Maternal adaptations to pregnancy are essential for reproductive success in mammals. These gestational-dependent adjustments are helpful to the development of embryos/fetuses in the reproductive tract. Our previous study indicated that the activities of sympathetic nerves might be favorable to fetal survival and development during early pregnancy through influencing on immune function and decidua formation of uterus in mice. After pregnant mice were treated with 6-hydroxydopamine (6-OHDA), the number of implanted embryos was reduced to 64.4% of that in the control group at embryonic day (E) 7. The CD4+/CD8+ T cell ratio of the uterus was decreased, and the uterine glands and blood vessels were underdeveloped (Dong et al.,2007b). The decreased litter size and the increased number of fetuses resorbed or stillborn have also been reported in the 6-OHDA-treated rat (MacDonald et al.,1981). The monoamine, for example, norepinephrine (NE), might regulate several steps of the reproductive processes such as embryo implantation and decidua formation (Bottalico et al.,2003). All these data suggest that sympathetic nerve activity plays a critical role on the regulation to the immune function and endometrium reconstruction, which are important for the implantation and development of the fetus in early pregnancy.
The spleen is an important organ for blood filtration and immunologic function, and it undergoes pregnancy-dependent adaptations in its structure, size, and functions during pregnancy (Fowler and Nash,1968; Maroni and de Sousa,1973; Mattsson et al.,1979; Sasaki et al.,1981; de Rijk et al.,2002; Norton et al.,2009). Pregnancy-dependent increase in splenic growth has been found in mice and rats, accompanied by an expansion in splenic red pulp. Maternal spleen expression of erythroid-associated genes (erythroid Krüppel-like factor, erythroid 5-aminolevulinate synthase-2, and β-major globin) has been shown to be influenced by pregnancy (Bustamante et al.,2008). Takabatake et al. (1997) demonstrated that administration of splenocytes derived from early pregnant mice facilitated murine embryo implantation. These observations showed that the maternal spleen appeared to adapt to the demands of pregnancy. The spleen has an extensive innervations of sympathetic neurons, and the splenic nerve contains approximately 98% sympathetic C-fibers (Klein et al.,1982). In nonpregnant female mice, the proliferation of both B and T lymphocytes decreased after chemical sympathectomy (Dong et al.,2007a). It is uncertain whether these changes in the process of pregnancy are dependent on the regulation of sympathetic nerve activity.
Moreover, early pregnancy is associated with a low oxygen (O2) environment, which can keep the developing fetus from the deleterious and teratogenic effects of oxygen-free radicals (OFRs). Extremely increased oxidative stress has been reported to be a potential risk factor for placental function (Myatt et al.,2004). Placental oxidative stress and a disrupted antioxidant system are involved in a variety of pregnancy complications such as preterm labor, fetal growth restriction, pre-eclampsia, and miscarriage (Burton and Jauniaux,2004; Myatt et al.,2004). Oxidative stress is caused by an imbalance between the production of reactive oxygen and the ability of biological system to readily detoxify the reactive intermediates or to easily repair the resulting damage. However, little work has been done to evaluate whether this process is correlated with the sympathetic nervous system activity.
Based on these data, this study was conducted to investigate the effects of sympathetic nerves on the proliferation of splenic lymphocytes, structure, oxidant status, and antioxidant function of the maternal spleen in early pregnancy by using 6-OHDA-induced chemical sympathectomy model of mice.
MATERIALS AND METHODS
Eighty female (25–32 g) and 20 male (35–40 g) Institute for Cancer Research (ICR) mice of 8 weeks of age were purchased from Vital River Laboratory Animal Technology (Beijing, China). After an adaptive period of 1 week, the females were injected intraperitoneally for 5 consecutive days with 100 mg/kg body weight (BW) of 6-OHDA (Sigma Chemical, St. Louis, MO) diluted in a sterile saline solution containing 0.01% L-ascorbic acid to establish a peripheral sympathectomy model of mice (Dong et al.,2007b). The control females were treated with vehicle (0.01 mL/g BW). After completion of treatments, females were paired with males of proven fertility following vaginal smears detection (Wright's staining) daily. On the next day, animals were confirmed for pregnancy by vaginal plug or vaginal smear. The presence of a plug in the vagina or sperm detected in the vaginal smear was designated as embryonic day (E) 1. All experimental protocols were carried out in accordance with the Guidelines for Animal Experimentation of China Agricultural University.
