Feature: Weapons of mass infection

This feature appeared in the May/June 2010 issue of Australian Life Scientist. To subscribe to the magazine, go here.

As the 2009 Fenner Award winner, Associate Professor Elizabeth Hartland presented the prestigious Fenner lecture at the Australian Society of Microbiology’s (ASM 2010) annual conference in Sydney in July. Hartland and her team at the University of Melbourne study highly specialised systems used by many pathogenic bacteria to boost their own virulence. One such bacterial system is the Type III secretion system (T3SS), which was identified in the early 1990s in Yersinia, a genus of bacteria infamous for causing plague. It turns out T3SS is a widely used mechanism in many gram-negative bacteria that infect humans, animals and plants.

The T3SS is, in fact, central to the pathogenicity of those bacteria that employ it. Bacteria use the system to not only secrete virulence factors, but also to inject them directly into the host cells being infected. The injection part of the system is a complex and sophisticated syringe structure that spans the bacterial inner and outer membranes, thus setting up direct access from the bacterial cytosol to the host cell interior.

A bacterium may have hundreds of these structures lined up against the membrane of an invading organism ready to inject proteins directly into the host cell when the time is right and, in this way, hijack crucial host cell mechanisms for their own sinister purposes. These injected proteins often resemble the host’s own host signalling components and so are able to function in or subvert a normal cellular pathway.

T3SS proteins – and there may be 20 to 40 encoded by a single bacteria – are grouped into three categories: structural, effector and chaperone proteins. Many groups, including Hartland’s, focus on the effector proteins, which actually do the dirty work. T3SS effectors use one or more common pathogenic mechanisms to manipulate their host biology such as interfering with the host cell cytoskeleton and/or cellular trafficking to promote attachment and invasion, inducing cytotoxicity and hijacking the immune system.

A fascinating aspect of this system is that while the T3SS is largely conserved across very different bacterial species in structural terms, the effector molecules they deliver are unique for each species. However, the effectors are the least characterised of all those proteins making up the T3SS with many remaining a virtual black box. Understanding the biology of these proteins is thus a goal of many research groups around the world, including Hartland’s, with the ultimate view to identifying novel targets for prevention and treatment of several infectious diseases.

Hooked from the beginning

Hartland has been interested in these secretion systems since they were first identified, which was during her PhD at the University of Melbourne in the group of Roy Robins-Browne. She was working on pathogenic Yersinia bacteria, although gladly not the famous species responsible for bubonic plague.

“This bacteria contained a plasmid that was essential for virulence, but no-one knew anything else about it,” says Hartland says. “It then came out that this plasmid encoded a new type of secretion system: the T3SS.”

From that point Hartland became very interested in host-pathogen interactions and, in particular, what happens to the effector proteins once injected into a host cell and how they advantage the infecting bacteria.

At the end of her PhD, Hartland undertook a postdoc in the UK on a Royal Society NHMRC Howard Florey fellowship. At the Imperial College in London she started working on enteropathogenic E. coli, which also have a T3SS. Her focus was nutting out molecular functions for the E. coli effector proteins, how and where these novel virulence factors interact with host cell systems to boost pathogenicity.

---PB---

Now that she’s an independent researcher back in Australia with her own group and many different projects on the go, Hartland still plays with pathogenic E. coli that cause diarrhoea in humans, and it is these that she will discuss in Sydney.

Enteropathogenic E. coli is an extracellular gastrointestinal pathogen, so they are not invasive as such but remain stuck onto the lumen of the gut. Quite a bit is known about how these bugs colonise the intestine in the first place, says Hartland.

“They form very interesting lesions on the gut wall, and then even though they are outside the intestinal cells, they use their very neat secretion system to inject effector proteins and manipulate what goes on inside their target cells.” And it is this part of the infection that still poses many questions.

