Cnr2 Deficiency Confers Resistance to Inflammation-Induced Preterm Birth in Mice
Abstract
Infection-induced inflammation, frequently associated with increased production of proinflammatory cytokines, is considered a significant contributor to preterm birth. A G protein-coupled cannabinoid receptor 2 (CB2), encoded by Cnr2, is expressed in various immune cells and was shown to modulate immune responses. We show here that Cnr2, but not Cnr1, deficient mice are resistant to lipopolysaccharide (LPS)-driven preterm birth and suppression of serum progesterone levels. After LPS challenge, Cnr2−/− mice exhibited increased serum levels of IL-10 with decreased IL-6 levels. These changes were associated with reduced LPS-induced Ptgs2 expression at the maternal-conceptus interface on day 16 of pregnancy. LPS stimulation of Cnr2−/− dendritic cells in vitro resulted in increased IL-10 with reduced IL-6 production and correlated with increased cAMP accumulation. Collectively, our results suggest that increased IL-10 production occurring via augmented cAMP accumulation represents a potential mechanism for the resistance of Cnr2−/− mice to LPS-induced preterm birth. These results may have clinical relevance, because currently, there are limited options to prevent preterm birth.
In humans, preterm birth, defined as parturition occurring before 37 weeks of gestation, accounts for 12% of all births. Despite advances in obstetrics and neonatology, the rate of preterm birth in the United States has increased from 10.6% to 12.7% between 1990 and 2005 (1). Socioeconomic, environmental, medical, and genetic risk factors are all correlated with preterm birth (2). Infection and inflammation, however, are considered a major contributor to preterm birth (3); 40% of preterm births are attributable to intrauterine infection (3). There is a strong association between elevated circulating proinflammatory cytokine levels and preterm birth. In rhesus macaques, infusion of proinflammatory cytokines induces premature myometrial contractions leading to preterm birth (4, 5). Furthermore, pathogen-driven increases in proinflammatory cytokine production induces preterm birth in mice (6), whereas exogenous administration of IL-10, an antiinflammatory cytokine, attenuates uterine and placental inflammation, preventing preterm birth (7). This latter observation is consistent with the finding that IL-10−/− females are highly susceptible to inflammation-induced preterm birth (8).
Activation of Toll-like receptor (TLR) signaling by conserved microbial structures, or endogenous molecular structures generated or unmasked during tissue injury and inflammation, induces nuclear factor of kappa light polypeptide gene enhancer in B cells (NF-κB), MAPK, and interferon regulatory factor (IRF) signaling pathways, leading to proinflammatory and immunoregulatory cytokine production (9). In addition to immune cells, TLRs are expressed at the maternal-fetal interface by the uterus, placenta, amniotic membranes, and trophoblast cells (10). Diverse TLR ligands have been used to model inflammation-induced preterm birth: exposure to these ligands were shown to induce proinflammatory cytokine release in the uterus and fetal membrane, and recruit immune cells to the cervix, inducing preterm birth in mice and nonhuman primates (11). As might be expected, antagonism of the TLR that signals the presence of lipopolysaccharide (LPS) (TLR4) was shown to reduce LPS-driven preterm birth (12), and Tlr4−/− mice are resistant to LPS-induced preterm birth (13).
Natural cannabinoids and endocannabinoids signal through cannabinoid receptors CB1 and CB2. CB1, encoded by Cnr1, is expressed in the central nervous system and peripheral tissues, including heart, testis, liver, small intestine, and uterus. In contrast, CB2 (Cnr2) is expressed by a variety of immune cells (14). Cnr1and Cnr2 share over 40% sequence homology and are coupled to G proteins of the Gi/o and Gq families. Activation of these receptor subtypes leads to different, often cell-specific, biological effects. Signaling via CB receptors modulates multiple pathways, including inhibition of adenylyl cyclase (AC) and MAPK pathways that are also known to modulate cytokine production (15).
