Distinct polymicrobial populations in a chronic foot ulcer with implications for diagnostics and anti-infective therapy
© Imirzalioglu et al.; licensee BioMed Central Ltd. 2014
Received: 22 August 2013
Accepted: 27 March 2014
Published: 29 March 2014
Polymicrobial infections caused by combinations of different bacteria are being detected with an increasing frequency. The evidence of such complex infections is being revealed through the use of novel molecular and culture-independent methods. Considerable progress has been made in the last decade regarding the diagnostic application of such molecular techniques. In particular, 16S rDNA-based sequencing and even metagenomic analyses have been successfully used to study the microbial diversity in ecosystems and human microbiota. Here, we utilized denaturing high-performance liquid chromatography (DHPLC) as a diagnostic tool for identifying different bacterial species in complex clinical samples of a patient with a chronic foot ulcer.
A 45-year-old female suffered from a chronic 5x5cm large plantar ulcer located in the posterior calcaneal area with subcutaneous tissue infection and osteomyelitis. The chronic ulcer developed over a period of 8 years. Culture and DHPLC revealed a distinct and location-dependent polymicrobial infection of the ulcer. The analysis of a superficial biopsy revealed a mixture of Staphylococcus aureus, Proteus vulgaris, and Fusobacterium nucleatum, whereas the tissue-deep biopsy harbored a mixture of four different bacterial species, namely Gemella morbillorum, Porphyromonas asaccharolytica, Bacteroides fragilis, and Arcanobacterium haemolyticum.
This clinical case highlights the difficulties in assessing polymicrobial infections where a mixture of fastidious, rapid and slow growing bacteria as well as anaerobes exists as structured communities within the tissue architecture of chronic wound infections. The diagnosis of this multilayered polymicrobial infection led to a microbe-adapted antibiotic therapy, targeting the polymicrobial nature of this infection in addition to a standard local wound treatment. However, a complete wound closure could not be achieved due to the long-lasting extensive destruction of tissue.
KeywordsPolymicrobial infection Foot ulcer Microbiome Gemella morbillorum
Polymicrobial diseases caused by combinations of different bacteria, as well as viruses, fungi, and parasites are being detected with an increasing frequency [1, 2]. There are different ecological features underlying the induction of polymicrobial diseases, such as microbial interference, where (i) one microorganism generates a niche in the host that suppresses the colonization of other microorganisms or synergistic polymicrobial infections, where (ii) one microorganism generates a niche favorable for the infection and colonization of other microorganisms, or where (iii) one microorganism predisposes the host to colonization by other microorganisms as evidenced between some periodontal pathogens [2, 3]. Another example of a polymicrobial infection is an additive infection implying that two or more microorganisms can synergistically cause infection as seen in different clinical entities, such as bacteraemia, abdominal abscess, secondary peritonitis , soft-tissue infection or fasciitis . Furthermore, many bacteria are able to form biofilms which provide a perfect niche, a bastion, to protect themselves from the host immune system and anti-infective treatment .
The evidence of such complex infections is being revealed through the use of novel molecular and culture-independent methods. Indeed, in the clinical case presented in this study, a mixture of fastidious, rapid and slow growing bacteria as well as anaerobes existed as structured communities within the tissue architecture of a chronic wound infection with G. morbillorum being the predominant bacterial species. This mixture would probably not have been detected without the use of an additional molecular technology.
