due to the restriction of mouse illness appearance thereafter. The anatomic organs of the tested mice were analyzed in parallel for IVIS imaging.
Pair-wise comparison in one-way ANOVA (ANalyses Of VAriance between groups) test was performed using the software program of Statistical Product And Service Solutions (SPPS 12.0), and followed by a Tukey’s HSD (Honestly Significant Difference) test to determine the significance of difference between p3ZLux4 and pSE-Lux1 in the tests of plasmid stabilities.
Plasmid pSE-Lux1-mediated conjugation tests between Gram-negative bacteria
Conjugation efficiency of pSE-Kux1 among various Gram-negative species
A. baumannii AB08
1.2 ± 0.6 × 10−7
A. baumannii AB23
8.1 ± 3.8 × 10−6
E. coli TOP10
1.8 ± 1.6 × 10−4
K. pneumoniae CG354
1.9 ± 1.3 × 10−6
S. Choleraesuis SC7085
7.5 ± 2.4 × 10−3
S. Typhimurium OU5045
1.0 ± 0.8 × 10−6
Serratia marcescens CB40
5.9 ± 5.1 × 10−7
Serratia marcescens CB47
1.2 ± 0.4 × 10−3
S. Typhimurium OU5046/pSE-Lux1
E. coli TOP10
2.2 ± 1.1 × 10−6
K. pneumoniae CG354
8.4 ± 4.6 × 10−7
S. Typhimurium OU5045
1.4 ± 1.6 × 10−3
Difference in plasmid stability of pSE-Lux1 and p3ZLux4 within various bacteria
Long-term monitoring of bioluminescence bacterium in mice with a systemic bacterial infection
Bacterial distributions of the route of gastrointestinal infection in mice
No significant difference in bioluminescence patterns was observed among the four isogenic strains within gastrointestinal tracts of mice. The bioluminescence signals were matched the amount of living bacteria from the anatomic organs. However, the periods of illness appearance differed between the sipB-normal and the sipB-deletion strains (four hours vs. two days post-Salmonella inoculations). Noteworthily, the bioluminescence signals present in the portion of stomach organs for the virulence plasmid-less OU5046, but not in its wild-type OU5045.
The green fluorescence protein (GFP)-associated fluorescence method and the luciferase-associated bioluminescence method are the two major methods for IVIS imaging system in animal models. Compared to GFP fluorescence, two advantages of the luxABCDE- mediated bioluminescence method are A) only metabolically active and living bacteria may present light, but the dormant or dead bacteria may not, or weakly, because of their little production of aldehyde substrates for luciferase reactions; B) bioluminescence background of this system is low in animal models . However, the GFP fluorescence method has higher sensitivity than bioluminescence methods in IVIS system and this advantage is only available to superficial organs less than 6 mm depth from the surface of test animals, or fluorescence signals would be faded . Moreover, auto-fluorescence backgrounds emitted from animals are high . To overcome auto-fluorescence, the test animals should get starved for 3 to 24 hours prior to fluorescence imaging, because diets may cause significant auto-fluorescence . The precaution by starvation may limit the application of GFP-associated methods in IVIS imaging systems, particularly at the detection sites close to gastrointestinal tracts.
First bioluminescence image of the lux operon was developed to study the pathogenesis of S. Typhimurium in C57BL/6 or BALB/c mice using an artificial plasmid that contains this operon from Photorhabdus luminescens. However, the replication origin of this plasmid was derived from the ColE1 replicon and was unstable without antibiotic pressure. Therefore, three stable photonic plasmids pCGLS-1 (carrying ColE1 replicon), pAK1-lux (carrying pBBR1 replicon) and pXEN-1 (carrying both pC194 and ColE1 replicons) were constructed for the stability in S. Typhimurium . In this study, a pSE34-based pSE-Lux1 showed better plasmid stability than ColE1 replicon-based p3ZLux4 in E. coli, K. pneumoniae, and S. enterica. This is probably because pSE-Lux1 carries important genetic elements, including ColE1 and IncX replicons, conjugation-associated pil operon, and plasmid maintenance-associated genes pir, parG, parF, stbD, and stbE. Some other bioluminescence systems (such as fluc, gluc, or rluc) may be more appropriate than luxABCDE in various bacteria ,,. However, firefly Fluc and Gaussia Gluc require the intravenous addition of substrate luciferins for bioluminescence catalyzation in vivo, therefore, it only remains relatively as short as 30 minutes when the peak of bioluminescence signal reaches a plateau .
