Importance of the Vertebrate Complement System

This colonization assay allowed us to evaluate the specificity of the symbiosis experimentally. For these experiments, a portion of the inoculated blood was incubated in vitro under the same conditions as the animal. This modification allowed us to compare the growth potential of the strain in blood and perhaps detect modifications of the ingested blood inside the leech (Indergand and Graf 2000). One Escherichia coli strain, EcR1, had an interesting colonization phenotype (Fig. 3). The number of CFU per ml decreased both inside the leech and in the in vitro control 18 and 42 h after inoculation. At 42 h, EcR1 had decreased 1,000-fold in concentration. This concomitant decrease was consistent with an antimicrobial property of the ingested blood remaining active inside the leech and killing EcR1. Heat-treatment of the blood prior to inoculation allowed the E. coli strain to proliferate both inside the animal and in the in vitro control, indicating that this antimicrobial property was heat sensitive and also responsible for the reduction of viable E. coli inside the leech. One powerful antimicrobial property of vertebrate blood is the complement system. The complement system uses the classical or alternative pathway to activate the formation of the membrane-attack-complex, which is inserted into the membrane of sensitive cells. This leads to a permeablization of the membrane and the eventual loss of viability. One can prevent the activation of the membrane-attack-complex by pre-treating the blood with chemicals or heat. In vitro experiments, where the blood was pretreated with EDTA, EGTA with Ca2+ or heat treatment allowed the strain to proliferate. These treatments interfere with the activation of the complement system by the classical system and are consistent with a role of the complement system in killing E. coli. The genetic background of EcR1 is not known and it seemed reasonable to assume that Aeromonas strains that can colonize the leech would have to be resistant to the complement system.

Previously it was shown that the lipopolysaccharides (LPS) of the outer membrane is important in conferring resistance to the complement system in Aeromonas (Merino et al. 1992, 1994, 1996). Using well-characterized A. veronii biovar sobria mutants, we decided to test the hypothesis that the

Fig. 3. Colonization of H. medicinalis by E. coli EcR1. During the first 24 h after feeding, the concentration of strain EcR1 decreased inside the animal and the in vitro control. The native microbiota and the introduced strain EcR1 was recovered on blood agar (open bar), strain EcR1 alone on LB containing rifampin (solid bar) and strain EcR1 from the in vitro control (hatched bar). The arrow depicts the cell concentration of strain EcR1 in the inoculum. Reprinted with permission from Indergand and Graf (2000)

Fig. 3. Colonization of H. medicinalis by E. coli EcR1. During the first 24 h after feeding, the concentration of strain EcR1 decreased inside the animal and the in vitro control. The native microbiota and the introduced strain EcR1 was recovered on blood agar (open bar), strain EcR1 alone on LB containing rifampin (solid bar) and strain EcR1 from the in vitro control (hatched bar). The arrow depicts the cell concentration of strain EcR1 in the inoculum. Reprinted with permission from Indergand and Graf (2000)

complement system of vertebrates remains active for some time inside the leech by testing the ability of complement-sensitive Aeromonas mutants to colonize the leech (Braschler et al. 2003). Mutants were available that exhibited a defect in the biosynthesis of LPS and were shown to have an increased sensitivity to the complement system (Merino et al. 1992, 1994, 1996). The LPS is thought to prevent the membrane-attack-complex from reaching the lipid bilayer. These A. veronii biovar sobria mutants were devoid of the O34 antigen LPS and their defect could be complemented with a cosmid carrying the biosynthetic genes for rhammose which are present in the wb gene cluster for the O-antigen. Accordingly, we predicted that if the complement system remained active inside the leech, these mutants should have a reduced ability to colonize their host. As expected, the serum-sensitive LPS mutants had a severe colonization defect that could be reversed either by the heat-inactivation of the blood or by complementing the mutant with the biosynthetic genes for rhammose on a plasmid (Fig. 4; Braschler et al. 2003). These results provide further support to the hypothesis that the complement system of vertebrates contributes to the specificity of the association by preventing other, sensitive bacteria from colonizing the medicinal leech. The mutants with a defect in the ability to synthesize the LPS represent the first mutant class known that is defective in colonizing the digestive tract of the leech.