Although there is evidence that cannabinoid signaling plays an important role in modulating immune responses, the effects of endocannabinoids and their respective receptors on immune responses remain underdefined. In this context, augmented cannabinoid signaling suppresses the production of proinflammatory cytokine after LPS-driven TLR4 activation (16) but potentiates generation of antiinflammatory cytokines (17). In contrast, antagonism of cannabinoid signaling was reported to inhibit LPS-induced proinflammatory cytokine production (18). Furthermore, studies with Cnr2−/− mice in the model of cecal ligation and puncture-induced sepsis have also yielded conflicting results: Cnr2−/− mice had either decreased IL-6 production and sepsis-induced mortality (19) or increased IL-6 production and sepsis-induced mortality (20).
Considering that TLR-induced inflammation drives preterm birth and that signaling by CB2 modulates TLR-dependent immune responses, we hypothesized that cannabinoid/endocannabinoid signaling via CB2 plays a critical role in the timing of parturition in mice through regulation of TLR-driven cytokine production. To test this hypothesis, we studied LPS-induced incidence of preterm birth in Cnr1−/− and Cnr2−/− mice. We observed that Cnr2−/−, but not Cnr1−/−, females were resistant to LPS-induced preterm birth. Cnr2−/− females produced higher levels of IL-10 but less IL-6 in response to LPS. Furthermore, this modulation of cytokine production was associated with protection from LPS-driven decline in serum progesterone (P4) levels and prevention of increases in LPS-induced Ptgs2 expression at the maternal-conceptus interface. These results were associated with increased levels of intracellular cAMP levels in LPS-stimulated bone marrow-derived dendritic cells (BMDCs) from Cnr2−/− mice compared with wild-type (WT) littermates, suggesting increased accumulation of cAMP as a candidate mechanism for driving increased IL-10 production that protects against preterm birth caused by inflammatory insults.
Materials and Methods
Animals and treatments
WT, Cnr1−/−, and Cnr2−/− mice and their littermates were housed in the animal care facility at the Cincinnati Children’s Hospital Medical Center according to National Institutes of Health and institutional guidelines for laboratory animals. Cnr1−/− and Cnr2−/− mice on C57BL/6J genetic background were generated as described (21, 22). All experiments used WT littermates as controls. All protocols of the present study were approved by the Cincinnati Children’s Hospital Research Foundation Institutional Animal Care and Use Committee.
Female mice were mated with fertile males to induce pregnancy (vaginal plug, day 1 of pregnancy). Parturition events were monitored from day 17 through day 21 by observing mice daily in the morning and evening. Preterm birth was defined as birth occurring earlier than day 19. All inflammatory agents were given ip in the morning of day 16. LPS (Escherichia coli 0111:B4, 437627; Calbiochem), ultrapure LPS (E. coli 0111:B4, tlrl-3pelps; Invivogen), and ultrapure TLR2 ligand Pam3Cys (Pam3CSK4, tlrl-pms; Invivogen) were all dissolved in saline. Control groups received vehicle alone.
In situ hybridization
In situ hybridization was performed as previously described (23). In brief, frozen sections (12 μm) were mounted onto poly-L-lysine-coated slides and fixed in cold 4% paraformaldehyde in PBS. The sections were prehybridized and hybridized at 45°C for 4 hours in 50% (vol/vol) formamide hybridization buffer containing 35S-labeled antisense RNA probes. The probe for Ptgs2 was synthesized as previously described (24). The probe for Il6, a gift from Frank Lee (DNAX Research Institute of Molecular and Cellular Biology, Inc) was previously described (25, 26). Ribonuclease A-resistant hybrids were detected by autoradiography. Sections were poststained with hematoxylin and eosin.
Measurement of estradiol-17β (E2) and progesterone (P4) levels
Serum levels of E2 and P4 were measured by enzyme immunoassay kits (Cayman).
In vivo cytokine capture assay (IVCCA)
IVCCA was used to measure systemic cytokine production as previous described (27). Briefly, biotinylated capture antibodies (anti-IL-10, IL-6, TNF-α, and Interferon gamma (IFNγ); eBioscience) were injected via tail vein 3 hours before challenge with TLR ligands, ultrapure LPS (25 μg), or Pam3Cys (100 μg), and sera were collected 24 hours later.
