Near-term pregnant rats show a suppressed fever response to LPS that is associated with reduced induction of cyclooxygenase (COX)-2 in the hypothalamus. The objective of this study is to explore whether the LPS-activated signaling pathways in the fever-controlling region of the hypothalamus are specifically altered at near term. Three rat groups consisting of 15-day pregnant rats, near-term 21- to 22-day pregnant rats, and day 5 lactating rats were injected with a febrile dose of LPS (50 μg/kg ip). The hypothalamic preoptic area and the organum vasculosum of the lamina terminalis (OVLT) were collected 2 h after LPS injection. The activation of three transcription modulators, nuclear factor-κB (NF-κB), extracellular signal-regulated kinase 1/2 (ERK1/2), and signal transducer and activator of transcription 5 (STAT5), was assessed using semiquantitative Western blot analysis. LPS activated the NF-κB pathway in all rat groups, and this response was not altered at near term. ERK1/2 and STAT5 were constitutively activated during all reproductive stages, and their levels were not significantly affected by LPS injection. Plasma levels of the proinflammatory cytokines (IL-1β, IL-6, TNF-α, and IFN-γ), anti-inflammatory cytokines (IL-4, IL-10, and IL-1 receptor antagonist), and corticosterone were unaffected during the three reproductive stages after LPS challenge. We observed a sharp decrease in the expression of a prostaglandin-producing enzyme called lipocalin-prostaglandin D2 synthase in near-term pregnant and lactating rats. Thus fever suppression at near term is not due to an alteration in either LPS-activated intracellular signaling pathways or LPS-induced pro- and anti-inflammatory cytokine production.
- nuclear factor-κB
- extracellular signal-regulated kinase 1/2
- signal transducer and activator of transcription 5
the febrile response is part of the body's natural defense against invading pathogens (37). Upon viral or bacterial infection, immune competent cells produce a series of proinflammatory cytokines (IL-1β, IL-6, TNF-α, and IFN-γ) that regulate the immune responses through the activation of various transcription factors within the fever-controlling region of the hypothalamus (22, 29, 54, 60). This fever-controlling region consists of the preoptic area and the organum vasculosum of the lamina terminalis (POA/OVLT) (4, 62). These transcription factors, especially the nuclear factor-κB (NF-κB), activate the inducible cyclooxygenase (COX-2) enzyme (71) in the endothelial and perivascular cells of the brain (6, 44, 63). COX-2 converts arachidonic acid to prostaglandin H2 (PGH2). Subsequently, a downstream microsomal prostaglandin synthase converts this unstable PGH2 into PGE2 (35, 61). The resulting PGE2 binds to its receptors to reduce heat loss and to augment heat retention, thus increasing body temperature (56). Another downstream prostaglandin synthase called lipocalin-type PGD2 synthase (L-PGDS) can also convert PGH2 into PGD2, which subsequently leads to the production of anti-inflammatory prostaglandins (49, 68).
In addition to the proinflammatory stimulation, the immune challenge activates other antipyretic/anti-inflammatory molecules, including glucocorticoids (16, 48), anti-inflammatory cytokines [IL-1 receptor antagonist (IL-1ra), IL-4, and IL-10] (7, 8), and neuropeptides (α-melanocyte-stimulating hormone and arginine vasopressin) (57, 66).
It is well established that the febrile response to various pathogens is suppressed in pregnant animals at near term (10, 36, 42, 43, 69). However, the mechanisms that underlie this suppressed neuroimmune response are still elusive (10, 12, 17, 18). We and others have observed that fever suppression at near term was associated with a reduction in the level and the activity of COX-2 in the hypothalamus (19, 33, 51). Thus it was suggested that the refractoriness of near-term pregnant rats to pyrogens is, at least in part, due to a general reduction of the synthesis and activity of the inducible COX-2 in the rat brain. Because LPS, and subsequently the LPS-induced proinflammatory cytokines, activate cox-2 gene expression through a multitude of signaling pathways such as the NF-κB pathway, the extracellular signal-regulated kinases (ERK1/2) pathway, and the signal transducer and activator of transcription (STAT5) pathway (65, 72), we explored whether LPS activation of these pathways is affected in the POA/OVLT of near-term pregnant rats. Concomitantly, we assessed whether the LPS-induced proinflammatory cytokine production is also altered at near term.
