| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Physiology (H.H., H.F., M.K., T.S., M.S., H.O., Y.U.), School of Medicine, University of Occupational and Environmental Health, Kitakyushu 807-8555, Japan; and Laboratory of Physiology (Y.T.), Department of Marine Bioscience, Ocean Research Institute, University of Tokyo, Tokyo 164-8639, Japan
Address all correspondence and requests for reprints to: Yoichi Ueta, M.D., Ph.D., Department of Physiology, School of Medicine, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807-8555, Japan. E-mail: yoichi{at}med.uoeh-u.ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most species show a close relationship between drinking and feeding (6, 7, 8). About 80% of spontaneous daily water intake is temporally associated with feeding in rats (9, 10). Ghrelin, a 28-amino-acid neuropeptide and an endogenous ligand for the GH secretagogue receptor, was first isolated from the stomach (11). However, ghrelin is found in the brain and is now also recognized as a neuropeptide. In addition to the stimulation of GH secretion from the anterior pituitary, central and peripheral administration of ghrelin strongly stimulates feeding in mammals (12, 13, 14) and nonmammals (15, 16). It has been further shown that peripheral administration of ghrelin activates hypothalamic orexigenic neurons and inhibits anorectic neurons to induce hunger (17, 18, 19). Therefore, ghrelin is established as a major orexigenic hormone acting not only from the periphery but also locally in the brain (11). Furthermore, Ishizaki et al. (20) reported that intracerebroventricular (icv) and iv injection of ghrelin increased plasma arginine vasopressin (AVP) levels in conscious rats. AVP is well known as an important hormone involved in body fluid balance. Thus, there is a possibility that ghrelin may have a potent effect on drinking behavior and body fluid balance in mammals and other nonmammalian species. Kozaka et al. (21) reported that the central administration of ghrelin inhibited water drinking in eels, although there has been no study thus far that examined the effects of ghrelin on water intake in mammals. Interestingly, the antidipsogenic effects of ghrelin on eel were more potent than those of ANP, the most potent antidipsogenic hormone that has been reported in eels as well as in mammals (21, 22).
The purpose of the present study was to examine whether centrally and peripherally administered ghrelin modulates drinking in mammals, as observed in eels. To highlight the inhibitory effect, 24-h-dehydrated rats were afforded with water just after icv and iv injection of ghrelin. In addition to water intake, food intake and urine volume were measured after the injections because these parameters influence drinking. The effects of ghrelin were compared with those of ANP to evaluate the potency of the two hormones. Finally, using immunohistochemistry for Fos, we examined the effects of icv and iv administration of ghrelin on the induction of c-fos protein (Fos) in the rat central nervous system, including the organum vasculosum of the lamina terminalis (OVLT), median preoptic nucleus (MnPO), subfornical organ (SFO), supraoptic nucleus (SON), paraventricular nucleus (PVN), arcuate hypothalamic nucleus (Arc), area postrema (AP), and nucleus of the tractus solitarius (NTS). The expression of the c-fos protein has been widely used to detect neuronal activity in the central nervous system (23).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All procedures in the present study were done in accordance with the guidelines on the use and care of laboratory animals as set out by the Physiological Society of Japan and under the control of the Ethics Committee of Animal Care and Experimentation, University of Occupational and Environmental Health, Japan.
Surgical procedures
For icv injection of solutions, the animals were implanted with stainless steel cannulae aimed at the lateral ventricle. The animals were anesthetized (sodium pentobarbital, 50 mg/kg body weight, ip injection) and then placed in a stereotaxic frame. A stainless steel guide cannula (550 µm outer diameter, 10 mm length) was implanted stereotaxically at the following coordinates: 0.8 mm posterior to the bregma, 1.4 mm lateral to midline, and 2.0 mm below the surface of the left cortex such that a tip of the cannula was 1.0 mm above the left cerebral ventricle (24). Two stainless steel anchoring screws were fixed to the skull, and the cannula was secured in place by acrylic dental cement.
For iv injection of solutions, venous catheters (PE-50) filled with heparinized saline (200 U/ml) were inserted into the jugular vein in rats under anesthesia with isoflurane (3% via a mask at a flow rate of 450 ml/min air). They were tunneled under the skin to exit at the nape of the neck.
After the surgical procedure, the animals were handled every day, housed in a plastic cage each, and allowed to recover for at least 5 d. Thereafter they were handled every day and housed individually in metabolic cages for 2 d before the start of the experiments.
Central administration of ghrelin, ANP, and vehicle
For icv injection of ghrelin, ANP, or vehicle, a stainless steel injector (300 µm, outer diameter) was introduced through the cannula at a depth of 1.0 mm beyond the end of the guide. The total volume of injected solution of ghrelin, ANP, and saline into the lateral ventricle was 10 µl. Rat ghrelin and rat ANP1–28 were purchased from the Peptide Institute (Minoh, Japan). Ghrelin and ANP1–28 were dissolved in pyrogen-free sterile 0.9% saline solution (Otsuka Pharmaceutical Co. Ltd., Japan).