Tissue Harvesting and Processing
Fifty spleen samples were immediately removed from the abdominal cavity of the mice under Nembutal deep anesthesia (50 mg/g BW) at E1, E3, E5, E7, and E9. The spleen weight of each animal was measured. The spleen samples were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4, 4°C) for 48 hr. Paraffin sections (6-μm thickness) were made by using standard procedures and were stained with hematoxylin and eosin (H&E) or by proliferating cell nuclear antigen (PCNA) immunohistochemistry.
In H&E staining, photographs of spleen were taken with Olympus microscope (BX51, Japan) under 20× magnifications. Areas of 10 biggest splenic nodules (SNs) and periarterial lymphatic sheath (PLS) in each spleen position were measured in six cross-sections for each mouse by using Scion Image (Scion Image for Windows, Version Beta 4.0.2; Scion, Frederick, MD). The mean areas of SN and PLS were calculated.
In immunohistochemical staining, the sections were immunostained with primary antibodies for PCNA (Sigma, 1:2,000) overnight at 4°C. Slides were incubated with biotinylated horse anti-mouse IgG secondary antibodies (Sigma, 1:200) and streptavidin-horseradish peroxidase (Sigma, 1:200) for 2 h at 25°C, respectively. Immunoreactivity was visualized by incubating the sections in 0.01 M PBS containing 0.05% 3,3′-diaminobenzidine tetrahydrochloride (Sigma) and 0.003% hydrogen peroxide for 10 min. Nuclei were counterstained with hematoxylin. Control slides without primary antibody were examined in all cases. The number of positive nuclei per 100 total nuclei was counted in 25 random microscopic fields from five cross sections of the spleen of each animal at each sampling point.
Splenocyte Isolation and Preparation
Spleens were aseptically removed, minced into small pieces, and passed through a tissue sieve (200 mesh per 2.5 cm) to prepare single-cell suspension in 3 mL Hank's solution. The suspension was layered onto 3-mL Ficoll-Hypaque (density = 1.077; manufactured by Tianjin Blood Research Center, Tianjin, China) and separated by density-gradient centrifugation at 1,500g for 30 min. The precipitates were resuspended in 3 mL RPMI 1640 base media (Gibco BRL, Grand Island, NY) and centrifuged at 1,000g for 10 min for two times. Splenocytes were suspended in 2 mL RPMI 1640 complete media (Gibco BRL). The cell activity was determined by the trypan blue dye exclusion test, and then the cells were counted.
Splenocyte Proliferation Assay
Mitogen-induced splenocyte proliferation was selected as in vitro indices of immune function. Concanavalin A (ConA)- and lipopolysaccharide (LPS)-induced splenocyte proliferation was assayed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Mosmann,1983) with some modifications. Each cell suspension sample was assigned into ConA (Sigma) or LPS (Sigma) treatment and control wells, with three sequential triplicate wells individually. Briefly, 100 μL of the spleen cell suspension (2 × 106 cells per milliliter) was added into individual wells of 96-well microtiter plate (Costar 3599; Corning, Corning, NY). The wells of the plates with splenocyte samples were added with ConA, LPS and RPMI 1640 complete Medium to get ConA, LPS treatment and control wells. The final ConA, LPS concentration of the wells were 15 μg/mL, and each well was filled with RPMI 1640 complete Medium to final volume of 120 μL. Then plates were incubated at 37°C in a 5% CO2 incubator for 72 hr. Subsequently, 10 μL of the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) solution was added to each well to make a final concentration of 5 mg/mL (in 1/15 M PBS, pH 7.4). The plates were incubated at 37°C for another 4 hr. Following incubation, 100 μL of the 10% sodium dodecyl sulfate (Shanghai Chemical Factory, Shanghai, China) in 0.04 M HCl solution was added to lyse the cells and to solubilize the MTT crystals. The plates of each sample were read at 570 nm wavelength using an automated microplate reader (Bio-Rad, USA). The stimulation index was calculated for each sample as absorbance values for cells with mitogen divided by absorbance values for cells without mitogen.
