Effects of Clostridium perfringens Alpha-Toxin (PLC) and Perfringolysin O (PFO) on Cytotoxicity to Macrophages and on Escape from Phagosomes

Citation

O'Brien, D.K., and Melville, S.B. (2004). Effects of Clostridium perfringens alpha-toxin (PLC) and perfringolysin O (PFO) on cytotoxicity to macrophages, on escape from the phagosomes of macrophages, and on persistence of C. perfringens in host tissues. Infect Immun. 72, 5204-15.

Link

http://iai.asm.org/cgi/content/full/72/9/5204

Introduction/Motivation

Clostridium perfringens is a gram-positive bacterium that is responsible for a host of human diseases including enteritis necroticans (Pigbel), acute food poisoning, antibiotic-associated diarrhea, and most notably gas gangrene (clostridial myonecrosis). It is a non-motile anaerobe that can cause signs of gangrene to appear in anoxic conditions (if blood flow to the wound is disrupted) only 6 hours after C. perfringens bacterial infection in a wound.

A distinguishing characteristic of gangrene is the absence of phagocytic cells (polymorphonuclear cells (PMNs), monocytes and macrophages) that would normally be expected at the site of a wound. This suggests C. perfringens not only evades the phagocytic cells of the innate immune system, but also plays a part in destroying them. Moreover, it is noteworthy that appreciable levels of C. perfringens can accumulate even if a small number of bacteria infect a wound site that presumably has many more phagocytic cells from the immune system.

It is believed that the two main toxins in C. perfringens that allow it to cause gas gangrene are alpha toxin (PLC) theta toxin (perfringolysin O or PFO). PFO is a cholesterol-dependent cytolysin that forms large pores when it comes into contact with cholesterol-containing membranes. PLC, a phospholipase C and sphingomyelinase, is also a membrane-active cytoxin, which may also play a part in mediating the escape from a macrophage’s phagosome. However, previous studies have shown that PLC alone is not sufficient for phagosomal escape.

This paper documents the study of the roles of PLC and PFO in the survival of C. perfringens in the presence of phagocytic cells and in its ability to destroy such cells given enough time to persist in the site of a wound.

Summary of Methods and Results

In order to test the effects of PFO and PLC, the research group needed to obtain strains of C. perfringens where the genes for each of these proteins was disrupted or deleted as well as the wild type (strain 13) where both were functional and a double knockout. They could then compare how well these different strains survived in the presence of phagocytic cells and how well they killed phagocytic cells. They obtained a strain (PLC-) that did not express PLC from another laboratory, but they needed to construct their own strains where PFO was disrupted as well as a combination of the two. They accomplished this by screening for cells that had recombined a plasmid containing a non-functional PFO gene fragment with the endogenous PFO gene found on the C. perfringens chromosome (they did this to both the wild-type and PLC- strains). They characterized the strains that they created with Southern hybridization analysis to confirm that the recombination was successful (this also revealed to them that plasmid dimers had recombined with the genome) using probes for both the mutant PFO gene and the plasmid resistance marker. They also did functional testing of the strains that they created by plating on sheep blood agar and looking for the lack of inner zone hemolysis in the case of PFO- strains and lack of outer zone hemolysis in the case of PLC-strains. When they tested how well their PFO- strains lysed sheep blood cells, they found that the PFO- (henceforth called pfoA mutant) strain showed only 0.96% of the wild type PFO activity and the PLC-PFO- (henceforth called plc pfoA mutant) strain showed only 0.02% os the wild type PFO activity.

The experiments performed with the four different strains (wild type, PLC-, pfoA mutant, and plc pfoA mutant) demonstrated the following major results:

Both PFO and PLC are necessary for C. perfringens survival in the presence of mouse peritoneal macrophages (primary cells). They put the different strains of C. perfringens into cultures of primary macrophages (in 24-well tissue culture plates), lysed cultures (using Triton-X) with each strain at various time points, and plated dilutions the bacteria that they obtained in appropriate bacterial growth conditions to determine how many had survived. In aerobic conditions (figure 2 at the left), they found that the number of wild type bacteria increased at first before decreasing to ~15% of the initial level at 24 hours after inoculation. In contrast, the PLC- and pfoA mutant strains decreased immediately and did not survive more than 7 hours after inoculation. Surprisingly, the plc pfoA mutant strain survived better than the strains that had only one of the two genes disabled. The authors attribute that to the higher amount of this strain that persists in the culture media without any cells (figure 2B). Under anaerobic conditions, all of the strains grew exponentially (regardless of whether or not macrophages were present) and completely lysed the macrophages after 5 hours.

PFO is the main factor that allows C. perfringens to kill J774 (macrophage-

like immortal cell line) and mouse peritoneal macrophages under aerobic conditions. Mouse macrophages incubated with the cell-free supernatents of each of the different strains have a much higher rate of cell death in strains containing PFO (wild type and PLC-) than in strains that do not contain PFO. This is quantified in figure 3 at the right using an assay that measures the amount of

the cytoplasmic protein LDH released into the cell media (the protein is released when the macrophage membrane is disrupted and its contents are spilled into the surrounding solution). The pictures at the top of the accompanying figure also show that macrophages that are exposed to supernatent from bacteria producing PFO become rounded and dissociate from the plate much more than cells exposed to other bacterial supernatent.

Critique

This paper does demonstrate some aspects of the roles of PFO and PLC in C. perfringens. However, there are definitely places where they dismiss unusual or unexpected results without enough investigation about why they occurred. These places have a potential for creating questions regarding their data and essentially, the results that they concluded.

The most gaping lapse of explanation occurs in the data from figure 2 where the plc pfoA mutant strain survived longer than both the PLC- and pfoA strains. Even though disabling each of these genes individually is detrimental to C. perfringens survival, it seems that disabling both of them provides some viability. The authors state that they believe that this is because the plc pfoA mutant is more viable in DMEM cell media. However, they fail to note that the plc pfoA mutant survives better in the presence of macrophages than it does in media alone. This begs the question of how the cells that are supposed to be killing them actually help them survive. Thus, it seems that the media is not the only contributor to the increased cell survival of the plc pfoA mutant strain. It is possible that disabling both genes has altered other aspects of the cell via cell pathways that it no longer makes sense to do a direct comparison in the way that they have done throughout the paper.

With respect to that same figure, they also mention that the data for 1 hour and 2 hours after inoculation is omitted because of an adverse affect on the C. perfringens from the Triton-X used to lyse the macrophages. From their explanation, it is not clear how the bacteria were affected or how they ensured that the other data points were not skewed by the same effects. This omission of details causes one to question why the data was omitted and what it actually showed.