Peripheral administration of ghrelin and vehicle
For iv injection of ghrelin or vehicle, ghrelin was dissolved in pyrogen-free sterile 0.9% saline solution (Otsuka Pharmaceutical). After iv injection of ghrelin, we administered heparinized saline (200 U/ml, 0.1 ml).
Experimental procedures
The animals were housed individually in metabolic cages for 2 d before the start of the experiments. We repeated the use of the metabolic cages four times.
In the first experiment, the animals, which had been deprived of water for 24 h, were put into metabolic cages after icv injection of ghrelin (0.1, 1, and 10 nmol/rat), ANP (0.1, 1, and 10 nmol/rat), or vehicle. The number of rats was six to nine in each group. We measured the cumulative water intake, food intake, and urine volume 30–180 min after icv injection of the solutions.
In the second experiment, animals, which had been deprived of water for 24 h, were put into metabolic cages after iv injection of ghrelin (0.1, 1, and 10 nmol/kg) or vehicle. The number of rats was six in each group. We measured the cumulative water intake, food intake, and urine volume 30–180 min after iv injection of the solutions.
In the third experiment, animals, which had been deprived of water for 24 h, were put into metabolic cages after icv injection of ghrelin (1 nmol/rat), iv injection of ghrelin (10 nmol/kg), or vehicle. The number of rats was six in each group. We measured the cumulative water intake, food intake, and urine volume 15 min, 30 min, 60 min, 120 min, 180 min, 6 h, and 24 h after icv and iv injection of the solutions.
In the fourth experiment, animals, which had been deprived of water for 24 h, were put into metabolic cages after icv injection of ghrelin (0.1 and 1 nmol/rat), ANP (0.1 and 1 nmol/rat), or vehicle. The number of rats was six to eight in each group. We measured the cumulative water intake and urine volume without food to remove the effects of a prandial drinking 30 and 60 min after icv injection of the solutions.
In the final experiment, ghrelin (1 nmol/rat icv or 10 nmol/kg iv) or vehicle was administered icv and iv (n = 3 in each group). Ninety minutes after icv and iv administration of the solution, the animals were anesthetized deeply (sodium pentobarbital, 75 mg/kg body weight, ip) after perfusion, and then the fixed brains were used for immunohistochemistry for Fos.
Fos-like immunoreactivity (LI)
The deeply anesthetized animals were perfused transcardially with 0.1 M phosphate buffer (PB) (pH 7.4) containing heparin (1000 U/liter saline) followed by 4% paraformaldehyde and 0.2% picric acid in 0.1 M PB. The brains were then removed and divided into three blocks that included the forebrain, hypothalamus, and brain stem. The blocks were postfixed with 4% paraformaldehyde and 0.2% picric acid in 0.1 M PB for 48 h at 4 C. The tissues were then cryoprotected in 20% sucrose in 0.1 M PB for 48 h at 4 C. Serial sections of 40 µm were cut using a microtome. The sections were rinsed twice with 0.1 M PBS containing 0.3% Triton X-100 and incubated in 0.1 M PBS containing 0.3% Triton X-100 with 1% hydrogen peroxidase for 60 min. They were then rinsed twice with 0.1 M PBS containing 0.3% Triton X-100. The floating sections were incubated with a primary Fos antibody (sc-52; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at a dilution of 1:500 in 0.1 M PBS containing 0.3% Triton X-100 at 4 C for 4 d. After washing for 20 min in 0.1 M PBS containing 0.3% Triton X-100, the sections were incubated for 120 min with a biotinylated secondary antibody solution (1:250) and finally with an avidin-biotin peroxidase complex (Vectastain ABC kit; Vector Laboratories, Inc., Burlingame, CA) for 120 min. The peroxidase in the sections was visualized with 0.02% diaminobenzidine in a Tris buffer containing 0.05% hydrogen peroxidase for 1.5 min.
The sections were mounted onto gelatin-coated slides, air dried, dehydrated in 100% ethanol, cleared using xylene, and then finally coverslipped and examined under a light microscope. The presence of a dark brown label that appeared in round structures was judged to be indicative of Fos-LI-positive nuclei (25).