“We know that there are at least 20 to 40 of these effector proteins, depending on the type of E. coli, and we became interested in a few of them with no known function. We have since discovered that these factors actually interfere with inflammatory signalling pathways in infected cells. For instance, some interfere with the activation of the crucial inflammatory mediator, NFkB, which leads to diminished IL-8 secretion.”

IL-8’s job is to stimulate a defensive inflammatory response in the host when the colonising bacteria are first detected. Hartland’s findings thus indicated that these E. coli T3SS effectors act very early on in the host-pathogen interaction to suppress inflammation so that the bacteria can get a better foothold and boost their numbers before meeting the full force of the host immune response.

“Because this is a fairly recent finding, we don’t yet know what happens subsequently, such as whether these proteins also have a role in how the infection resolves. We do think though that this early action assists the bacteria in getting the infection established.

“Of course, normally the host does win,” Hartland adds, “once the inflammation and the full immune response kick in, the bacteria are cleared, but in that time the colony has had time to gather significant numbers and cause symptoms such as diarrhoea. So, the bacteria are shed and can move on to infect the next person.”

It therefore seems that the secretion system is important for both the ongoing pathogenicity of the infection and its transmission to other hosts. “These bacteria are more sophisticated than we think.”

Where, when and how

Hartland’s team have found several proteins that do this, the best characterised of which are called NleB and NleE - these were selected initially from a bank of bioinformatically identified effector proteins. Hartland suspected that they might act in different signalling pathways at different points to affect inflammation.

“We are now trying to dissect where these proteins act, in which pathways, which targets they inhibit and how they do it. So we have now become experts in these terribly complicated NFkB and MAP kinase pathways and I have charts of them all over my wall!”

For two of the three proteins, they know where they act and are fairly sure what they act on. Now they are just trying to nail the ‘how’. “For instance, do they interfere with phosphorylation or proteosomal degradation? We want to identify the exact biochemical mechanisms involved.”

So, just how does Hartland and her team work out what the bacterial proteins are doing among the host signalling networks? “Because we know that these proteins get injected by bacteria into the host cell and act there, we can study their effects by themselves without the bacteria by transfecting genes encoding these proteins into cultured cells and assaying the effects,” Hartland explains.

---PB---

A common approach for the group is to clone a GFP fusion of the virulence protein of interest, and express it in HeLa cells or similar. The cells are then hit with something like TNF or IL-1ß to stimulate the cell’s inflammatory signalling pathway, and luciferase reporter assays provide different readouts of effects, such as NFКB or interferon gamma production. They might also look at phosphorylation or degradation of different known cellular proteins.

“Each of these readouts tells us something about the status of a specific signalling pathway, and how it functions in the presence and absence of the bacterial virulence factor.”

The next step is to test the effect of the protein in isolation with what really happens during a bacterial infection. Cell culture models are again used, usually Hela cells, and these are infected with live enteropathogenic E. coli or mutants thereof.

“We can test the wild-type bacteria and ones deleted for each or combinations of these virulence proteins, which we can then reintroduce by plasmid expression as needed,” Hartland adds. “We also have mutants that allow us to study the influence of specific regions of the proteins.”

Hartland’s group are also setting up an in vivo infection model in mice using a bacterium called Citrobacter rodentium, which is basically a mouse version of human enteropathogenic E. coli that has all the same genes. “It is a really good natural infection model that is close enough to the human system.” This work has just started and there is a lot to do, but Hartland hopes that studying the mouse immune response to the Citrobacter infection will back up their in vitro findings.

Ultimately, they would like to identify targets with potential for therapeutic applications.

Hartland is also interested in the emerging generalised effects of these effector proteins on hosts. “Finding out exactly what these proteins do in a couple of species might explain why so many bacteria have them and what are they doing for bacteria more generally in terms of host-pathogen interactions, which revolve around this subversion of the host processes.”

This feature appeared in the May/June 2010 issue of Australian Life Scientist. To subscribe to the magazine, go here.