Moreover, the immune-activated NF-κB signaling pathway is also involved in the transcriptional control of the l-pgds gene that encodes for L-PGDS (21, 53). Because L-PGDS activation leads ultimately to the production of anti-inflammatory cytokines (49, 68), we tested whether L-PGDS protein expression is equally reduced in pregnant rats at near term, as was the case for COX-2 protein, or whether its levels are augmented to account for the reduced fever at near term. Furthermore, we tested whether plasma levels of anti-inflammatory cytokines were altered in immune-challenged near-term pregnant rats compared with 15-day pregnant or lactating rats.
On the other hand, it is well known that the basal hypothalamic-pituitary-adrenal (HPA) activity, and consequently the plasma levels of corticosteroids, are significantly enhanced at near term (3, 14). Because corticosteroids attenuate the febrile response (48), we tested whether plasma corticosterone levels are also altered in immune-challenged near-term pregnant rats.
MATERIALS AND METHODS
Sprague-Dawley timed pregnant rats from Charles River Laboratories were individually housed in a temperature-controlled room (22 ± 1°C) under a 12:12-h light-dark cycle (lights on 0700). All experimental protocols were approved by the University of Calgary Animal Care Committee and were carried out in accordance with the Canadian Council of Animal Care guidelines.
Two hours after intraperitoneal injection of LPS (50 μg/kg ip, Escherichia coli, serotype 026:B6; Sigma) or 0.9% pyrogen-free saline, rats at pregnancy day 15 (G15 group), rats at pregnancy day 21–22 (G22 group, near term), and rats at the 5th day of lactation (L5 group) were perfused with PBS composed of the following chemicals (in mM): 137 NaCl, 2.7 KCl, 10 Na2HPO4, and 1.8 KH2PO4, to remove the blood. This time point was chosen because the peak of NF-κB activity in the hypothalamus is at ∼2 h after immune challenge (54). The brain region that consists of the OVLT and the POA was cut in a triangle shape where the upper angle is located at the midpoint of the anterior commissure and the base is delimited by the cortex invaginations (with the tip of the piriform cortex excluded). Both the OVLT and the POA have been shown to be important in the generation of fever (4, 62). The brain tissues were quickly dissected and put in lysis buffer [20 mM MOPS, 4.5 mM Mg(C2H3O2)2, 150 mM KCl, and 1% Triton] supplemented with a mixture of protease and phosphatase inhibitors, as previously described (49).
Western blot analysis.
Western blotting was performed as previously described (49) with slight modifications. Briefly, protein extracts from each individual rat were separated on 10% SDS-PAGE (15% SDS-PAGE for L-PGDS detection). The proteins were transferred onto nitrocellulose membranes. The membranes were then incubated for 2 h at room temperature with 5% fat-free milk in Tris-buffered saline containing Tween 20 (TBS-T) composed of 20 mM Trizma base (Sigma), 0.15 M sodium chloride (Fisher Scientific), and 0.1% polyoxyethylenesorbitan monolaurate (Sigma). The membranes were then incubated with one of the following primary antibodies: rabbit antibodies for phospho-IκB (Santa Cruz Biotechnology; 1:500) and phospho-STAT5 (Cell Signaling Technology; 1:1,000), monoclonal antibody for phospho-ERK1/2 (Cell Signaling Technology; 1:1,000), or goat antibody for L-PGDS (Santa Cruz Biotechnology; 1:1,000) overnight at 4°C. After being washed in TBS-T, the membranes were incubated for 1 h at room temperature with secondary antibodies conjugated with horseradish peroxidase as follows: goat anti-rabbit for phospho-IκB and phospho-STAT5 (Santa Cruz Biotechnology; 1:4,000), donkey anti-goat for L-PGDS (Santa Cruz Biotechnology; 1:2,000), or goat anti-mouse for phospho-ERK1/2 (Santa Cruz Biotechnology; 1:4,000). A chemiluminescence substrate was applied to the membrane (ECL kit; Amersham Biosciences, Arlington, IL), and protein bands were visualized using Kodak X-Omat film (Eastman Kodak, Rochester, NY). Because the phospho-IκB signal was very weak, the ECL kit was substituted with an ECL Plus kit (Amersham Biosciences, Amersham, UK) to enhance the sensitivity of the protein detection.
