Biofilm formation in invasive disease caused by fungus of the genus Candida

Candida species are ranked as the fourth leading cause of nosocomial bloodstream infection. The association of the infection with several risk factors has been reported. The most predominant risk factor, however, is the presence of an intravenous catheter in patients, which provides an artificial surface for Candida spp. to form biofilms and subsequently initiate infection.

A total of 940 yeast isolates, mostly Candida spp., causing bloodstream infection in Sweden were collected between September 2005 – August 2006 and species identified through sequence polymorphisms in 40 nucleotides in the internal transcribed spacer region 2 (ITS2) of the ribosomal RNA gene obtained by pyrosequencing (paper I). Using this newly develop molecular approach, all isolates could be reliably identified down to the species level. In addition, intraspecies sequence variation in Candida albicans and Candida glabrata suggests subclassification of isolates. As Candida specification is associated with differences in susceptibility to antifungal drugs, accurate species identification aids treatment decisions. Pyrosequence analysis of the 40 nucleotides in the ITS2 region is fast and reliable, and can therefore contribute to the high quality management of patients with invasive fungal infections.

Biofilm formation of Candida spp. isolates was determined with a model mimicking the clinical conditions of the intravenous catheter surface through using a silicone elastomer, the host contribution through conditioning the surface with 10% human serum and the parenteral nutrition solution by a defined growth medium containing 10% glucose (paper II). Under these experimental conditions, biofilms of all Candida spp. with the exception of Candida parapsilosis were found to be composed of basal yeast cells. As a correlation between the amount of biofilm and the metabolic activity was observed, biofilm formation could be directly compared among Candida spp. isolates with isolates forming no, low and high biofilm. Biofilm formation was less prevalent in C. albicans isolates compared to non-albicans Candida spp. isolates. Thus, biofilm formation seems to be more significant for infections caused by non- albicans Candida species than by C. albicans. It is, therefore, possible that C. albicans uses mechanisms other than biofilm formation to cause bloodstream infection (paper II).

Among Candida spp., C. parapsilosis is the second/third most common cause of bloodstream infection. Biofilm formation of C. parapsilosis was highly variable among 33 epidemiologically independent isolates (paper II and paper V). A nosocomial outbreak involving C. parapsilosis has been reported from a hospital in Southern Sweden (paper III). The clonal origin of the C. parapsilosis isolates was confirmed. All outbreak isolates robustly showed high level of biofilm formation with a compelx structure on three surface materials, while expression of major virulence factors was low (paper IV). This finding indicated the significance of biofilm formation of C. parapsilosis as a determinative factor to cause the outbreak infection.

Within 33 clinical isolates of C. parapsilosis causing bloodstream infection, two different biofilm architectures could be identified in the 15 isolates forming high level of biofilm (paper V). Of these 15 isolates, 11 showed a complex structure biofilm consisting of macro-colonies with a spider-like appearance composed of aggregated yeast cells and pseudohyphae, while four isolates formed monolayers of pseudohyphae. Surprisingly, biofilm formation of isolates with high biofilm formation including an outbreak isolate were independent of the transcription factor Bcr1, a major biofilm regulator, although BCR1 deletion affected colony switching and cell morphology in those isolates. As novel phenotypes, BCR1 regulated secretion of aspartyl proteinases and susceptibility to antimicrobial peptides irrespectively of the biofilm formation phenotype in all tested isolates of C. parapsilosis (paper V).

In conclusion, work in this thesis characterised biofilm formation in an extensive number of Candida spp. isolates causing bloodstream infection. Detailed analysis of biofilm development in C. parapsilosis revealed that high biofilm formation in C. parapsilosis is independent of the major biofilm regulator Bcr1. This finding will have an impact on the development of biofilm prevention strategies in C. parapsilosis.

List of scientific papers

I. Pannanusorn S, Elings MA, Römling U, Fernandez V. Pyrosequencing of a hypervariable region in the internal transcribed spacer 2 to identify clinical yeast isolates. Mycoses. 2012 Mar; 55(2): 172-80.
https://doi.org/10.1111/j.1439-0507.2011.02064.x

II. Pannanusorn S, Fernandez V, Römling U. Prevalence of bio lm formation in clinical isolates of Candida species causing bloodstream infection. Mycoses. 2013 May; 56(3): 264-72.
https://doi.org/10.1111/myc.12014

III. Brillowska-Dabrowska A, Schön T, Pannanusorn S, Lönnbro N, Bernhoff L, Bonnedal J, Haggstrom J, Wistedt A, Fernandez V, Arendrup MC. A nosocomial outbreak of Candida parapsilosis in southern Sweden verified by genotyping. Scand J Infect Dis. 2009; 41(2): 135-42.
https://doi.org/10.1080/00365540802585301

IV. Pannanusorn S, Schön T, Morschhäuser J, Römling U. BCR1 independent biofilm formation of outbreak related Candida parapsilosis isolates from nosocomial bloodstream infections. [Submitted]

V. Pannanusorn S, Ramírez-Zavala B, Lünsdorf H, Agerberth B, Morschhäuser J, Römling U. Characterization of biofilm formation and the role of BCR1 in clinical isolates of Candida parapsilosis. [Submitted]