Microb Ecol (1995) 30:219-225
MICROBIAL ECOLOGY © 1995Springer-VerlagNew York
Inc.
Evolution of Resident Oral Bacterial Biota in BALB/c Mice During Pregnancy and Lactation C. Coulombe, M.C. Lavoie Groupe de Recherche en Ecologic Buccale, Universit6 Laval, Qu6bec, Canada, G1K 7P4 Received: 23 August 1994; Revised: 22 December 1994
Abstract. To assess the influence of pregnancy and lactation on the oral microbial ecology of BALB/c mice, we followed the distribution of the predominant oral bacteria of four groups of these mice during these two periods. Compared with nonpregnant control female mice of the same age maintained under the same conditions, the distribution of the resident oral bacterial species differed significantly only during the lactation period (8-16 days after parturition). This difference could possibly be attributed to hormonal influences and/ or grooming habits. Introduction In humans, several factors of the oral ecosystem have been shown to vary during pregnancy: hormonal levels in gingival crevicular fluid [9,10] and gingival tissues [16, 31], the proportion of Prevotella intermedia (formerly Bacteroides intermedius) in subgingival plaque samples [12, 22], immunosuppression [17, 24, 26], and the proportions of Streptococcus mutans [15]. To assess the possible influences of the physiological variations associated with the periods of pregnancy and lactation on the oral biota, we followed the distribution of the resident oral bacteria in our mouse model. This model possesses a restricted oral microbiota, which facilitates the study of its variation [33]. Our present results indicate that the gestation period does not significantly affect the oral bacterial populations of the BALB/c mice but that a significant effect is observed during the lactation period.
Materials and Methods Mice BALB/c AnNcrl/BR (BALB/c) mice, 8-10 weeks old, were obtained from Charles River Canada (St. Constant, Qu6bec, Canada). After a rest period of 1 week, the mice were mated at our facilities
Correspondence to: M.C. Lavoie, Group de Recherche en l~cologie Buccale, Universit6 Laval, Qu6bec, Canada, G1K 7P4.
220
C. Coulombe, M.C. Lavoie
according to the indications of Whittingham and Wood [36]. After the appearance of the vaginal plug indicating copulation [29], each female was housed individually in a sterile plastic cage with wood chip bedding (Northeastern Product Corporation, Warrensburg, N.Y.) and kept under a laminar flow hood (Canadian Cabinets, Moquin, Montrral, Qurbec) to prevent contamination from the external environment. The animals were fed sterile rodent chow (Charles River diet RMH 4020) and sterile tap water ad libitum. Each experiment included five pregnant mice and five nonpregnant mice as controls. The experiment was repeated three times with three different deliveries of mice to assess the reproducibility of the observations. To further assess the influence of the lactation period, a fourth group of six 12-day pregnant mice and a control group of five nonpregnant female mice of the same age were obtained from Charles River.
Bacterial Sampling To reduce variability, the food was removed 1 hour before sampling. The oral cavity was thoroughly sampled using Calgiswab® type 4 swabs (Spectrum Laboratories Inc., Los Angeles, Calif.) as previously described [28]. Appropriate dilutions (usually 10 < and 10 -2) in phosphate-buffered saline (10 rnM, pH 7.0 with 0.1 mM NaC1) were plated on trypticase soy agar (BBL®, Becton Dickinson, Cockeysville, Md.) enriched with 0.3% yeast extract (BBL) (TSAYE plates). The plates were incubated for 2 days at 37°C under anaerobic conditions (80% N2: 10% H2: 10% CO2) in a Coy Glove Box (Coy Manufacturing, Ann Arbor, Mich.). These conditions were previously shown to recover more than 99% of the bacterial species present [33]. The oral cavities were sampled every 4 days for 44 days, beginning 4 days before mating and ending 4 days after the offspring were weaned. Preliminary experiments indicated that although six consecutive samplings reduced the total number of bacteria recovered, it did not affect the proportions of the bacterial species present.