After protein detection, membranes were stripped with β-mercaptoethanol (BDH) and reblotted with rabbit anti-actin antibody (Sigma; 1:10,000), rabbit anti-IκB (Santa Cruz Biotechnology; 1:4,000), or rabbit anti-STAT5 (Santa Cruz Biotechnology; 1:4,000) as appropriate (see results) and processed as described above.
Two hours after LPS injection (50 μg/kg ip), rats were anesthetized with pentobarbital sodium (60 mg/kg ip). The rats were rapidly decapitated, and trunk blood samples were collected in test tubes. The plasma samples were separated from the blood by centrifugation, quickly frozen in liquid nitrogen, and stored in a −80°C freezer until assayed. The plasma levels of IL-1β, IL-4, IL-6, IL-10, IFN-γ, and TNF-α were assayed using a multiplex assay (Linco Diagnostic Services). The sensitivity of this assay is 25 pg/ml. The plasma level of IL-1ra was determined using a specific rat ELISA kit (Biosource International). The sensitivity of the IL-1ra assay is 12 pg/ml. The amount of free corticosterone in the plasma was also assayed using a commercially available ELISA kit (R&D Systems). The sensitivity for the corticosterone assay is 27 pg/ml.
For Western blot analysis, the area under the intensity profile curve of a given band was quantified (Bio-Rad). The ratios of optical density values of either phospho-protein/protein or protein/actin were calculated. Immunoblot data were compared using a two-way analysis of variance followed by Student-Newman-Keuls post hoc comparisons whenever possible. All cytokine and corticosterone data were compared using one-way analysis of variance. The significance was accepted at P < 0.05.
Activation of NF-κB pathway at near term.
NF-κB is normally sequestered in an inactive form by an inhibitory molecule called IκB. Upon immune stimulation, IκB is phosphorylated and subsequently detached from NF-κB (40). The free and functionally active NF-κB is translocated into the nucleus and induces the transcription of its target genes. Thus we used the phosphorylation levels of IκB as a measure of NF-κB activity (23). As presented in Fig. 1A, the phosphorylated form of IκB (p-IκB) was virtually absent in the POA/OVLT of pregnant and lactating rats injected intraperitoneally with pyrogen-free saline, whereas the level of nonphosphorylated IκB was similar at all stages. LPS injection (50 μg/kg ip) significantly enhanced the level of p-IκB in the POA/OVLT of all three reproductive stages (Fig. 1B). The ability of LPS to stimulate the phosphorylation of IκB was unaffected by the pregnancy stage or lactation. The level of nonphosphorylated IκB was also unaltered in the all rat groups challenged with LPS. Similarly, as shown in Fig. 1C, the ratio p-IκB/κB, as evaluated using densitometric analysis, remained constant in all reproductive stages.
Activation of ERK and STAT5 pathways at near term.
Figure 2, A and B, shows the phosphorylated forms of ERK1 (p-ERK1) and ERK2 (p-ERK2) proteins. We found that these two proteins were constitutively phosphorylated in the POA/OVLT of G15, G22, and L5 rats. The constitutive activation of ERK1/2 was not significantly different between G15, G22, or L5 animals and was not altered by LPS.
STAT5 was also constitutively phosphorylated (p-STAT5) in the POA/OVLT of G15, G22, and L5 rat groups (Fig. 2C). However, the ratio pSTAT5/STAT5 was significantly reduced in L5 rats compared with G15 rats. In LPS-injected rats (50 μg/kg ip), the ratio pSTAT5/STAT5 was not altered at any reproductive stage.
L-PGD2 synthase at near term.
The expression of L-PGDS is also activated by various immune stimuli (34). We therefore explored whether the L-PGDS response to an LPS challenge is altered at near term. As shown in Fig. 3A, G15 rats expressed a constitutively high level of L-PGDS in the POA/OVLT (saline injected). At near term (G22), there was an approximately fourfold decrease in the amount of L-PGDS. This reduced expression in L-PGDS was sustained during the lactation period (L5). In LPS-treated rats, L-PGDS level was also high at day 15 of pregnancy. It was slightly, but not significantly, decreased in rats close to term and in lactating rats. We observed that the LPS-injected near-term rats showed an apparent enhanced level of L-PGDS compared with their saline-injected counterparts. This enhanced response was not significant (P = 0.09).