Laboratory tests performed upon consultation
Leukocytes [giga/l] Eb
Erythrocytes [tera/l] Eb
Hemoglobin [g/l] Eb
Hematocrit [l/l] Eb
Thrombocytes [giga/l] Eb
MCV [fl] Eb
MCH [pg] Eb
MCHC [g/dl] Eb
Neutrophils [giga/l] Eb
Eosinophils [giga/l] Eb
Basophils [0.03] Eb
Lymphocytes [giga/l] Eb
Monocytes [giga/l] Eb
Erythrocytes sedimentation rate [mm/H] Eb
HbA1c [%] Eb
Sodium [mMol/l] Hp
Potassium [mMol/l] Hp
Creatinine [mg/dl] Hp
Urea [mg/dl] Hp
Uric acid [mg/dl] Hp
Protein [g/l] Hp
LDH [U/l] Hp
GOT (AST) [U/l] Hp
GPT (ALT) [U/l] Hp
GGT [U/l] Hp
CRP [mg/l] Hp
IgE [IU/ml] Hp
Antistreptolysin [U/ml] Hp
For microbiological analysis we obtained both a superficial and a deep subcutaneous tissue biopsy for cultural and molecular examinations (Figure 1A). The samples were routinely processed for culture and for PCR. Briefly, the biopsies were homogenized by using the FastPrep Cell Disrupter (MP Biomedicals, Germany) and firstly streaked on MacConkey, blood, chocolate, Schaedler, and Sabouraud agar plates. The agar plates were incubated at 37°C for two days except for Schaedler and Sabouraud agar plates which were incubated for two weeks and four weeks, respectively. Additionally, an enrichment culture was done in thioglycollate. Secondly, the homogenized samples were subjected to nucleic acid extraction, PCR, and subsequent denaturing high-performance liquid chromatography (DHPLC) as described . DHPLC is a technique proven to be appropriate for the molecular detection of polymicrobial infections. It is capable of separating mixed PCR amplicons derived from a mixture of bacteria in a sample by using a cartridge with microspheres, a fragment collector, amplicon sequencing, and bioinformatics [7–11].
G. morbillorum was identified to be the predominant species in this structured chronic wound infection. It is a gram-positive coccus and a member of the commensal organisms located on the mucous membranes of humans and warm-blooded animals . However, a number of reports have suggested that Gemella can cause both severe localized and generalized infections. These infections include abscesses [13, 14], endocarditis [15–18], endovascular infections [19, 20], infections of the central nervous system , septic arthritis , and skeletal infections . Gemella spp. possesses a typical gram-positive cell wall structure. However, during Gram staining cells easily lose their color and may therefore appear gram-negative (gram-labile). This could probably be the reason for frequent misidentification and the low number of clinical case reports associated with Gemella infections. Rapid identification systems based on biochemical features or MALDI-TOF mass spectrometry require cultured bacteria and do not offer reliable identification of all strains of these species once cultivated [24–26].
All of the other bacteria were detected in significantly lower amounts (Figure 1): From the deep-tissue biopsy P. asaccharolytica, B. fragilis, and A. haemolyticum were isolated, which have all been reported to be capable of causing chronic cutaneous or even necrotizing abscesses [27, 28]. From the superficial biopsy, which represented the upper level of the wound the clinically well-known and frequently isolated bacteria S. aureus, P. vulgaris, and F. nucleatum were detected. All of these bacteria are known to cause severe wound infections and even necrotizing abscesses [27, 29, 30].
Interestingly, there were no similarities in bacterial species from either sample suggesting two distinct polymicrobial populations, one located superficially and one located in the deep subcutaneous tissue. Therefore, we postulated that a polymicrobial infection – although not necessarily the primary cause - with predominance of G. morbillorum aggravated the lesion, and thereby complicated the healing process. The diagnosis of this multilayered polymicrobial infection prompted a local wound treatment with AQUACELL® Ag (ConvaTec GmbH, Munich, Germany) and Mepilex® Lite (Mölnlycke Health Care GmbH, Vienna, Austria). AQUACELL® Ag is based on an advanced Hydrofiber® technology which locks in fluid and traps bacteria. Incorporated ionic silver exhibits antimicrobial activity and kills a wide spectrum of wound associated bacteria. Mepilex® Lite consists of a thin flexible sheet of absorbent hydrophilic polyurethane foam which is designed for the management of a wide range of exuding wounds such as leg and foot ulcers and which provides a moist wound-healing environment. Furthermore, the diagnosis enabled a microbe-adapted antibiotic therapy, targeting the polymicrobial nature of this infection with emphasis on the predominant pathogen G. morbillorum (based on previously reported Gemella resistance traits), with clindamycin (daily dose: 3×600 mg) and levofloxacin (daily dose: 1×500 mg) over a period of 6 weeks. Although the ulcer was profuse, deep, and already persisting for nearly a decade the induced treatments prevented further dissemination of the infection, eliminated the potential threat of amputation, and supported tissue regeneration. However, a complete wound closure could not be achieved during the course of this study due to the long-lasting extensive destruction of tissue.