The conjugation efficiencies differed between bacterial species probably due to difference in enzymatic restriction and modification system. However, the two recipient S. marcescens strain CB40 and CB47 showed dramatically different conjugation efficiencies and this difference may be attributed to the different genomic background or bacterial capsule. With regard to the bioluminescence patterns of K. pneumoniae mucoid strain CG354 for a systemic bacterial infection in this study, the gradual reduction of bioluminescence signals over the time indicated that strain CG354 was non-pathogenic to mice. Similar to strain CG354, strain IA565 of K. pneumoniae, a human clinical isolate, is known to be non-pathogenic to mice . However, survey of International Klebsiella Study Group reported that 69% mucoid clinical strains are pathogenic to murine .
In the bioluminescence patterns of anatomic gastrointestinal organs of mice, pSE-Lux1-carrying OU5046 strain produced more signals in the stomach than by the pSE-Lux1-carrying OU5045 strain were found. It is likely because the more virulent strain caused more severe inflammatory diarrhea to mice, more Salmonella shed away from stomach and other gastrointestinal organs, and therefore, the less virulent △sipB mutant remained more in gastrointestinal tracts. Although Salmonella are known to enable survive in the acidic environment of stomach through the induction of the acid tolerance response, it still remains unclear why Salmonella can colonize in the stomach; however, its colonization may explain why Salmonella can cause stomach cramps in humans ,.
Bioluminescence-related publications have been increasing in application to study in the area of pathogenicity, tumorigenicity, biofilm, and dermatology –. Moreover, the bioluminescence vehicle can be genetically engineered to carry some other potential exogenous genes, such as anticancer agents for therapeutic purposes .
In this study, a novel stable and conjugative bioluminescence pSE-Lux1 vehicle system available in a broad range of bacteria, even for encapsulated bacteria, is well developed and applied to investigate the infection route of pSE-Lux1-carrying bacteria in living mice
The authors thank Pei-Chun Tu and Hsin-Ju Chang, Chang Gung Memorial Hospital for their assistance in bacterial conjugation test and the IVIS detection in mice. This study was supported by grants from Chang Gung Memorial Hospital, Taoyuan, Taiwan (CMRPG381051-2, CMRPG390701-2, CMRPG6C0341, CMRPG6B0501, CMRPG490141-3 and CMRPG3A1111-3).