Distribution of the Bacterial Species The distributions of the bacterial species isolated were established as the percentage of the total cultivable biota using the colony-immunoblot method, as previously described [27, 28]. The isolates that were not detected with our specific antisera, raised against the predominent species isolated from the oral cavity of BALB/c mice [33], were purified by restreaking twice on TSAYE plates and identified using the appropriate AP! galleries (API Laboratory Products, St. Laurent, Qurbec).
Statistical Analyses For every sampling day, the distribution of the oral bacterial species in the pregnant group was compared to the control nonpregnant group using the minimal percentage of similarity and the Lambda of Good [7, 8, 14, 21, 23]. These statistical analyses permit taking into account the large fluctuations often observed in indigenous bacterial populations. The limit of acceptability level was fixed at P < 0.05.
Results T h e e v o l u t i o n o f the r e s i d e n t oral b a c t e r i a l p o p u l a t i o n s f r o m the three e x p e r i m e n t s for m i c e m a t e d at our facilities s h o w e d s i m i l a r trends. T h e results o f t h e s e e x p e r i -
Mouse Oral MicrobiotaDuring Pregnancy and Lactation
221
ments are presented in Tables 1-3. The results from the experiment in which pregnant mice were obtained from Charles River are presented in Table 4. Although only the proportions of the major species detected are presented in the tables, all species observed were used in the calculation of the minimal percentage of similarity and the Lambda of Good. In all four experiments, the only statistically significant differences in the distribution of the oral bacterial populations between the pregnant group and the control group were observed during the lactation period (8-16 days after parturition). Lactobacillus murinus usually predominated in the total cultivable oral bacterial biota of BALB/c mice, but between 8 and 16 days after parturition, another species became predominant in the mother's mouth. It was S. cohnii in two experiments (Tables 2 and 4) and S. intermedius (Table 3), and S. auricularis, respectively, in the two others (Table 1).
Discussion
In the controlled environment in our mouse model, the gestation period did not influence the resident oral bacterial populations. These results differ from those obtained in humans where pregnancy gingivitis seems to be induced by variations in physiological factors [9, 10, 25]. It should be noted, however, that gingivitis is generally attributed to the proliferation of strict anaerobes in the human oral bacterial biota [20] and that the mouse does not harbor such anaerobic species [31]. Furthermore, although serum progesterone levels are similar in humans and mice during the gestation period [1, 3, 4, 21, 35], estradiol levels are 500 times higher in humans [1, 3, 4]. Thus higher level could perhaps have an influence on the levels of certain bacterial species. Although the salivary concentrations of different hormones have been correlated with their serum levels in humans [2, 5, 13, 18, 19, 21], to our knowledge, salivary hormonal levels have never been measured in mice. In humans, gingivitis symptoms and the increase in P. intermedia (associated with pregnancy gingivitis) are observed between the third and the sixth month of pregnancy [12]. In the mouse, the whole pregnancy period lasts only 20 days. It is thus possible that no effect was observed, because the bacteria were not exposed to the hormones for a long enough period of time. During the lactation period in each trial, we systematically observed a statistically significant modification of the distribution of the resident oral bacterial populations between days 8 and 16 postpartum (Tables 1--4). This period corresponds to the maximal lactogenic period in the mouse [36], where serum prolactin is at its highest concentration [3]. We do not know, however, if prolactin and the other hormones associated with lactogenesis such as adrenocorticotropic hormone (ACTH) and growth hormones [34] are found in the saliva or whether they would influence the chemical and/or microbiological salivary composition. Mouse grooming habits could also explain the observed oral bacterial population shifts in our experiments. Mothers frequently lick the pups after parturition with the possible consequence of an increase in the detection of the bacterial skin populations in the mother's mouth. It is well-known that staphylococci, which increased in our experiments, are a major component of the skin flora [11]. We
C. Coulombe, M.C. Lavoie
222
+1 +1 +i
+i +1 +1 +1 o ~
+l +1 +1 +1
+1 +i +1 +1 +1 +1
~9 0"~ ("
8
+1 ~1 +1 +1 ~
-8
+l +i +1 +1 o
o
8
+~ +~ +~ +~
©
,D
MICROBIAL ECOLOGY © 1995Springer-VerlagNew York
Inc.