Chronic skin and soft tissue ulcers present challenges regarding diagnosis and treatment. These wounds are frequently caused by a diverse mixture of bacteria encompassing fastidious, rapid and slow growing microbes as well as anaerobes. The traditional detection of such diverse microbes by culture requires dedicated media and culture conditions and is frequently unsatisfactory although the conditions used were appropriate. Therefore, many laboratories also use molecular methods such as PCR to detect pathogens involved in polymicrobial infections. The combination of traditional culture methods and molecular techniques enhances pathogen detection and therefore enables a pathogen-adapted anti-infective therapy. Interestingly, the bacterial community in the superficial biopsy could be cultivated and also detected via PCR. However, the microorganisms in the tissue-deep biopsy were nonculturable. Therefore, the polymicrobial nature of this tissue-deep biopsy was only detectable by culture-independent, molecular methods (PCR and DHPLC). It is known that anaerobes die quickly after sample collection and transport into the laboratory and that diagnosis by culture is therefore difficult . But in our case the transportation times to the laboratory were short and the samples were immediately processed. One would expect that G. morbillorum and A. haemolyticum were cultivable since the first one is a microaerophilc bacterium, although rather slow growing, and the second one is facultative anaerobe. We hypothesize that the reason for the failure of the culture lies in the potential presence of a polymicrobial biofilm in the deep wound where the microbes existed in a viable but nonculturable (VBNC) status [32, 33]. Furthermore, discrepancies between superficial and deep wound samples are often observed in diabetic foot infections, as a model for chronic wound situations, as stated in several publications [34–36]. However, in some of these cases the deep sample or even the bone biopsy remained negative despite clinical evidence of infection. Unfortunately, these samples were not evaluated by molecular methods, as a comparable situation as in this case could be the underlying cause of the culture negativity.
This clinical case highlights the difficulties in assessing polymicrobial infections where a mixture of fastidious, rapid and slow growing bacteria as well as anaerobes exists as structured communities within the tissue architecture of chronic wound infections. The evidence of such complex infections is being revealed through the use of novel molecular and culture-independent methods. Here, subjecting biopsies obtained from various regions of the wound to DNA-based detection methods provides additional information not available by conventional culture methods. The identification of G. morbillorum as a predominant bacterial species associated with the pathogenesis of the wound infection would have probably been missed without the use of an additional molecular technique. Considerable progress has been made in the last decade regarding the diagnostic application of molecular techniques. In particular the use of the 16S rDNA-based sequencing and even metagenomic analyses have been successfully used to study the microbial diversity in ecosystems and human microbiota [37–40]. However, these approaches require sophisticated instrumentation and technologies employing next-generation sequencing machines with subsequent bioinformatic analyses of metagenomic data , . At least two hurdles must be overcome for the implementation of these techniques in the clinical diagnostic laboratory. Firstly, these techniques are still time consuming, difficult to reproduce, not yet standardized, and labor-intensive, thereby hindering their routine use in a diagnostic laboratory. Secondly, data linking detection of polymicrobial infection and their utility in assessing clinical states is largely missing [42–44]. Here, we utilized DHPLC as an additional diagnostic tool (staged diagnostics) for identifying different bacterial species in a complex clinical sample. This technology not only allows resolution of mixed amplicons quantitatively, but also permits subsequent sequencing of the individually separated amplicons [7, 9]. Falling costs in high-throughput next-generation sequencing technologies and the implementation of enhanced bioinformatic tools, to enable “next-generation diagnostic capabilities” will considerably improve the detection potential of polymicrobial infections in the standard clinical microbiology laboratory.
Written informed consent was obtained from the patient for publication of this Case report and any accompanying images. A copy of written consent is available for review by the Editor of this journal. The study was approved by the Ethics Board of the Justus-Liebig-University of Giessen.
The authors thank the patient for her consent to publish the case and Silke Zechel-Gran and Kirsten-Susann Bommersheim for excellent technical assistance. This work was supported by the research grant “Metagenomics and host-pathogen interactomics in diabetic foot infections” to E.D. and T.C. (FKZ: 0315830C) from the Bundesministerium fuer Bildung und Forschung, Germany.