BMDCs ELISA
BMDCs were generated from the bone marrow precursors from femurs and tibias of WT and Cnr2−/− mice as previously described by us (28). BMDCs were stimulated for 24 hours, and cell-free supernatants were collected. Cytokine levels were determined by ELISA (R&D Systems).
Reverse transcription-polymerase chain reaction
RNAs from Cnr2+/+ and Cnr2−/− BMDC samples were analyzed as described previously (29, 30). In brief, total RNA was extracted with TRIzol (Invitrogen) according to the manufacturer’s protocol. After deoxyribonuclease treatment (Ambion), 1 μg of total RNA was reverse transcribed with Superscript II (Invitrogen). RT-PCR was performed using primers 5′-ACAAGGCTCCACAAGACCCT-3′ (sense) and 5′- CCGTTGGTCACTTCTGTCT-3′ (antisense) for Cnr2.
cAMP assay
Intracellular cAMP assays were performed using an immunoassay kit from GE Healthcare (RN225) according to the manufacturer’s recommendations.
Statistical analysis
Comparison of means was performed by the Student’s t test or χ2test. P < .5 was considered significant. Values are mean ± SEM.
Results
Cnr2−/− females are resistant to LPS-induced preterm birth
CB2-mediated signaling is important in modulating immune responses, including those driven by TLR signaling. We challenged WT and Cnr2−/− pregnant females on day 16 of pregnancy with a standard commercial LPS preparation (25 μg/dose; Calbiochem) to examine the incidence of preterm birth. Although 80% of WT females had preterm birth on day 17, only 42% of Cnr2−/− females gave preterm birth after LPS challenge (Figure 1A). Notably, TLR4−/− pregnant mice are fully protected from LPS-driven preterm birth (13). Because LPS is isolated from bacteria, the standard commercial LPS often contains other bacterial components, such as lipopeptides, and therefore stimulates both TLR4 and TLR2. To exclude confounding effects of other TLRs except TLR4, we challenged WT and Cnr2−/− pregnant mice on day 16 of pregnancy with ultrapure LPS (37 μg/dose; Invivogen) known to specifically activate TLR4. We observed that 89% of WT mice had preterm delivery, whereas only about 44% of Cnr2−/−females showed preterm birth (Figure 1B and Supplemental Table 1). Females of both genotypes injected with sterile saline on day 16 showed normal term delivery occurring between day 19 afternoons and day 20 mornings with the birth of live pups (Supplemental Figure 1). These results suggest that Cnr2−/− females are resistant to LPS-driven preterm birth and the effect of LPS are primarily mediated by TRL4.
In this context, previous reports suggested a potential role for CB1 in modulating cytokine production. In rats and mice, modulation of LPS-driven cytokine production by CB1 receptor agonists, WIN55,212–2 or HU-210, was antagonized by SR141716A, a selective CB1 receptor antagonist (17). Thus, to determine the role of CB1 in LPS-driven preterm birth, we injected Cnr1−/− and WT littermate pregnant females with ultrapure LPS (37 μg/dose) on day 16 of pregnancy. Although approximately 80% of WT littermate females had preterm delivery, all Cnr1−/− females exhibited preterm birth (Figure 1C), suggesting that Cnr1−/− mothers are more prone to TLR4-driven preterm delivery. Taken together, these results provide evidence that Cnr2−/− females are more resistant to LPS-induced preterm birth when compared with their WT littermates or Cnr1−/− females.