Evolution of Resident Oral Bacterial Biota in BALB/c Mice During Pregnancy and Lactation C. Coulombe, M.C. Lavoie Groupe de Recherche en Ecologic Buccale, Universit6 Laval, Qu6bec, Canada, G1K 7P4 Received: 23 August 1994; Revised: 22 December 1994
Abstract. To assess the influence of pregnancy and lactation on the oral microbial ecology of BALB/c mice, we followed the distribution of the predominant oral bacteria of four groups of these mice during these two periods. Compared with nonpregnant control female mice of the same age maintained under the same conditions, the distribution of the resident oral bacterial species differed significantly only during the lactation period (8-16 days after parturition). This difference could possibly be attributed to hormonal influences and/ or grooming habits. Introduction In humans, several factors of the oral ecosystem have been shown to vary during pregnancy: hormonal levels in gingival crevicular fluid [9,10] and gingival tissues [16, 31], the proportion of Prevotella intermedia (formerly Bacteroides intermedius) in subgingival plaque samples [12, 22], immunosuppression [17, 24, 26], and the proportions of Streptococcus mutans [15]. To assess the possible influences of the physiological variations associated with the periods of pregnancy and lactation on the oral biota, we followed the distribution of the resident oral bacteria in our mouse model. This model possesses a restricted oral microbiota, which facilitates the study of its variation [33]. Our present results indicate that the gestation period does not significantly affect the oral bacterial populations of the BALB/c mice but that a significant effect is observed during the lactation period.
Materials and Methods Mice BALB/c AnNcrl/BR (BALB/c) mice, 8-10 weeks old, were obtained from Charles River Canada (St. Constant, Qu6bec, Canada). After a rest period of 1 week, the mice were mated at our facilities
Correspondence to: M.C. Lavoie, Group de Recherche en l~cologie Buccale, Universit6 Laval, Qu6bec, Canada, G1K 7P4.
220
C. Coulombe, M.C. Lavoie
according to the indications of Whittingham and Wood [36]. After the appearance of the vaginal plug indicating copulation [29], each female was housed individually in a sterile plastic cage with wood chip bedding (Northeastern Product Corporation, Warrensburg, N.Y.) and kept under a laminar flow hood (Canadian Cabinets, Moquin, Montrral, Qurbec) to prevent contamination from the external environment. The animals were fed sterile rodent chow (Charles River diet RMH 4020) and sterile tap water ad libitum. Each experiment included five pregnant mice and five nonpregnant mice as controls. The experiment was repeated three times with three different deliveries of mice to assess the reproducibility of the observations. To further assess the influence of the lactation period, a fourth group of six 12-day pregnant mice and a control group of five nonpregnant female mice of the same age were obtained from Charles River.
Bacterial Sampling To reduce variability, the food was removed 1 hour before sampling. The oral cavity was thoroughly sampled using Calgiswab® type 4 swabs (Spectrum Laboratories Inc., Los Angeles, Calif.) as previously described [28]. Appropriate dilutions (usually 10 < and 10 -2) in phosphate-buffered saline (10 rnM, pH 7.0 with 0.1 mM NaC1) were plated on trypticase soy agar (BBL®, Becton Dickinson, Cockeysville, Md.) enriched with 0.3% yeast extract (BBL) (TSAYE plates). The plates were incubated for 2 days at 37°C under anaerobic conditions (80% N2: 10% H2: 10% CO2) in a Coy Glove Box (Coy Manufacturing, Ann Arbor, Mich.). These conditions were previously shown to recover more than 99% of the bacterial species present [33]. The oral cavities were sampled every 4 days for 44 days, beginning 4 days before mating and ending 4 days after the offspring were weaned. Preliminary experiments indicated that although six consecutive samplings reduced the total number of bacteria recovered, it did not affect the proportions of the bacterial species present.