- Brogden KA, Guthmiller JM: Polymicrobial diseases. 2002, Washington, D.C: ASM PressView ArticleGoogle Scholar
- Brogden KA, Guthmiller JM, Taylor CE: Human polymicrobial infections. Lancet. 2005, 365: 253-255. 10.1016/S0140-6736(05)17745-9.PubMedView ArticleGoogle Scholar
- Palmer RJ, Kazmerzak K, Hansen MC, Kolenbrander PE: Mutualism versus independence: strategies of mixed-species oral biofilms in vitro using saliva as the sole nutrient source. Infect Immun. 2001, 69: 5794-5804. 10.1128/IAI.69.9.5794-5804.2001.PubMedPubMed CentralView ArticleGoogle Scholar
- Cheadle WG, Spain DA: The continuing challenge of intra-abdominal infection. Am J Surg. 2003, 186: 15S-22S. 10.1016/j.amjsurg.2003.10.003.PubMedView ArticleGoogle Scholar
- Fontes RA, Ogilvie CM, Miclau T: Necrotizing soft-tissue infections. J Am Acad Orthop Surg. 2000, 8: 151-158.PubMedGoogle Scholar
- Percival SL, Emanuel C, Cutting KF, Williams DW: Microbiology of the skin and the role of biofilms in infection. Int Wound J. 2012, 9: 14-32. 10.1111/j.1742-481X.2011.00836.x.PubMedView ArticleGoogle Scholar
- Domann E, Hong G, Imirzalioglu C, Turschner S, Kühle J, Watzel C, Hain T, Hossain H, Chakraborty T: Culture-independent identification of pathogenic bacteria and polymicrobial infections in the genitourinary tract of renal transplant recipients. J Clin Microbiol. 2003, 41: 5500-5510. 10.1128/JCM.41.12.5500-5510.2003.PubMedPubMed CentralView ArticleGoogle Scholar
- Caillaux G, de la Cochetiere MF, Carton T, Le Vacon F, Roze JC, Potel G, Caillon J, Gras-L Guen C: Application of denaturing high-performance liquid chromatography for intestinal microbiota analysis of newborns. J Perinat Med. 2010, 38: 339-341.PubMedGoogle Scholar
- Imirzalioglu C, Hain T, Chakraborty T, Domann E: Hidden pathogens uncovered: metagenomic analysis of urinary tract infections. Andrologia. 2008, 40: 66-71. 10.1111/j.1439-0272.2007.00830.x.PubMedView ArticleGoogle Scholar
- Skraban J, Dzeroski S, Zenko B, Mongus D, Gangl S, Rupnik M: Gut microbiota patterns associated with colonization of different Clostridium difficile ribotypes. PLoS One. 2013, 8: e58005-10.1371/journal.pone.0058005.PubMedPubMed CentralView ArticleGoogle Scholar
- Wagner AO, Malin C, Illmer P: Application of denaturing high-performance liquid chromatography in microbial ecology: fermentor sludge, compost, and soil community profiling. Appl Environ Microbiol. 2009, 75: 956-964. 10.1128/AEM.01411-08.PubMedPubMed CentralView ArticleGoogle Scholar
- Ruoff KL: Aerococcus, Abiotrophia, and other aerobic catalase-negative, gram-positive cocci. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA. 2007, Washington, D.C: ASM Press, 1: 443-454. 9Google Scholar
- Liberto MC, Matera G, Puccio R, Barbieri V, Quirino A, Capicotto R, Guadagnino V, Pardatscher K, Foca A: An unusual case of brain abscess by Gemella morbillorum. Jpn J Infect Dis. 2006, 59: 126-128.PubMedGoogle Scholar
- Villamil I, Villar A, Masa LA: Cutaneous abscess due Gemella morbillorum. Rev Chilena Infectol. 2009, 26: 464-465.PubMedView ArticleGoogle Scholar
- Al-Hujailan G, Lagace-Wiens P: Mechanical valve endocarditis caused by Gemella morbillorum. J Med Microbiol. 2007, 56 (Pt 12): 1689-1691.PubMedView ArticleGoogle Scholar
- FitzGerald SF, Moloney AC, Maurer BJ, Hall WW: Gemella endocarditis: consider the colon. J Heart Valve Dis. 2006, 16: 833-835.Google Scholar
