Genetic content of clinical pneumococcal isolates and its relation to disease outcome

Abstract

Streptococcus pneumoniae is a Gram-positive bacterium that can cause a wide range of diseases. These include otitis media, sinusitis, pneumonia and meningitis. It can, however, also cause asymptomatic carriage thereby behaving almost like a commensal. It is of great interest to try to determine factors playing a role for if the bacterium will cause carriage or whether it will cause an invasive disease (meaning that the bacterium is sampled from the blood or the meninges).

The aim of this thesis is to investigate differences in genetic content in invasive and carriage isolates. In the first study we set out to see how well a molecular typing method, MLST, and genetic content as determined by microarray correlated. We found that isolates differing in up to two MLST alleles and belonging to the same clonal complex clustered together using microarray data. Hence the method of MLST, measuring differences in 7 house-keeping genes, correlates well with genetic content as determined by microarray. We also suggest an alternative typing method, a PCR consisting of 25 accessory genes, which gives the same discrimination as MLST clonal complexes. This method must, however be investigated further.

In paper II we wanted to see how well isolates differing in more than two alleles, and still belonging to the same clonal complex correlated genetically. We found that both the number of alleles differing and if the isolates compared were of the same CC and / or Serotype affected the number of genes differing. For the three larger CCs investigated we saw that for one of them the number of genes increased with the number of alleles differing. For the other two there was a large increase when the serotype of the isolates compared shifted, indicating the possibility that these CC:s in fact come from more than one ancestor ST.

Paper III investigated the genetic content of isolates of different invasive disease potential to see if any particular accessory region was associated with either carriage or invasiveness. No obvious candidate was found. ARs 6 and 34 were found in most of the invasive isolates and were absent in most of the isolates of low invasive disease potential and were tested for invasiveness in a mouse model of infection. We were not able to see any differences between knock-out mutants of these regions and the wild type. This, in addition to the fact that many genes found to be important for virulence in STM screens were absent, suggest that there is a great genetic redundancy affecting the ability of different isolates to cause invasive disease.

In the fourth study we determined the invasive disease potential of isolates from Stockholm, Sweden 1997-2004. Invasiveness for serotype well matched previous studies with serotype 1, 4, 7F having a high invasive disease potential and 6A, 19F and 23 F having a low invasive disease potential. We did, however, find differences between clonal types of the same serotype, for serotypes 14 and 6B. We further investigated 6B by whole genome microarray and whole genome sequencing and found a number of differences. Among these are two different variants of the important protein PspA and the absence of PcpA, important for disease in the lung, from one of our isolates. We were also able to determine that the four serotype 6B isolates differed in presence/ absence of prophages, a factor that may be of importance for virulence.