Distribution of the Bacterial Species The distributions of the bacterial species isolated were established as the percentage of the total cultivable biota using the colony-immunoblot method, as previously described [27, 28]. The isolates that were not detected with our specific antisera, raised against the predominent species isolated from the oral cavity of BALB/c mice [33], were purified by restreaking twice on TSAYE plates and identified using the appropriate AP! galleries (API Laboratory Products, St. Laurent, Qurbec).
Statistical Analyses For every sampling day, the distribution of the oral bacterial species in the pregnant group was compared to the control nonpregnant group using the minimal percentage of similarity and the Lambda of Good [7, 8, 14, 21, 23]. These statistical analyses permit taking into account the large fluctuations often observed in indigenous bacterial populations. The limit of acceptability level was fixed at P < 0.05.
Results T h e e v o l u t i o n o f the r e s i d e n t oral b a c t e r i a l p o p u l a t i o n s f r o m the three e x p e r i m e n t s for m i c e m a t e d at our facilities s h o w e d s i m i l a r trends. T h e results o f t h e s e e x p e r i -
Mouse Oral MicrobiotaDuring Pregnancy and Lactation
221
ments are presented in Tables 1-3. The results from the experiment in which pregnant mice were obtained from Charles River are presented in Table 4. Although only the proportions of the major species detected are presented in the tables, all species observed were used in the calculation of the minimal percentage of similarity and the Lambda of Good. In all four experiments, the only statistically significant differences in the distribution of the oral bacterial populations between the pregnant group and the control group were observed during the lactation period (8-16 days after parturition). Lactobacillus murinus usually predominated in the total cultivable oral bacterial biota of BALB/c mice, but between 8 and 16 days after parturition, another species became predominant in the mother's mouth. It was S. cohnii in two experiments (Tables 2 and 4) and S. intermedius (Table 3), and S. auricularis, respectively, in the two others (Table 1).
Discussion
In the controlled environment in our mouse model, the gestation period did not influence the resident oral bacterial populations. These results differ from those obtained in humans where pregnancy gingivitis seems to be induced by variations in physiological factors [9, 10, 25]. It should be noted, however, that gingivitis is generally attributed to the proliferation of strict anaerobes in the human oral bacterial biota [20] and that the mouse does not harbor such anaerobic species [31]. Furthermore, although serum progesterone levels are similar in humans and mice during the gestation period [1, 3, 4, 21, 35], estradiol levels are 500 times higher in humans [1, 3, 4]. Thus higher level could perhaps have an influence on the levels of certain bacterial species. Although the salivary concentrations of different hormones have been correlated with their serum levels in humans [2, 5, 13, 18, 19, 21], to our knowledge, salivary hormonal levels have never been measured in mice. In humans, gingivitis symptoms and the increase in P. intermedia (associated with pregnancy gingivitis) are observed between the third and the sixth month of pregnancy [12]. In the mouse, the whole pregnancy period lasts only 20 days. It is thus possible that no effect was observed, because the bacteria were not exposed to the hormones for a long enough period of time. During the lactation period in each trial, we systematically observed a statistically significant modification of the distribution of the resident oral bacterial populations between days 8 and 16 postpartum (Tables 1--4). This period corresponds to the maximal lactogenic period in the mouse [36], where serum prolactin is at its highest concentration [3]. We do not know, however, if prolactin and the other hormones associated with lactogenesis such as adrenocorticotropic hormone (ACTH) and growth hormones [34] are found in the saliva or whether they would influence the chemical and/or microbiological salivary composition. Mouse grooming habits could also explain the observed oral bacterial population shifts in our experiments. Mothers frequently lick the pups after parturition with the possible consequence of an increase in the detection of the bacterial skin populations in the mother's mouth. It is well-known that staphylococci, which increased in our experiments, are a major component of the skin flora [11]. We
C. Coulombe, M.C. Lavoie
222
+1 +1 +i
+i +1 +1 +1 o ~
+l +1 +1 +1
+1 +i +1 +1 +1 +1
~9 0"~ ("
8
+1 ~1 +1 +1 ~
-8
+l +i +1 +1 o
o
8
+~ +~ +~ +~
©
,D