- Murai M, Fukumoto H, Negoro N, Sano K, Nishimoto M, Muraoka H, Hazui H, Morita H, Hanafusa T: Evidence of active endocarditis, caused by Gemella morbillorum, related to acute embolic stroke. Int J Cardiol. 2006, 112: e17-e18. 10.1016/j.ijcard.2006.01.071.PubMedView ArticleGoogle Scholar
- Zakir RM, AlDehneh A, Dabu L, Kapila R, Saric M: Mitral bioprosthetic valve endocarditis caused by an unusual microorganism, Gemella morbillorum, in an intravenous drug user. J Clin Microbiol. 2004, 42: 4893-4896. 10.1128/JCM.42.10.4893-4896.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Brack MJ, Avery PG, Hubner PJ, Swann RA: Gemella haemolysans: a rare and unusual cause of infective endocarditis. Postgrad Med J. 1991, 67: 210-PubMedPubMed CentralView ArticleGoogle Scholar
- Buu-Hoi A, Sapoetra A, Branger C, Acar JF: Antimicrobial susceptibility of Gemella haemolysans isolated from patients with subacute endocarditis. Eur J Clin Microbiol. 1982, 1: 102-106. 10.1007/BF02014200.PubMedView ArticleGoogle Scholar
- Aspevall O, Hillebrant E, Linderoth B, Rylander M: Meningitis due to Gemella haemolysans after neurosurgical treatment of trigeminal neuralgia. Scand J Infect Dis. 1991, 23: 503-505. 10.3109/00365549109075101.PubMedView ArticleGoogle Scholar
- Roche M, Smyth E: A case of septic arthritis due to infection with Gemella morbillorum. J Infect. 2005, 51: e187-e189. 10.1016/j.jinf.2005.01.009.PubMedView ArticleGoogle Scholar
- Omran Y, Wood CA: Endovascular infection and septic arthritis caused by Gemella morbillorum. Diagn Microbiol Infect Dis. 1993, 16: 131-134. 10.1016/0732-8893(93)90007-T.PubMedView ArticleGoogle Scholar
- Appelbaum PC, Jacobs MR, Heald JI, Palko WM, Duffett A, Crist R, Naugle PA: Comparative evaluation of the API 20S system and the automicrobic system gram-positive identification card for species identification of streptococci. J Clin Microbiol. 1984, 19: 164-168.PubMedPubMed CentralGoogle Scholar
- Carbonelle E, Grohs P, Jacquier H, Day N, Tenza S, Dewailly A, Vissouarn O, Rottman M, Herrmann JL, Podglajen I, Raskine L: Robustness of two MALDI-TOF mass spectrometry systems for bacterial identification. J Microbiol Meth. 2012, 89: 133-136. 10.1016/j.mimet.2012.03.003.View ArticleGoogle Scholar
- Facklam RR, Rhoden DL, Smith PB: Evaluation of the Rapid Strep system for the identification of clinical isolates of Streptococcus species. J Clin Microbiol. 1984, 20: 894-898.PubMedPubMed CentralGoogle Scholar
- Citron DM, Poxton IR, Baron EJ: Bacteroides, Porphyromonas, Prevotella, Fusobacterium, and other anaerobic gram-negative rods. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA. 2007, Washington, D.C: ASM Press, 1: 880-901. 9Google Scholar
- Funke G, Bernard KA: Coryneform gram-positive rods. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA. 2007, Washington, D.C: ASM Press, 1: 485-514. 9Google Scholar
- Abbott SL: Klebsiella, Enterobacter, Citrobacter, Serratia, Plesiomonas, and other Enterobacteriaceae. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA. 2007, Washington, D.C: ASM Press, 1: 698-715. 9Google Scholar
- Bannerman TL, Peacock SJ: Staphylococcus, Micrococcus, and other catalase-positive cocci. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA. 2007, Washington, D.C: ASM Press, 1: 390-411. 9Google Scholar