Profile of pro- and anti-inflammatory cytokines after LPS challenge.
Figure 4 shows the cytokine profile of G15, G22, and L5 rats 2 h after LPS injection (50 μg/kg ip). LPS-induced proinflammatory cytokine production (IL-β, IL-6, TNF-α, and IFN-γ) was not altered in pregnant rats at near term (G22) compared with G15 or L5 rats. Similarly, the plasma levels of the anti-inflammatory cytokines (IL-1ra and IL-10) in LPS-injected rats at near term (G22) were not significantly different from those of LPS-injected G15 or L5 rats. However, we observed that IL-10 in the plasma of LPS-injected lactating rats was below the detection limit. The anti-inflammatory cytokine IL-4 was virtually absent in the plasma of all three animal groups (data not shown).
Response of plasma corticosterone to LPS at near term.
The basal HPA activity, and consequently the production of corticosteroids, was specifically enhanced at near term (3, 14). We therefore explored whether the corticosterone response to LPS was also altered in near-term pregnant rats, which could account for the fever suppression. Figure 5 shows the plasma levels of free corticosterone in G15, G22, and L5 rat groups 2 h after LPS injection (50 μg/kg ip). The plasma corticosterone level in near-term pregnant rats did not differ from that in either G15 or L5 rats (P > 0.05).
Neuroimmune responses of near-term pregnancy have been widely studied by looking at the febrile response to inflammatory stimuli, such as LPS. To our knowledge, this is the first integrative study in which immune challenges’ effects on 1) intracellular signaling pathways in the POA/OVLT, 2) the HPA axis, and 3) peripheral inflammatory cytokines were explored in nonterm-pregnant (G15), near-term pregnant (G22), and lactating rats (L5). This study showed that plasma levels of corticosterone and inflammatory cytokines (both pro- and anti-inflammatory) were not altered in near-term immune-challenged rats. Concordantly, the signaling pathways of these inflammatory cytokines were similarly unchanged in the POA/OVLT of near-term pregnant rats compared with G15 or lactating rats. These observations lead us to believe that the suppression of fever at near term was not due to an enhanced immune-activated HPA axis. The immune activation at near term is affected by factors that do not involve either the depression of proinflammatory or stimulation of anti-inflammatory cytokines.
We and several other investigators have shown that the expression and activity of the immune-induced COX-2 enzyme was reduced at near term (19, 33, 51). Knowing that LPS-induced COX-2 is largely mediated through the activation of the NF-κB signaling pathway, one would expect to see a reduction in LPS-induced NF-κB activity at near term. Using the phosphorylation of IκB as a marker of the NF-κB activity in the POA/OVLTs of the hypothalamus (40), we did not see any significant change in the activation of the NF-κB pathway. We explored phosphorylation of IκB 2 h after LPS injection. This time point was chosen because the peak of NF-κB activity in the hypothalamus occurs at ∼2 h after immune challenge (54). We did not explore NF-κB activation at later time points after LPS injection because later NF-κB activation would occur after the fever peak (2.5–4 h after LPS injection) and would not be important in events leading to fever generation.
It can be argued that a better measure of NF-κB activity is its actual binding to its DNA motifs. However, it should be kept in mind that measurement of NF-κB binding to its DNA is not always associated with gene transcription. In many circumstances, NF-κB can be translocated to the nucleus and binds to its DNA with no measurable transcriptional effect. This phenomenon can be due, at least in part, to mutual inhibitory feedbacks between NF-κB and the immune-modulating steroids (sex steroids and glucocorticoids), where these steroids interfere with the already DNA-bound NF-κB via physical or steric effect (13, 45). Furthermore, there was no significant change in the LPS-induced inflammatory cytokines in the plasma of all reproductive stages, which may explain why the activation of the NF-κB signaling pathway was unaltered. This observation substantiates the validity of our approach.
ERK and STAT5 pathways.