- Moulton K, Ryan P, Lay D, Willard S: Photonic plasmid stability of transformedSalmonella typhimurium: a comparison of three unique plasmids.BMC Microbiol 2009, 9:152.,Google Scholar
- Andreu N, Zelmer A, Wiles S: Noninvasive biophotonic imaging for studies of infectious disease. FEMS Microbiol Rev. 2010, 35: 360-394. 10.1111/j.1574-6976.2010.00252.x.PubMed CentralView ArticlePubMedGoogle Scholar
- Hutchens M, Luker GD: Applications of bioluminescence imaging to the study of infectious diseases. Cell Microbiol. 2007, 9: 2315-2322. 10.1111/j.1462-5822.2007.00995.x.View ArticlePubMedGoogle Scholar
- Llosa M, de la Cruz F: Bacterial conjugation: a potential tool for genomic engineering. Res Microbiol. 2005, 156: 1-6. 10.1016/j.resmic.2004.07.008.View ArticlePubMedGoogle Scholar
- Chiu CH, Chen CL, Huang YK, inventor: Chang Gung Memorial Hospital, assignee: Method for tracing Gram-negative bacteria inside animal model using stable and bioluminescence-based expression system therefor. In United States Patent. 2010:US8268616B2.Google Scholar
- Chiu CH, Chen CL, Huang YK, inventor: Chang Gung Memorial Hospital, assignee: Method for tracing Gram-negative bacteria inside animal model using stable and bioluminescence-based expression system therefor. In 2011:US8263366B2.Google Scholar
- Chen CL, Wang CY, Chu C, Su LH, Chiu CH: Functional and molecular characterization of pSE34 encoding a type IV secretion system in Salmonella enterica serotype Enteritidis phage type 34. FEMS Immunol Med Microbiol. 2009, 57: 274-283. 10.1111/j.1574-695X.2009.00612.x.View ArticlePubMedGoogle Scholar
- Vuong C, Kocianova S, Yu J, Kadurugamuwa JL, Otto M: Development of real-time in vivo imaging of device-related Staphylococcus epidermidis infection in mice and influence of animal immune status on susceptibility to infection. J Infect Dis. 2008, 198: 258-261. 10.1086/589307.PubMed CentralView ArticlePubMedGoogle Scholar
- Kado CI, Liu ST: Rapid procedure for detection and isolation of large and small plasmids. J Bacteriol. 1981, 145: 1365-1373.PubMed CentralPubMedGoogle Scholar
- Ou JT, Chu C, inventor: Crystal Biotechnology Research and Development Co., Ltd., assignee. In Chicken leucocytozoon vaccine. United States Patent. 1999:6207167.Google Scholar
- Doublet B, Douard G, Targant H, Meunier D, Madec JY, Cloeckaert A: Antibiotic marker modifications of lambda Red and FLP helper plasmids, pKD46 and pCP20, for inactivation of chromosomal genes using PCR products in multidrug-resistant strains. J Microbiol Methods. 2008, 75: 359-361. 10.1016/j.mimet.2008.06.010.View ArticlePubMedGoogle Scholar
- Myeni SK, Wang L, Zhou D: SipB-SipC complex is essential for translocon formation.PLoS One 2013, 8:e60499.,Google Scholar
- Uzzau S, Figueroa-Bossi N, Rubino S, Bossi L: Epitope tagging of chromosomal genes in Salmonella. Proc Natl Acad Sci U S A. 2001, 98: 15264-15269. 10.1073/pnas.261348198.PubMed CentralView ArticlePubMedGoogle Scholar
- Trevors JT, van Elsas JD, Starodub ME, van Overbeek LS: Survival of and plasmid stability in Pseudomonas and Klebsiella spp. introduced into agricultural drainage water. Can J Microbiol. 1989, 35: 675-680. 10.1139/m89-110.View ArticlePubMedGoogle Scholar
- Olivier V, Queen J, Satchell KJF: Successful small intestine colonization of adult mice byVibrio choleraerequires ketamine anesthesia and accessory toxins.PLoS One 2009, 4:e7352.,Google Scholar
- Yu VL, Hansen DS, Ko WC, Sagnimeni A, Klugman KP, von Gottberg A: Virulence characteristics of Klebsiella and clinical manifestations of K. pneumoniae bloodstream infections. Emerg Infect Dis. 2007, 13: 986-993. 10.3201/eid1307.070187.PubMed CentralView ArticlePubMedGoogle Scholar