Serum levels of IL-10 are elevated in Cnr2−/− females after LPS challenge compared with WT littermate controls
Inflammatory responses arising from infection/inflammation are thought to be a leading cause of preterm birth (3, 11). Increased levels of proinflammatory cytokines lead to uterine activation and preterm labor, whereas production of antiinflammatory cytokines causes uterine relaxation during gestation (31). Therefore, we evaluated the possible role of CB2 in regulating the production of cytokines after TLR4 stimulation. Specifically, we quantified serum IL-10, IL-6, TNF-α, and IFNγ levels in littermate WT and Cnr2−/− females 12 hours after challenge with ultrapure LPS on day 16 of pregnancy. The results showed that IL-6 and TNF-α levels were significantly lower in Cnr2−/− mice compared with WT littermates. In contrast, serum IL-10 levels were significantly elevated in Cnr2−/− mice as compared with WT controls (Figure 2A). These results provide evidence that CB2 plays an important role in balancing the levels of proinflammatory (IL-6) and antiinflammatory (IL-10) cytokines and affirm the importance of CB2 in modulating immune responses in the context of preterm birth. Because immune responses are altered in pregnancy (32), we also examined whether CB2-mediated modulation of LPS-driven cytokine production is specific to pregnancy. We observed a similar pattern of changes in serum IL-6 and IL-10 levels in both nonpregnant and pregnant Cnr2−/− females, although no significant changes were noted for levels of TNF-α and IFNγ in nonpregnant littermates (Figure 2B). Our findings suggest that modulation of TLR4 signaling via CB2 is not limited to pregnancy.
Cnr2−/− females are resistant to LPS-induced drop in P4 levels and show restricted expression of Il6 and Ptgs2 at the maternal-fetal interface
Previous reports suggested that bacteria-induced placental inflammation is associated with preterm birth (13). To examine whether the maternal-conceptus interface in Cnr2−/− pregnant females displays signs of inflammation, we examined Il6 mRNA localization by in situ hybridization on day 16 of pregnancy 12 hours after LPS challenge. Il6 transcripts were primarily localized in endothelial cells of blood vessels at the mesometrial side in WT females; the signals were much weaker in endothelial cells at similar regions in Cnr2−/− females (Figure 3A). No detectable signals were noted in the decidua or placenta. These results suggest that although LPS and/or inflammatory cytokines reach implantation sites and activate the endothelial cells, the response is much weaker in Cnr2−/− females.
Prostaglandins (PGs) play a major role in parturition. PG biosynthesis is catalyzed by cyclooxygenase enzymes, which are encoded by Ptgs genes (33). There are 2 isoforms of Ptgs genes: Ptgs1 is constitutively expressed in many tissues, whereas Ptgs2 is induced by many growth factors, cytokines, and various inflammatory stimuli. PGs are up-regulated in the uterus and decidua at the time of parturition in animals (34). In fact, increased local generation of PGs, especially PGF2α, are attributed to increased Ptgs2 expression before parturition (35). Expression of Ptgs2 and PGF synthase is also up-regulated in the uterus in preterm birth, with local increases in PGF2α levels resulting from infection/inflammation or genetic mutation (35). Therefore, we examined the expression of Ptgs2 on day 16 by in situ hybridization in WT and Cnr2−/− implantation sites 12 hours after LPS challenge. We found that Ptgs2 expression was induced at the maternal-conceptus interface, primarily in the decidua in WT females, whereas Ptgs2 expression was greatly compromised in Cnr2−/− females (Figure 3A). These data suggest that exacerbated IL-10 production in Cnr2−/− mice after LPS challenge also decreases inflammatory responses locally at the implantation sites. The observation corroborates previous findings that IL-10 plays a role in regulating Ptgs2 expression after LPS challenge in mice (36).
During pregnancy, the myometrium is relatively quiescent due to increased levels of circulating P4. It is known that premature decline in P4 levels can predispose pregnant females to preterm delivery (37). LPS-induced inflammation can cause P4 levels to decline within hours after exposure (38). To examine whether the protection from LPS-induced preterm birth in Cnr2−/− mice correlates with ovarian hormone production, we quantified serum levels of E2 and P4. Although P4 levels were comparable in WT and Cnr2−/− females after saline (vehicle) injections, LPS challenge significantly decreased P4 levels in WT but not in Cnr2−/− pregnant females (Figure 3B). In contrast, E2 levels in WT and Cnr2−/− females were comparable after LPS injection (Figure 3C). The results suggest that the ovaries are a target for LPS-induced inflammatory insults in WT females, but Cnr2−/− females can maintain serum P4 levels under such conditions.