Whereas NF-κB pathway was inactive in the hypothalamus of saline-injected pregnant rats, both ERK1/2 and STAT5 pathways were constitutively active. This observation suggests that the physiological states of pregnancy are accompanied by constitutive production of ERK1/2 and STAT5 transcription activators. There are several potential candidates for ERK1/2 and STAT5 activation in an immune-independent fashion. For example, prolactin can activate both ERK1/2 and STAT5 pathways (1, 11, 27, 73), and both prolactin plasma levels and hypothalamic prolactin receptor expression are high during pregnancy (2, 25, 26). Another candidate for the constitutive activation of the ERK1/2 and STAT5 pathways is leptin (30, 31), whose circulating levels are increased during pregnancy (38, 64). These, and likely other ERK and STAT5 activators, may respond to the physiological demand of pregnant and lactating rats in a nonimmune-dependent fashion.
The reduction of the activated STAT5 in the hypothalamus of saline-injected lactating rats was rather surprising, because the high prolactin levels during lactation should result in a highly activated STAT5 signaling pathway. It is possible that the precipitous fall in progesterone levels at parturition results in reduced hypothalamic STAT5 activation by prolactin, because progesterone has been shown to enhance prolactin-induced STAT5 activation (58).
Upon LPS injection, there is an increase in a battery of proinflammatory (pyretic) and anti-inflammatory (antipyretic) cytokines that ultimately affect the febrile response. Thus the reduction in the circulating levels of proinflammatory cytokines or an increase in the levels of anti-inflammatory cytokines in immune-challenged, near-term pregnant rats would result in an attenuated febrile response. In a study by Fofie et al. (20), it has been shown that, whereas a relatively high dose of LPS (160 μg/kg) stimulated the release of both proinflammatory (TNF-α, IL-1β, and IL-6) and anti-inflammatory (IL-1ra) cytokines in nonpregnant rats (as expected from numerous previous studies), LPS-challenged near-term rats were not able to produce a significant amount of some proinflammatory cytokines (IL-1β and IL-6) but were still able to respond via an increase in the anti-inflammatory cytokine IL-1ra. In our study, near-term pregnant rats injected with a more modest febrile dose of LPS (50 μg/kg) produced quantities of proinflammatory (IL-1β, IL-6, TNF-α, and IFN-γ) and anti-inflammatory (IL-1ra, IL-10, and IL-4) cytokines that were not significantly different among G15, G22, and L5 rats. This apparent discrepancy may stem from the fact that we used a lower LPS dose than that used by Fofie et al. (20), which may account for different levels of activation of cytokines. In addition, we compared LPS-induced cytokine release between 15-day pregnant, near-term pregnant, and lactating rats, whereas in the study by Fofie et al., the comparison was made between near-term pregnant and nonpregnant rats. We choose to compare cytokine production at near term to that of 15-day pregnant and lactating rats for two reasons: 1) nonpregnant rats are randomly cycled rats whose changing levels of ovarian hormone during the estrous cycle can affect both LPS-induced inflammatory cytokines and the fever response (50, 52), and 2) the febrile response is reduced specifically at near term and is unaltered in G15 pregnant rats or during lactation (42). Thus pregnant rats that are not close to term would appear to be the best physiological control group for the near-term pregnant rat group.
The plasma levels of inflammatory cytokines measured in this study are similar to the cytokine levels detected in immune-challenged randomly cycling females or ovariectomized rats that were supplemented with estrogen and progesterone (52, 67). However, it seems that LPS-induced IL-1β and IL-6, but not LPS-induced TNF-α, is reduced in late-term pregnant rats compared with nonpregnant female rats (20), a finding that is in agreement with a generally reduced immune response during pregnancy (70). On the other hand, plasma levels of corticosterone in immune-challenged pregnant rats measured in the present study are of the same order of magnitude as those corticosterone levels reported recently in late pregnancy in rats (5). However, the corticosterone levels were higher in immune-challenged nonpregnant rats compared with late-term pregnant rats (5). This observation explains why the activation of HPA axis is also dampened at near term (14).
Our experimental results also corroborate with clinical studies that have shown no increase in either IL-1ra (59) or IL-10 (32) plasma levels as a function of the stage of pregnancy in women. Similarly, there was no change in basal or LPS-induced proinflammatory cytokine production by immune-competent cells of women at different stages of pregnancy (41).