- Miller JM, Krisher K, Holmes HT: General principles of specimen collection and handling. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgensen JH, Landry ML, Pfaller MA. 2007, Washington, D.C: ASM Press, 1: 43-54. 9Google Scholar
- Pinto D, Santos MA, Chambel L: Thirty years of viable but nonculturable state research: unsolved molecular mechanisms. Crit Rev Microbiol. 2013, [Epub ahead of print]Google Scholar
- Trevors JT: Viable but non-culturable (VBNC) bacteria: gene expression in planktonic and biofilm cells. J Microbiol Methods. 2011, 86: 266-273. 10.1016/j.mimet.2011.04.018.PubMedView ArticleGoogle Scholar
- Kessler L, Piemont Y, Ortega F, Lesens O, Boeri C, Averous C, Meyer R, Hansmann Y, Christmann D, Gaudias J, Pinget M: Comparison of microbiological results of needle puncture vs. superficial swab in infected diabetic foot ulcer with osteomyelitis. Diabet Med. 2006, 23: 99-102. 10.1111/j.1464-5491.2005.01764.x.PubMedView ArticleGoogle Scholar
- Mutluoglu M, Uzun G, Turhan V, Gorenek L, Ay H, Lipsky BA: How reliable are cultures of specimens from superficial swabs compared weith those of deep tissue in patients with diabetic foot ulcers. J Diabetes Complications. 2012, 26: 225-229. 10.1016/j.jdiacomp.2012.03.015.PubMedView ArticleGoogle Scholar
- Senneville E, Morant H, Descamps D, Dekeyser S, Beltrand E, Singer B, Caillaux M, Boulogne A, Legout L, Lemaire X, Lemaire C, Yazdanpanah Y: Needle puncture and transcutaneous bone biopsy cultures are inconsistent in patients with diabetes and suspected osteomyelitis of the foot. Clin Infect Dis. 2009, 48: 888-893. 10.1086/597263.PubMedView ArticleGoogle Scholar
- Amann RI, Ludwig W, Schleifer KH: Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev. 1995, 59: 143-169.PubMedPubMed CentralGoogle Scholar
- Fredericks DN, Relman DA: Sequence-based identification of microbial pathogens: a reconsideration of Koch’s postulates. Clin Microbiol Rev. 1996, 9: 18-33.PubMed CentralGoogle Scholar
- Gonzalez A, Stombaugh J, Lozupone C, Turnbaugh PJ, Gordon JI, Knight R: The mind-body-microbial continuum. Dialogues Clin Neurosci. 2011, 12: 55-62.Google Scholar
- Gordon JI: Honor thy gut symbionts redux. Science. 2012, 336: 1251-1253. 10.1126/science.1224686.PubMedView ArticleGoogle Scholar
- Dunne WM, Westblade LF, Ford B: Next-generation and whole-genome sequencing in the diagnostic clinical microbiology laboratory. Eur J Clin Microbiol Infect Dis. 2012, 31: 1719-1726. 10.1007/s10096-012-1641-7.PubMedView ArticleGoogle Scholar
- Xuan J, Yu Y, Qing T, Guo L, Shi L: Next-generation sequencing in the clinic: promises and challenges. Cancer Lett. 2012, doi:10.1016/j.canlet.2012.11.025. [Epub ahead of print]Google Scholar
- Fromin N, Hamelin J, Tarnawski S, Roesti D, Jourdain-Miserez K, Forestier N, Teyssier-Cuvelle S, Gillet F, Aragno M, Rossi P: Statistical analysis of denaturing gel electrophoresis (DGE) fingerprinting patterns. Environ Microbiol. 2002, 4: 634-643. 10.1046/j.1462-2920.2002.00358.x.PubMedView ArticleGoogle Scholar
- Zoetendal EG, Akkermans AD, de Vos WM: Temperature gradient gel electrophoresis analysis of 16S rRNA from human fecal samples reveals stable and host-specific communities of active bacteria. Appl Environ Microbiol. 1998, 64: 3854-3859.PubMedPubMed CentralGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.