, a major etiologic agent of dental caries in humans, produces organic acids during fermentation and must survive the resulting acidic milieu. One result of decreasing environmental pH is an increase in the proportion of monounsaturated membrane fatty acids (UFAs) (4
). We have shown, via insertional inactivation of fabM
in S. mutans
, that the gene is solely responsible for unsaturated fatty acid production (3
). Moreover, the fabM
mutant strain exhibited several markedly acid-sensitive attributes that differed from the wild type, S. mutans
UA159: it was approximately 3.5 log units more sensitive to extreme acid stress, it produced approximately 1.5 log units less acid, and it was able to maintain approximately half of the transmembrane ΔpH (3
). In addition, the construct was highly stable for at least 20 generations in our chemostats (data not shown). The ability to survive low-pH conditions is thought to be critical for S. mutans
to persist in the oral cavity and cause disease. The availability of a defined strain, defective in unsaturated fatty acid biosynthesis and acid resistance, provided an opportunity to investigate whether acid sensitivity influences transmissibility from host to host and pathogenesis in a rodent model of dental caries (1
Transmission of a fabM
) was measured by the ability of the organism to pass from dams to pups and from infected cage mate to uninfected cage mate as outlined in Fig. . For the transmission experiment, eight litters of Sprague-Dawley rats, aged 15 days, and their dams were obtained from Harlan Laboratories. Dams were determined to be mutans streptococci free and sialoacroadenitis virus free by previously described methods (1
). Mid-logarithmic cultures of either wild-type UA159StR
) or UR117StR
, a spontaneous streptomycin-resistant strain of UR117, were used to infect two dams on three consecutive days, while the remaining dams were uninfected. It should be noted here that while the streptomycin resistance genotype of the two test strains may not be identical, growth of the UR117StR
strain was virtually identical to that of the UR117 strain in vitro and that streptomycin resistance has not previously been shown to affect results from rodent infection studies (2
). In the present study, in all cases, infected animals received Diet-2000 (5
) and drinking water containing 5% (wt/vol) sucrose ad libitum; uninfected animals were fed lab chow and water ad libitum. On day 6, pups (aged 21 days) associated with infected dams were screened for successful infection of either strain by oral swabbing and plating on mitis-salivarius medium (Difco) containing 500 μg/ml streptomycin (MSS). None of the pups caged with infected dams became infected by day 6 (Table ).
Transmission experiment outline. SDV, sialoacroadenitis virus.
Transmission of S. mutans strains UA159StR and UR117StR from infected to uninfected cage matesa
After we determined that the pups had not become infected with detectable numbers of bacteria (group I and group II animals [Fig. ]), the animals were directly infected by oral swabbing with UA159StR or UR117StR, respectively. Following oral swabbing, all pups became infected by experimental day 8 (data not shown), demonstrating that both strains, UA159StR and UR117StR, were capable of productive infection. The infected pups from both groups were paired with uninfected cage mates on experimental day 10. We observed that within 1 day of being paired with pups infected with the wild type, UA159StR, 6 out of 16 recipient pups had become infected (Table ). By day 17, all of the recipients had become infected with the wild-type strain (group IV). In contrast, only four of the uninfected recipient pups became infected with the fabM defective strain (UR117StR; group V).
The experiment was concluded on experimental day 19, at which time the rat pups were aged 34 days. The average number of bacteria recovered from wild-type donors, expressed as CFU/ml of jaw sonicate, was higher than that determined for mutant strain donors (Table ). The average recovered CFU from the 16 recipient animals successfully infected with strain UA159StR was approximately 3 log units higher than the average recovered CFU from the four animals that had detectable infection by strain UR117StR (Table ).
Recovery of strains UA159StR and UR117StR from pups on day 19a
A separate caries study was undertaken to test the hypothesis that the fabM mutant strain, UR117StR, would be less cariogenic than the wild-type strain, UA159StR. Four litters of pups, aged 15 days, and their dams were obtained from Harlan Laboratories. Dams were screened as in the transmission study. Two dams were infected with actively growing UA159StR or with UR117StR on days 1 and 2 (pups were aged 16 and 17 days), and infection was confirmed by oral swabbing. On day 6 (pups aged 21 days), the pups were weaned and infected with either UA159StR or UR117StR (based on their initial exposure) for two consecutive days. Infection was confirmed via oral swabbing. Rats were fed Diet-2000 and water containing 5% sucrose (wt/vol) ad libitum for 5 weeks. Animals were killed by CO2 asphyxiation, and bacterial counts were determined from lower left jaw sonicates. Similar to what was observed in the transmission experiment, the average CFU recovered was significantly higher in animals infected with UA159StR (3.1 × 108 ± 1.2 × 108) than in those infected with UR117StR (7.7 × 107 ± 4.0 × 107) as determined by Student's t test (P < 0.01).
Caries were scored by the method of Keyes (5
) as modified by Larson (6
), and data were evaluated following arcsine transformation (10
). Animals infected with strain UR117StR
experienced far fewer smooth-surface carious lesions than animals infected with UA159StR
. As the severity index increased (from slight to moderate to extensive), the severity score was approximately 90% reduced in animals infected with the fabM
deficient strain (Table ). The smooth-surface caries scores in animals infected with UR117StR
were striking, in that they were even lower than the scores reported previously from experiments involving the well-established virulence factor encoded by gtfB
Incidence and severity of dental caries in rats infected with S. mutans UA159StR (wild type) and S. mutans UR117StR (ΔfabM)a
Typically, sulcal caries scores do not as readily reveal strong differences between infecting strains because the bacteria compact into the fissures of teeth. Nevertheless, we recorded sulcal caries scores from the animals infected with the wild-type or fabM mutant strains. Not surprisingly, the differences were not as pronounced as those seen with smooth-surface caries. However, the sulcal scores were significantly different (at the P < 0.05 level), when the severity of the lesions was taken into account (from moderate to extensive severity [Table ]).
From these studies, it is clear that membrane fatty acid composition plays a significant role in the acid resistance phenotype of S. mutans and has a significant role in the virulence of the organism. We attribute the differences in infectivity and caries-forming ability of the wild-type and fabM strains to the inability of the mutant to produce monounsaturated membrane fatty acids. Our previous reports provided strong physiological evidence that production of UFAs, during growth at low pH, directly impacts the ability of the organism to withstand acid stress. We have shown here that disrupting the ability of S. mutans to produce UFAs also leads to the inability of the organism to be transmitted from infected to uninfected animals, thereby not fulfilling a key component of Koch's postulates. Importantly, the severity of caries in animals infected with the fabM strain was clearly less than those infected with the wild-type strain. The results of this study provide a link between the ability of S. mutans to produce acid, survive acidic environments, and cause severe disease to a single gene product, FabM.