The fact that LPS-induced inflammatory cytokines were not affected in all three reproductive stages questions the notion of the necessary role of endogenous inflammatory cytokines in mediating all of the LPS effects. In light of the recent evidence for a direct effect of LPS on the brain (9), it is possible that the brain sensitivity to a direct effect of LPS is reduced at near term. We are not aware of any study that has explored the expression of LPS receptors, such as Toll-like receptor 4 (TLR4), in rat hypothalamus during pregnancy. However, the expression of TLR4 is reduced in peripheral tissues such as the placenta of near-term pregnant mice (28). The dampened response to the immune challenge can also be brought about, at least in part, by the inhibitory effect of the opioidergic system, which is highly activated at near term (5).
The HPA axis activation results in the production of the antipyretic/anti-inflammatory corticosterone. The fact that basal HPA activity is specifically enhanced at near term (3, 14) led us to explore the importance of HPA involvement in the observed fever reduction at near term. In the present study, there was no difference in LPS-stimulated circulating free corticosterone levels of pregnant and lactating rats. In addition, late pregnancy is already known to be associated with a reduced HPA response to emotional stress (14, 55). Therefore, neither possible emotional stress response to the injection nor the immune stimulation itself extraordinarily enhances HPA activity at term that could account for the reduced fever.
COX-2 and L-PGDS.
It is still unclear how LPS-induced COX-2 is reduced at near term (19, 33, 51). It is possible that the dampened noradrenergic activity at near term (15) affects LPS-induced COX-2 at this reproductive stage. It is now accepted that COX-2 activity can also lead to the production of anti-inflammatory prostaglandins (PGD2 pathway). Indeed, the prostaglandin product of the enzymatic activity of COX-2 (PGH2) is the substrate for downstream prostaglandin synthase L-PGDS. L-PGDS converts PGH2 into PGD2, which subsequently leads to the production of anti-inflammatory prostaglandins (49, 68). Because L-PGDS leads to the production of anti-inflammatory prostaglandins, we hypothesized that L-PGDS would be increased at near term. In contrast, we observed a rather sharp reduction in L-PGDS level at near term and during lactation. However, there was a high variability in the levels of L-PGDS, probably due to individual differences. Our results strongly suggest that the fever reduction at near term is not due to enhanced production of anti-inflammatory prostaglandins.
To our knowledge, this is the first study to show that L-PGDS expression is altered in the POA/OVLT of near-term pregnant rats. The suppression of L-PGDS expression matches what happens in the expression pattern of COX-2 at near term and during lactation (51). It seems counterintuitive that both COX-2 and L-PGDS are downregulated at near-term pregnancy, because COX-2 plays a key role in fever, whereas the prostaglandin product of L-PGDS (PGJ2) possesses anti-inflammatory/antipyretic properties. However, as mentioned above, both cox-2 and l-pgds genes are under the control of the same transcription factors (21, 53, 71). Thus it is conceivable that the regulation of both of these genes occurs in parallel. Another example of this parallel regulation is the concomitant upregulation of both COX-2 and L-PGDS enzymes during the resolution of inflammation (24).
Recent data suggest that sex hormones (namely, estrogen) reduce the expression levels of L-PGDS in the preoptic region (part of the thermoregulatory region of the hypothalamus) (47). Thus the high levels of estrogens at late pregnancy and during lactation might be responsible for the observed reduction in L-PGDS in the POA/OVLT. The physiological reason for the reduced hypothalamic L-PGDS at near term and during lactation is not well understood. PGD2, a product of L-PGDS, is known for its somnogenic effects (68). The observed reduced L-PGDS in the hypothalamus may account, at least in part, for the sleep disorder that is known to occur during late pregnancy and during lactation (39, 46).
The work described in this report clearly shows that the suppression of fever in near-term pregnant rats is not due to an alteration of the endotoxin-induced inflammatory cytokines and is not the result of a deficiency in the activation of immune-related signaling pathways in the thermoregulatory regions of the hypothalamus. Although these results do not explain why the COX-2 response to endotoxin is reduced at near term, they do suggest that mechanisms other than the signaling pathways explored in the present work may contribute to COX-2 reduction and fever suppression at near term.
This work was supported by the Canadian Institutes of Health Research. S. Ellis is an Alberta Heritage Foundation for Medical Research (AHFMR) and Natural Sciences and Engineering Research Council Student. E.-M. Harré is an AHFMR Fellow. Q. J. Pittman is an AHFMR Medical Scientist.
We thank Mio Tsutsui for technical support.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society