Cnr2−/− females exhibit higher levels of serum IL-10 in response to TLR2 and TLR4 ligands compared with WT littermate controls
Infection-driven inflammation during pregnancy is complex and can involve activation of multiple TLRs. There are reports that diverse TLR ligands can drive preterm birth in mice (39). To determine whether CB2-mediated modulation of cytokine production is TLR4 specific, we challenged littermate WT and Cnr2−/− nonpregnant mice with a TLR2 ligand, Pam3Cys. Although no differences in IL-6, TNF-α, and IFNγ levels were observed between the 2 genotypes, Cnr2−/− females had significantly elevated levels of IL-10 compared with WT counterparts (Figure 4). In the same vein, an ip injection of Pam3Cys even at 500 μg on day 16 of pregnancy failed to induce preterm birth in WT females; all 3 females examined delivered full complement of pups (9.0 ± 0.7 pups/litter) on day 20.
Cnr2−/− BMDCs show increased accumulation of cAMP levels after LPS challenge in vitro
Endocannabinoids, arachidonoylethanolamide (AEA) and 2-arachidonylglycerol (2-AG), and their receptor CB2 are present in a variety of immune cells, including dendritic cells, macrophages, microglia, and lymphocytes (14), making them potential targets of endocannabinoid signaling. In the same vein, a variety of immune and nonimmune cells produces IL-10 after LPS stimulation (40). These results suggest a possible link between endocannabinoid signaling via CB2 and IL-10 secretion. In this context, we also compared the overall infiltration of hematopoietic cells into Cnr2−/− and WT implantation sites 12 hours after saline or ultrapure LPS (37 μg/dose) injection on day 16 of pregnancy. Immunofluorescence of CD45, a marker of hematopoietic cells, did not show much variation in the number of CD45-positive cells in saline-injected Cnr2+/+ and Cnr2−/− decidua basalis. However, LPS challenge increased the presence of these cells in the blood vessel and myometrium in both genotypes without much difference in their populations between the 2 groups (Supplemental Figure 2).
Dendritic cells are thought to play a crucial role in mediating a balance between immunity and tolerance during pregnancy (41). BMDCs express a wide range of TLRs (42), making them a suitable in vitro model to further investigate CB2’s role in regulating immune responses. We used BMDCs to assess whether in vivo findings of CB2-mediated modulation of IL-10 production could be corroborated in vitro. RT-PCR confirmed the expression of Cnr2 in WT but not in Cnr2−/− BMDCs (Figure 5B). We found that Cnr2−/− BMDCs produced higher levels of IL-10 with reduced levels of IL-6 compared with WT controls in response to LPS (Figure 5A). Similar results were observed when WT and Cnr2−/−BMDCs were stimulated with Pam3Cys. These results are consistent with our in vivo observations and suggest that BMDCs can be used for further mechanistic studies. Although peritoneal macrophages did not show similar changes under similar conditions (Figure 5C), we still cannot exclude the possibility that CB2 influences IL-10 production in other immune cells after LPS challenge.
CB2-mediated signals are primarily transduced via Gi proteins (15). Activation of CB2 decreases cellular accumulation of cAMP by inhibiting AC activity. The AC-cAMP signaling pathway plays an important role in immune cell function. Specifically, cholera toxin, an activator of AC, inhibits IL-12 and TNF-α production in bone marrow macrophages by increasing IL-10 production (43). Therefore, we speculated that elevated IL-10 production in the absence of CB2 is a consequence of increased intracellular cAMP accumulation. We compared the levels of cAMP in WT and Cnr2−/− BMDC and found that Cnr2−/− BMDCs had elevated levels of cAMP after LPS challenge compared with WT BMDC. These results suggest that increased accumulation of cAMP represents a likely mechanism of CB2-dependent modulation of IL-10 production (Figure 6).