Measurement of Antioxidant Status and Lipid Peroxidation
The endogenous antioxidant components, superoxide dismutase (SOD), glutathione peroxidase (GPx), and malondialdehyde (MDA), an end product of lipid peroxidation, were analyzed by kinetic and spectrophotometric methods. Spleen samples were homogenized with ice-cold PBS (pH 7.4) and centrifuged at 3,000g for 15 min, and the supernatant was used immediately for the assays of SOD, GPx, and MDA levels using available commercial kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). SOD assay was conducted on the basis of the ability to inhibit oxidation of oxyamine by the xanthine–xanthine oxidase system. GPx activities were detected by the consumption of glutathione, and MDA content was measured by the thiobarbituric acid colorimetric method. Values were calculated by using optical density (550 nm for SOD, 412 nm for GPx, and 532 nm for MDA) according to the formula described in the manufacturer's instruction. The results were expressed as follows: units (U) per mg protein for SOD and GPx, and nmol/mg protein for MDA.
All results were expressed as mean ± SD, and statistical analysis were carried out with ANOVA or independent samples t-test of SPSS 16.0. P value of less than 0.05 was considered statistically significant.
Mouse Maternal Spleen Weight
The weight of spleen was gradually increased as gestation advanced in the control group during early pregnancy (Fig. 1), and statistically significant difference was found between E1–E3 and E5–E9 (P < 0.05). When compared with the control group, the spleen weight in the 6-OHDA-treated group was decreased by 10.9%–13.0% at E5–E9 (P < 0.05), especially in E7 (Fig. 1).
Sizes of SN and PLS of Maternal Spleen
Control animals exhibited normal spleen morphology on microscopic examination. As pregnancy advanced, the ratio of red pulp to the white pulp was increased (not shown in figure). Here, we put emphasis on SN and PLS (Fig. 2).
The distribution of SN was similar in both groups and was less prominent in 6-OHDA-treated mice. The SN of control mice was elliptical in shape, its area was increased at E5 (P < 0.05), and then maintained at E9 (P > 0.05). In the 6-OHDA-treated group, the SN was smaller in size than that of the control group. When compared with control mice, the SN areas of 6-OHDA-treated mice were decreased significantly (24.0%–34.1%, P < 0.05) at E1–E9, especially in E1 (Fig. 3A). Similarly, the area of PLS peaked at E5–E9 (P > 0.05) in the control group. PLS in 6-OHDA-treated mice was smaller than control mice (Fig. 2B), and its area was kept at a lower level during E1–E9. The area of PLS in the 6-OHDA-treated group was significantly decreased by 20.3% (P < 0.05) at E1, 26.1% at E5 (P < 0.05), and 20.1% at E9 (P < 0.05) when compared with the control group (Fig. 3B).
Expressional Characteristics of PCNA in Maternal Spleen
Immunohistochemical staining revealed brown nuclear staining (Fig. 4) on all sections except for those run as negative controls (not shown). PCNA-positive cells were mainly found in the SN, and few PCNA-positive cells were scattered in other regions, such as PLS, the red pulp, and marginal zone. In the control group, the positive cells of SN appeared densest and even dyeing (Fig. 4A), whereas they were dispersedly distributed in the 6-OHDA-treated group (Fig. 4B).
The PCNA-positive cells of SN were increased as gestation advanced in the two groups. In the control group, the increase was gradual from E1 to E5 (P > 0.05) and dramatic at E9 (P < 0.05; Fig. 5). The PCNA-positive cells were decreased by 15.6%, 10.9%, and 16.2% in the 6-OHDA-treated group at E1, E5, and E9, respectively (P < 0.05; Fig. 5).
Lymphocyte Transformation Ability
ConA-induced splenocyte proliferation.
The ConA-induced splenocyte proliferation remained relatively unchanged during early pregnancy in the control group (Fig. 6A) other than E5 (compared with E1, P < 0.05).
When compared with control mice in early pregnancy, the ConA-induced proliferation ability of splenocytes in 6-OHDA-treated mice was decreased (Fig. 6A). The difference was significant at E1, E3, E5, E7, and E9 (P < 0.05), and the reduction rates were 9.3%, 7.4%, 14.0%, 12.3%, and 8.8%, respectively.
LPS-induced splenocyte proliferation.
The proliferation of splenocyte induced by LPS was decreased at E5 (compared with E1 and E3, P < 0.05) and increased at E7 and E9 (compared with E5, P < 0.05 for E7, and P > 0.05 for E9) in the control group (Fig. 6B) during early pregnancy.
When compared with corresponding control mice, the LPS-induced splenocytes proliferation ability of 6-OHDA-treated mice were significantly decreased (P < 0.05) by 7.5%, 6.1%, 9.6%, 13.0%, and 20.2% at each sampling time point (E1, E3, E5, E7, and E9; Fig. 6B).
