Nasopharyngeal colonization is the first step of invasive pneumococcal disease . However, it is not known what triggers the transition from colonization to invasiveness. Our on-going work on pneumococcal biology indicates that environmental factors, such as changing oxygen concentration, differences in metal and sugar composition of tissues, can have a fundamental impact on pneumococcal virulence [39, 40]. However, although these environmental factors are important, they do not explain fully what triggers the sudden change from colonization to invasiveness. Therefore, we investigated whether other host factors, such as stress hormones, might be important for transition of the pneumococcus from commensal to pathogen . The reason for this hypothesis stems from the rapid change in the concentration of stress hormones due to physical and emotional stress, from stress hormones’ adverse effect on immune system function , and from the microbial ability to recognize and process human stress hormone signals .
In this study we showed that S. pneumoniae responds to levels of catecholamine found within the circulation of inotrope-medicated patients  with increased growth and virulence, which could have a major impact on the progression of pneumococcal infection or transmission to new hosts. Many predisposing factors for pneumococcal diseases including emotional and cold stress, and overcrowding are known to increase stress hormone levels. In addition, catecholamine inotropes are administered up to 50% of patients in intensive care unit (ICU) , and up to 56% of patients with pneumococcal pneumonia are admitted to ICU . Hence, in addition to endogenously produced stress hormones, pneumococci are exposed to externally applied catecholamine inotropes. Growth stimulation of S. pneumoniae came about due to the inotrope providing essential Fe for growth from the host iron binding protein transferrin, which was directly bound by the bacteria. Interestingly, the supposedly simple in function PspA and PspC surface proteins were found to play a major role in NE mediated growth induction. When the genes for PspA and PspC were mutated, the ability of S. pneumoniae to utilize the additional Fe provided from transferrin by the catecholamine was reduced. The uptake of the radiolabelled NE was similarly reduced. Also, mutating pspA and pspC appeared to block NE-effects on gene expression, which agrees well with the non-growth responsiveness observed. Why PspA and PspC should be so important in mediating catecholamine responsiveness in the pneumococcus is unclear. The two proteins are important in virulence as they have been shown to play a pivotal role in the inhibition of complement-mediated opsonization [42, 43], in prevention of lactoferrin killing , and in facilitating the microbe’s attachment to the respiratory tissues and the brain microvascular endothelium [19, 30–32]. PspA is also known to bind to lactoferrin . In addition, PspA and PspC have been shown to elicit protective antibody response against invasive pneumococcal infection, hence they are considered to be promising vaccine candidates . Although their contribution to S. pneumoniae-host interaction is well studied, comparatively little is known about their role in pneumococcal physiology. Previously, using recombinant PspA and a strain mutated in pspA, it was shown that PspA, but not PspC, is responsible for pneumococcal binding to human lactoferrin, which was suggested to be important to overcome the iron limitation at mucosal surfaces [44, 45]. Contrary to previous reports [32, 45], in this study we consistently demonstrated that S. pneumoniae could bind to transferrin, and acquire iron from this glycoprotein, and that uptake of Fe from Tf was enhanced when NE was present. The reason for this discrepancy could be due to different culture conditions, and detection technology used for transferrin binding. For example, unlike Hakansson et al., (2001)  we used a serum based medium to prepare pneumococcal cultures, which can affect the synthesis of proteins involved in binding to Tf. Currently, the mechanism of PspA and PspC mediated pneumococcal response to NE is not known and so defining how PspA and PspC are mediating catecholamine responsiveness is a current focus of our laboratories. However, based on the available data it is clear that these surface proteins are required for recognition and/or internalisation of NE since the mutation of pspA or pspC abolished NE responsiveness, reduced NE uptake and blocked catecholamine-induced gene responsiveness. This clearly indicates that the proteins encoded by these genes may be acting as a sensor molecule. It is not surprising that both PspA and PspC are involved in stress hormone mediated effects in S. pneumoniae given these proteins are coded by paralogous genes, and previous studies have demonstrated their involvement in similar biological events [31, 42, 43]. In future experiments, we plan to investigate to which downstream targets PspA and PspC relay NE mediated messages.
A recent study by Marks et al.
 showed that NE treatment of biofilms formed in vitro, and in vivo in the nasopharynx leads to dispersion of S. pneumoniae, and the dispersed cells display distinct phenotypic traits that are different from those of both biofilm and broth-grown planktonic bacteria. The dispersed pneumococci were shown to have differential virulence gene expression, and had a significantly increased ability to disseminate and cause infection in the middle ear, lungs, and bloodstream. Our results are consistent with Marks et al.,  in that the pneumococcus responds to NE, and that treatment with the catecholamine leads to differential gene expression. On the other hand, contrary to the Marks et al. study, who used biotic surfaces to determine NE’s role in pneumococcal dispersion from biofilms, our results show that in host like serum-containing media the catecholamine aggregates the pneumococci and promotes biofilm formation on abiotic surfaces. The reason for this seeming discrepancy could be due to methodological differences and also be attributed to NE’s possible dual function in biofilm formation. In other words, NE can initially promote bacterial biofilm formation (our current study) and after a certain stage in the infection process, depending on the microbial growth phase, may also promote dispersion of the pneumococci . Interestingly, a recent paper from Gonzales et al.
 found that addition of a non-therapeutic level of NE (100 μM) stimulated several-fold increases in growth but in contrast to our data, had an inhibitory effect on pneumococcal biofilm formation, as measured by attachment to host cells. Therefore, it is clear that further work is required to understand this differential effect of the catecholamine on biofilm formation.
In this study NE mediated Fe uptake from Tf was identified as the mechanism responsible for the observed growth effect of NE in serum based media. However, our gene expression analysis in wild type D39 shows that NE has an even wider effect on pneumococcal physiology. For example, the expression of genes coding for glycosidases (nanA, nanB, bgaC and strH), which are responsible for deglycosylation of host glycans and play important role in pneumococcal colonization and invasiveness [36, 37], were significantly upregulated in the presence of NE. Moreover, differential expression of genes involved in transcriptional regulation (SPD_0939), competence development (comX), galactose metabolism (galK), and iron transport (piuA) was also detected, indicating the comprehensive effect of NE on pneumococcal metabolism. Currently it is not known how the pneumococcus detects and processes stress hormone signals, though there is a clear involvement of PspA and PspC in the response mechanism. Therefore, investigating the underlying genetic mechanisms for detection and processing of catecholamine signals is a priority. Also, in this study we found that the pneumococcus responds similarly to a variety of catecholamine stress hormones (NE, as well as dopamine and epinephrine), which is in contrast to the situation demonstrated in Mycoplasma hypopneumoniae
. This finding is also of clinical significance as 300 μM epinephrine may be administered directly to ventilated patients to reduce airway inflammation .
Bacteria have evolved mechanisms to sense the changes in the stress hormone levels using receptors, which appear to be specific and able to differentiate between different stress hormones [2, 4, 48]. Using α and β receptor antagonists, we showed the presence of putative adrenergic and dopaminergic receptors in three Gram-negative bacteria: Escherichia coli, Salmonella enterica and Yersinia enterocolitica
. Our results demonstrated that catecholamine- induced growth in these bacteria could be blocked by catecholamine α-receptor antagonists, but not by antagonists for β adrenergic receptors. But, so far, no comprehensive study has been conducted to investigate proteins responsible for stress hormone recognition in Gram positive bacteria. Identification of such receptors in the pneumococcus would enhance our understanding of S. pneumoniae-host interactions and may offer alternative therapeutic options against pneumococcal diseases.