Discussion
The highlights of the present investigation are that deletion of Cnr2confers protection against LPS-induced preterm birth and that Cnr2deficiency drives increased IL-10 production. The failure of LPS to diminish serum P4 levels in Cnr2−/− females suggests that ovaries deficient in CB2 are protected from adverse effects of inflammatory insults, which in turn prevent local inflammation and preterm birth. Alternatively, TLR4-driven increased serum levels of IL-10 in Cnr2−/− females offer protection to the ovary to sustain serum P4 levels after exposure to LPS-induced inflammatory insults. Normally, inflammatory insults lead to the demise of ovarian corpora lutea (luteolysis), resulting in drop of serum P4levels, terminating pregnancy (44). Notably, P4 is an absolute requirement for pregnancy maintenance until the approach of parturition. Our results with Cnr2−/− BMDCs suggest that TLR4-driven increased IL-10 production in the absence of CB2 could, at least in part, be dependent on increased accumulation of intracellular levels of cAMP.
It is interesting to note that although both CB1 and CB2 were shown to modulate LPS-driven cytokine release (45, 46), signaling mediated by CB2 alone appears to play a critical role in LPS-driven preterm birth. The differential phenotypes observed in Cnr1−/− and Cnr2−/− mice could be due to distinct biological functions executed by CB1 and CB2 in response to LPS in a context-dependent manner. It is also possible that these disparate effects are due to differential cell type-specific expression of CB1 and CB2. For example, CB1 is primarily expressed in the central nervous system, whereas the expression of CB2 is more predominant in immune cells. Our finding that CB2 deficiency confers resistance to preterm birth is intriguing, because this observation is different from the current notion that signaling via CB2 is immunosuppressive (reviewed in Ref. 47). Therefore, the current observation of protection against LPS-induced preterm birth in Cnr2−/− females would suggest that the role of CB2 is context and system dependent.
In the current study, the cytokine levels in both pregnant and nonpregnant Cnr2−/− and control females were measured, but the levels between pregnant and nonpregnant Cnr2+/+ and Cnr2−/−females cannot directly be compared due to different collection time (24 h in nonpregnant females vs 12 h in pregnant females) and doses of LPS used. Nonetheless, these results showed that serum levels of IL-10 in Cnr2−/− females were higher than those in Cnr2+/+ females in response to LPS challenge irrespective of their pregnancy status.
Our observation of low to undetectable expression of Il6 in the decidua, placenta, or myometrium after LPS challenge suggests that either these tissues had limited access to LPS on day 16 of pregnancy or inability of these tissues to stimulate Il6 induction after LPS exposure at this stage of pregnancy. However, it is possible that cytokines are generated at distant sites and influence pregnancy by acting at the maternal-fetal interface; adverse or beneficial effects on pregnancy outcome will depend on the balance between proinflammatory and antiinflammatory cytokines. This assumption is consistent with previous findings of decline in serum P4 levels and pregnancy termination in WT females after LPS exposure that drives primarily proinflammatory cytokines (38). In this context, our results showing resistance of Cnr2−/−females to LPS-induced preterm birth point toward an important concept that CB2 signaling is a contributing factor in regulating IL-10 levels and ovarian function after LPS challenge. Indeed, IL-10 was shown to stimulate P4 production in human luteal cells in vitro (48). In contrast, there is also evidence that inflammation/infection induces proinflammatory cytokines, which causes ovarian luteolysis and drop in serum P4 levels. The mechanism by which antiinflammatory and proinflammatory cytokines inversely influence ovarian function requires further investigation.