- Kastenmayer RJ, Moore RM, Bright AL, Torres-Cruz R, Elkins WR: Select agent and toxin regulations: beyond the eighth edition of the Guide for the Care and Use of Laboratory Animals. J Am Assoc Lab Anim Sci. 2012, 51: 333-338.PubMed CentralPubMedGoogle Scholar
- Gahan CGM: The bacterial lux reporter system: applications in bacterial localisation studies. Curr Gene Ther. 2012, 12: 12-19. 10.2174/156652312799789244.View ArticlePubMedGoogle Scholar
- Zacharakis G, Kambara H, Shih H, Ripoll J, Grimm J, Saeki Y: Volumetric tomography of fluorescent proteins through small animals in vivo. Proc Natl Acad Sci U S A. 2005, 102: 18252-18257. 10.1073/pnas.0504628102.PubMed CentralView ArticlePubMedGoogle Scholar
- Inoue Y, Izawa K, Kiryu S, Tojo A, Ohtomo K: Diet and abdominal autofluorescence detected by in vivo fluorescence imaging of living mice. Mol Imaging. 2008, 7: 21-27.PubMedGoogle Scholar
- Contag CH, Contag PR, Mullins JI, Spilman SD, Stevenson DK, Benaron DA: Photonic detection of bacterial pathogens in living hosts. Mol Microbiol. 1995, 18: 593-603. 10.1111/j.1365-2958.1995.mmi_18040593.x.View ArticlePubMedGoogle Scholar
- Tannous BA, Kim DE, Fernandez JL, Weissleder R, Breakefield XO: Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol Ther. 2005, 11: 435-443. 10.1016/j.ymthe.2004.10.016.View ArticlePubMedGoogle Scholar
- Chang MH, Cirillo SLG, Cirillo JD: Using luciferase to image bacterial infections in mice.J Vis Exp 2011, 18:2547.,Google Scholar
- Lau HY, Clegg S, Moore TA: Identification of Klebsiella pneumoniae genes uniquely expressed in a strain virulent using a murine model of bacterial pneumonia. Microb Pathog. 2007, 42: 148-155. 10.1016/j.micpath.2007.01.001.PubMed CentralView ArticlePubMedGoogle Scholar
- Alvarez-Ordóñez A, Begley M, Prieto M, Messens W, López M, Bernardo A: Salmonella spp. survival strategies within the host gastrointestinal tract. Microbiology. 2011, 157 (Pt 12): 3268-3281. 10.1099/mic.0.050351-0.View ArticlePubMedGoogle Scholar
- Gonose T, Smith AM, Keddy KH, Sooka A, Howell V, Jacobs CA: Human infections due toSalmonellaBlockley, a rare serotype in South Africa: a case report.BMC Res Notes 2012, 5:562.,Google Scholar
- Garcez AS, Núñez SC, Azambuja N, Fregnani ER, Rodriguez HMH, Hamblin MR: Effects of Photodynamic Therapy on Gram-Positive and Gram-Negative Bacterial Biofilms by Bioluminescence Imaging and Scanning Electron Microscopic Analysis. Photomed Laser Surg. 2013, 31: 519-525. 10.1089/pho.2012.3341.PubMed CentralView ArticlePubMedGoogle Scholar
- Gonzalez RJ, Weening EH, Frothingham R, Sempowski GD, Miller VL: Bioluminescence imaging to track bacterial dissemination ofYersinia pestisusing different routes of infection in mice.BMC Microbiol 2012, 12:147.,Google Scholar
- Iochmann S, Lerondel S, Bléchet C, Lavergne M, Pesnel S, Sobilo J: Monitoring of tumour progression using bioluminescence imaging and computed tomography scanning in a nude mouse orthotopic model of human small cell lung cancer. Lung Cancer. 2012, 77: 70-76. 10.1016/j.lungcan.2012.01.009.View ArticlePubMedGoogle Scholar
- Vecchio D, Dai T, Huang L, Fantetti L, Roncucci G, Hamblin MR: Antimicrobial photodynamic therapy with RLP068 kills methicillin-resistant Staphylococcus aureus and improves wound healing in a mouse model of infected skin abrasion PDT with RLP068/Cl in infected mouse skin abrasion. J Biophotonics. 2013, 6: 733-742. 10.1002/jbio.201200121.PubMed CentralView ArticlePubMedGoogle Scholar
- Forbes NS: Engineering the perfect (bacterial) cancer therapy. Nat Rev Cancer. 2010, 10: 785-794. 10.1038/nrc2934.PubMed CentralView ArticlePubMedGoogle 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/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.