Analysis of Oxidant and Antioxidant Status
Detection of SOD activity.
As shown in Fig. 7A, the SOD activity of the spleen was decreased around implantation period (E5–E7, P < 0.05) at first and then increased (E9, P < 0.05) in control mice during early pregnancy.
The SOD activity of 6-OHDA-treated mice was decreased by 18.2%, 19.7%, 20.1%, 14.3%, and 21.9% at E1, E3, E5, E7, and E9, respectively (P < 0.05).
Detection of GPx Activity.
As shown in Fig. 7B, the GPx activity of the spleen was increased at E3, and then decreased at E5 (P < 0.05 when compared with their prior sampling time point) during early pregnancy in the control group.
When compared with that of the control group, the GPx activity of the 6-OHDA group was decreased by 4.2%, 20.9%, 18.2%, 17.4%, and 25.0% at E1, E3, E5, E7, and E9, respectively (P < 0.05).
Detection of MDA content.
The MDA content of spleen was decreased in the control group at E3 (P < 0.05), but increased at E5 (compared with E3, P < 0.05), and then decreased gradually (P > 0.05; Fig. 7C). During early pregnancy, the smallest MDA content of spleen occurred at E3 and E9, and the biggest content occurred at E1 and E5.
The similar trend was revealed in the 6-OHDA group other than that of the biggest MDA content that occurred at E7. When compared with the control group, the MDA content of the 6-OHDA group was increased by 5.6%, 14.9%, 10.6%, 38.6%, and 17.1% at E1, E3, E5, E7, and E9, respectively (P < 0.05).
Role of Sympathetic Nerves on Pregnancy-Dependent Changes of Maternal Spleen
Pregnancy represents a physiological stress and is associated with changes in blood volume resulting in hemodilution, which could serve as an effective activator of splenic erythropoiesis (de Rijk et al.,2002). In this study, we observed pregnancy-dependent structural and functional changes in the spleen during early pregnancy. The splenic weight was increased with pregnancy. Similar findings have also been reported in humans (Sasaki et al.,1981) and rats (Bustamante et al.,2008). Splenic growth reached its maximum size by the end of pregnancy, exhibiting an approximately 40% increase in weight and regressed postpartum to their nonpregnant weight (Bustamante et al.,2008). Norton et al. (2009) suggested that the transient increase in the size of the spleen during pregnancy is due to the expansion of erythroid precursors. However, our data of PCNA immunohistochemical and H&E staining showed an increase in the number of PCNA-positive cells of SN and an expansion in the splenic red pulp, which corresponds with a decrease in the white pulp, as gestation advanced. These results indicated that the pregnancy-dependent structural and functional changes in the spleen maybe due to the elevated proportion of proliferating cells in the red pulp during early pregnancy.
On the other hand, sympathetic nerve plays an important role in fetal survival and development during early pregnancy (Dong et al.,2007b). It has been reported that 6-OHDA treatment delayed the development of uterine glands and vessels by significantly increasing the number of CD8+ T cells and only slightly increasing the number of CD4+ T cells and by altering the secreting pattern of cytokine in uterine T cells during early pregnancy. The sympathetic nerve is considered as an influencing factor that can affect the structure and function of the spleen in nonpregnant female mice (Dong et al.,2007a). Our study suggests that the sympathetic nerve activity can regulate the pregnancy-dependent changes of the maternal spleen during early pregnancy. Administration of 6-OHDA to pregnant mice resulted in a significant decrease in the spleen weight (10.9%–13.0%) and in the areas of SN (24.0%–34.1%) and PLS (20.1%–26.1%) at E1–E9. The early presence of noradrenergic innervation in specific compartments of the spleen during critical periods of development points toward a role of NE in the maturation/development of the spleen (Elenkov et al.,2000). It has been reported that stimulation of the splenic nerve results in a transient decrease in the splenic artery flow and splenic weight in cat; however, no slow change in weight was observed during prolonged periods of stimulation (Greenway et al.,1968). The possible explanations for this difference between Greenway's and our study may be the different animal species, physiological state, and methods. Greenway et al. (1968) used methods of vascular responses of the spleen to nerve stimulation during normal and reduced blood flow. However, we further demonstrated that the proliferation of splenocytes induced by ConA and LPS decreased (P < 0.05) at entire sampling period by 6-OHDA treatment when compared with control mice. T and B lymphocytes can express α- and β-adrenergic receptors (AR; Kavelaars,2002). The decline of splenocyte proliferation ability in the 6-OHDA group suggests that the sympathetic nerve can adjust lymphocyte functions by AR in the spleen during early pregnancy.