In this study, culture of BMDCs and peritoneal macrophages was used to study the role of CB2 in regulating cytokine production after LPS challenge. Both cell types express a wide range of TLR receptors, making them widely used model systems to elucidate mechanistic information of immune cell function in responses to LPS. Dendritic cells play a crucial role in mediating immune tolerance during pregnancy (41). In the context of our study, AEA and 2-AG and their receptor CB2 are present in dendritic and macrophage cells (14). Although we used dendritic cells as a model to show Cnr2−/− BMDCs produce more IL-10 after LPS challenge, we cannot exclude the possibility that other immune cells also play a role in the tolerance to LPS challenge in Cnr2−/− mice.
Our observation of increased accumulation of cAMP in Cnr2−/−BMDCs after LPS exposure suggests that cAMP not only plays an important role in immune cell function but may also influence muscle contractile activity by releasing factors from immune cells resident in the myometrium. In fact, there is evidence that cAMP regulates uterine muscle activity during pregnancy. For example, higher intracellular cAMP levels cause muscle relaxation (49, 50), and cAMP has been shown to mediate the role of P4 in maintaining myometrial quiescence (51). Thus, appropriate cAMP levels are critical for maintaining pregnancy to full term.
Our observation of much lower levels of proinflammatory cytokines in both Cnr2−/− and WT females after Pam3Cys challenge compared with TLR4-driven cytokine levels may explain the inability of Pam3Cys to induce preterm birth in WT females. There is evidence that intrauterine injection of a TLR2 ligand lipoteichoic acid (500 μg/mouse) and peptidoglycan (750 μg/mouse), but contaminated with TLR4 ligands, can induce preterm birth in mice (39). In this respect, our observation of Pam3Cys’s inability to induce preterm birth in WT mice could be due to different routes of administration used in these 2 studies. Collectively, these results suggest that TLR2 activation is less potent than TLR4-driven signaling in inducing preterm birth. Interestingly, TLR2-mediated IL-10 production is similar to that mediated by TLR4 in Cnr2−/− females. These results suggest that CB2 plays a crucial role in regulating proinflammatory and antiinflammatory cytokine balance during pregnancy.
The findings of this study appear to have high clinical relevance. The endocannabinoid system is evolutionarily conserved from invertebrates to vertebrates, including humans (52). The major components of the endocannabinoid system are present in the human pregnant endometrium and myometrium, making these tissues potential targets for AEA and 2-AG (53). Therefore, the results of our studies in mice could be applicable to humans. In addition, infection/inflammation are considered a major contributor to preterm birth in humans (3). Therefore, CB2’s role in regulating infection/inflammation could provide clues to design approaches for preventing preterm birth. However, parturition in rodent models after exposure to infection/inflammation is induced by ovarian luteolysis and subsequent drop in serum P4 output, which is believed not to occur in human parturition (54). In addition, systemic administration of LPS induces not only uterine but systemic inflammation, which does not always simulate human pathophysiology of labor. Nonetheless, genetic mouse models are important to obtain more mechanistic information regarding preterm birth, which is not possible to derive from human studies. In summary, this study represents a first report on the role of CB2 in regulating inflammation-driven preterm birth and opens up a new avenue for further exploration and development of novel therapeutic and/or preventive approaches to preterm birth.
Acknowledgments
We thank Serenity Curtis for editing the manuscript.
Present address for C.L.K.: The Bill & Melinda Gates Foundation, Seattle, WA.
This work was supported in part by Mar of Dimes Grants 21-FY12–127 and 22-FY14-470 and The National Institute on Drug Abuse/The National Institutes of Health Grant DA006668 (to S.K.D.). X.S. was supported by a Lalor Foundation postdoctoral fellowship.
Disclosure Summary: The authors have nothing to disclose.
Footnotes
- AC
- adenylyl cyclase
- AEA
- arachidonoylethanolamide
- 2-AG
- 2-arachidonylglycerol
- BMDC
- bone marrow-derived dendritic cell
- CB
- cannabinoid receptor
- E2
- estradiol-17β
- IFNγ
- Interferon gamma
- IVCCA
- in vivo cytokine capture assay
- LPS
- lipopolysaccharide
- P4
- progesterone
- PG
- prostaglandin
- TLR
- Toll-like receptor
- WT
- wild type.
References