In a random transposon mutant screen of the symbiotic strain, HM21R, mutants that had become sensitive to antimicrobial properties in the vertebrate blood were isolated and none of the ones tested was able to colonize the leech (Rabinowitz, N., A. Silver, S. Kuffer and J. Graf, unpubl. data). Initial characterization of the mutants suggests that several had transposon insertions not significant

Fig. 4. The serum-sensitive mutant AH-21 had a dramatically reduced ability to colonize the leech. This defect could be restored by complementing the mutant with pLA226 which carries biosynthetic genes for the O-antigen but not by the empty vector control, pLA2917. The P values were calculated with an unpaired T-test using Welch's correction not significant

Fig. 4. The serum-sensitive mutant AH-21 had a dramatically reduced ability to colonize the leech. This defect could be restored by complementing the mutant with pLA226 which carries biosynthetic genes for the O-antigen but not by the empty vector control, pLA2917. The P values were calculated with an unpaired T-test using Welch's correction in genes that showed significant similarity to glycosyl transferases. Preliminary comparisons of the amino acid sequence of these proteins suggest that they differ from the published ones from A. hydrophila and A. veronii biovar sobria.

The colonization phenotypes of S. aureus strain SaR1 and P. aeruginosa strain PaR1 differed dramatically from that of E.coli or the serum-sensitive mutants (Indergand and Graf 2000). Their ability to proliferate inside the leech was significantly inhibited as compared to an in vitro control; however, the mutants were able to survive at a constant level in the leech for 7 days (for example see Fig. 5). These data suggest that the ingested blood is modified in a manner that interferes with the ability of these two species to proliferate inside the leech. This is indicative of a second layer of defense that the symbionts need to overcome. The possible factors that may contribute could be simple modifications of the blood by either the host or symbionts, such as the removal of water and salts from the intraluminal fluid or that the native symbionts out compete the other bacteria for nutrients. Other possibilities are that either host or symbionts release antimicrobial compounds to which these bacteria are sensitive but that the symbionts are resistant

Fig. 5. Colonization of H. medicinalis by S. aureus. SaR1 was unable to increase in concentration inside the leech but was able to persist for at least seven days. The concentration reached was significantly lower than in the in vitro control. The native microbiota and the introduced strain SaR1 was recovered on blood agar (open bar), strain SaR1 alone on LB containing rifampin (solid bar) and strain SaR1 from the in vitro control (hatched bar). The arrow depicts the concentration of strain SaR1 in the inoculum. * and **, the mean concentration differed significantly from the values obtained inside the animal at P <0.05 and P <0.005, respectively (two-sides MannWhitney test). Reprinted with permission from Indergand and Graf (2000)

Fig. 5. Colonization of H. medicinalis by S. aureus. SaR1 was unable to increase in concentration inside the leech but was able to persist for at least seven days. The concentration reached was significantly lower than in the in vitro control. The native microbiota and the introduced strain SaR1 was recovered on blood agar (open bar), strain SaR1 alone on LB containing rifampin (solid bar) and strain SaR1 from the in vitro control (hatched bar). The arrow depicts the concentration of strain SaR1 in the inoculum. * and **, the mean concentration differed significantly from the values obtained inside the animal at P <0.05 and P <0.005, respectively (two-sides MannWhitney test). Reprinted with permission from Indergand and Graf (2000)

themselves. The results described so far begin to paint a picture of multilayered defenses with possible contributions from the leech, the symbionts as well as the vertebrate on which the leech feeds. These defenses together contribute to the unusual specificity of this symbiotic interaction.

A major goal is to identify the mechanisms that are responsible for the establishment and maintenance of the specificity. One powerful approach is to isolate mutants of the native symbiont which exhibit a reduced ability to colonize the leech. The molecular characterization of these mutants may provide us with clues that offer us with an insight into the host environment and may reveal active mechanisms that interfere with microbial growth.

Was this article helpful?

0 0

Post a comment