Consequently, our findings confirmed that the sympathetic nerve activity can regulate the maternal spleen by altering the splenic structure and function and proliferation of resident cells during pregnancy. This finding is in agreement with those of Bustamante et al. (2008), who considered that the enlarged red pulp and a relatively shrink white pulp during pregnancy were related to the changed bloodstream and cellular subsets of the spleen.
It should be pointed out that the sympathetic nerve can affect the splenic structure and function under normal conditions, for example, the proliferation of splenocytes induced by ConA and LPS have shown to decrease in nonpregnant postsympathectomy female mice (Dong et al.,2007a). In addition, several reports have suggested that the immune reactivity was subjected to the regulation by the sympathetic nervous system in the lymphoid organs (Murray et al.,1992; Baerwald et al.,1997; Kohm and Sanders,2001). These findings suggest that the effects of sympathetic nerve on the spleen structure and function are not pregnancy dependent.
Regulatory Mechanism of Sympathetic Nerves on the Pregnancy-Dependent Changes of Maternal Spleen
Pregnancy is a physiological occasion primarily causing oxidative stress (Casanueva and Viteri,2003). Low levels of reactive oxygen species (ROS) are required for normal cell functions; however, elevated ROS production can cause oxidative stress and cellular damage. During normal human pregnancy, serum lipid peroxidation products are elevated but counterbalanced by an increased antioxidant activity (Buhimschi and Weiner,2001). Increased maternal oxidative stress in early human pregnancy is associated with pre-eclampsia, shortened gestation duration, lower infant birth weight, and so forth (Stein et al.,2008). In our study, after 6-OHDA treatment, SOD and GPx activities were decreased (P < 0.05) in the maternal spleen during early pregnancy. It is obvious that the antioxidant function of pregnant mouse spleen is impaired by 6-OHDA treatment. In contrast, the MDA content of spleen in the 6-OHDA group was increased than that of the control group during early pregnancy (P < 0.05). The result suggests that more free radicals are produced by the 6-OHDA treatment and may result in significant damage to cells. Our results demonstrate that the sympathetic nerve can modulate maternal lipid peroxidation and OFRs by adjusting SOD and GPx activities and MDA content during early pregnancy.
By combining the results reported in our study and other studies, we speculate that the sympathetic nervous system may regulate the pregnancy-induced adaptations of the spleen through the following routes: (1) adjusting the blood volume of the spleen and modulating recruitment, differentiation, and distribution of lymphocytes. These changes may lead to altered weight, structure, and cellularity of the spleen (Cesta,2006; Bustamante et al.,2008; Norton et al.,2009). (2) Regulating immune function of the spleen, even entire body. Successful pregnancy is a Th2 phenomenon. As described by Elenkov et al. (2000), endogenous catecholamines, for example, NE may cause a selective suppression of Th1 responses and cellular immunity and a Th2 shift toward dominance of humoral immunity. (3) Tuning oxidant state and antioxidant function of the spleen. Macarthur et al. (2008) confirmed that oxidative stress is indeed involved in the changes that occur in sympathetic neurotransmission in hypertension. (4) Modulating inflammatory reaction of the spleen/body. It has been suggested that the success of implantation is secondary to the development of an injury-induced inflammatory reaction, and the sympathetic nervous system may exert proinflammatory or anti-inflammatory effects locally (Elenkov et al.,2000; Dekel et al.,2010).
Despite its physiological significance in pregnancy, the animal lacking the spleen will probably be able to undergo successful pregnancy. In humans, a case of splenectomy done in childhood and delivering a healthy baby in adulthood has been reported (Prabhu Thangappah and Leela,2008). Successful emergent splenectomy during pregnancy in a patient with life-threatening idiopathic thrombocytopenia has been reported by Stiemer et al. (1996). The possible explanation to this phenomenon includes complete reproductive system and incomplete immune system in these cases, which can still support pregnancy requirement and alleviate immunological rejection.
Taken together, the demands of pregnancy cause marked adaptations in the maternal spleen, and the sympathetic nerve activity contributes to this process through modulation of the splenic parameters mentioned above. However, the exact mechanism by which this modulation occurs remains unclear. Understanding the interactions between the sympathetic nervous system and the spleen can provide valuable insight in early pregnancy and help in account for fetal loss.