Statistical analysis
A mean deviation from control ± SEM was calculated from data obtained from the measurements of the cumulative water drinking, food intake, urine volume, and immunohistochemistry for Fos. Each group within an experiment was compared with the control group. The data were analyzed using a one-way fractional ANOVA followed by a Bonferroni correction for multiple comparisons. The statistical significance was set at P less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and D). The antidipsogenic effects were dose dependent, and the effect of ghrelin tended to be greater than ANP at 30 min (Fig. 1D). However, the effect of ghrelin was only transient, compared with ANP, and the effect was no more significant 120–180 min after icv injection than that of the vehicle-injected controls (Fig. 1A). Thus, the inhibition of ANP was greater than that of ghrelin 120 and 180 min after injection (Fig. 1A).
|
). These effects lasted 180 min after icv and iv injection of ghrelin (Fig. 1B). Food intake did not change after injection of ANP (0.1, 1, and 10 nmol/rat) in comparison with the vehicle (Fig. 1, B and E).
Urine volume did not change after icv injection of ghrelin (0.1, 1, and 10 nmol/rat) and ANP (0.1, 1, and 10 nmol/rat) in comparison with the vehicle, except for icv injection of ghrelin (1 nmol/rat) (Fig. 1, C and F).
Moreover, we examined a dose-response study after iv injection of ghrelin.
Effects of iv injection of ghrelin on water intake, food intake, and urine volume: a dose-response study
Water intake was significantly inhibited 30 min after iv injection of ghrelin (0.1, 1, and 10 nmol/kg) in comparison with the vehicle (Fig. 2, A and D). The effects of iv injection of ghrelin (1 nmol/kg) lasted 180 min after injection (Fig. 2A).
|
, B and E). These effects lasted 180 min after iv injection of ghrelin (Fig. 2B).
Urine volume did not change after iv injection of ghrelin (0.1, 1, and 10 nmol/kg) in comparison with the vehicle (Fig. 2, C and F).
Effects of icv and iv injection of ghrelin on water intake, food intake, and urine volume: a time-course study
The cumulative water intake was significantly inhibited 15–60 min by icv injection of ghrelin (1 nmol/rat) in comparison with the vehicle (Fig. 3A). However, the cumulative water intake was not significantly inhibited 120–180 min and 6 and 24 h after icv injection of ghrelin. The cumulative water intake was significantly inhibited at 15–30 min by iv injection of ghrelin (10 nmol/kg) in comparison with the vehicle (Fig. 3B). The cumulative water intake was no more inhibited 60–180 min and 6 and 24 h after iv injection of ghrelin.
|
, C and D). The effects of icv and iv injection of ghrelin lasted 6 h. The cumulative food intake was not greater than in the controls 24 h after icv and iv injection of ghrelin.
Urine volume did not change after icv and iv injection of ghrelin in comparison with the vehicle (Fig. 3, E and F).
We hypothesize that the primary inhibition of ghrelin on water intake was masked by the orexigenic effects of ghrelin. Then to remove the influence of prandial drinking, we examined the effects of centrally administered ghrelin on water intake without food in comparison with ANP.
Effects of icv injection of ghrelin on water intake and urine volume without food
Water intake was significantly inhibited at 30 and 60 min by icv injection of ghrelin and ANP in comparison with the vehicle (Fig. 4, A and C). The effects of ghrelin were significantly greater than those of ANP when food was not available (Fig. 4, A and C).
|
, B and D).
Effects of icv injection of ghrelin by Fos expression
Many Fos-LI were found in the SON (Fig. 5H), PVN (Fig. 5J), Arc (Fig. 5L), AP, and NTS (Fig. 5N) after icv injection of ghrelin. Only a few Fos-LI were observed in the OVLT (Fig. 5B), MnPO (Fig. 5D), and SFO (Fig. 5F). On the other hand, only a few Fos-LI were observed in corresponding areas in the controls injected icv with a vehicle (Fig. 5, A, C, E, G, I, K, and M).
|
), PVN (Fig. 6J), Arc (Fig. 6L), AP, and NTS (Fig. 6N) after iv injection of ghrelin. Only a few Fos-LI were observed in the OVLT (Fig. 6B), MnPO (Fig. 6D), and SFO (Fig. 6F). On the other hand, only a few Fos-LI were observed in corresponding areas in the controls injected iv with a vehicle (Fig. 6, A, C, E, G, I, K, and M).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The inhibitory effects of centrally administered ghrelin on water intake were as potent as those of ANP in dehydrated rats. Intravenous injection of ghrelin also inhibited water intake in dehydrated rats. The most general explanation for this similarity is that central and peripheral ghrelin act on identical sites within the brain that regulate drinking behavior. Because brain parenchymal neurons are isolated from the systemic circulation by the blood-brain barrier, iv administered ghrelin would only act on the specific control sites in which there is no blood-brain barrier. These specific regions are called circumventricular organs (CVOs), which include the OVLT, SFO, and AP. These are well known to be involved in the regulation of water and electrolyte balance and blood pressure. The neurons in the OVLT and SFO are osmosensitive in rats (27, 28, 29, 30). ANP suppressed water drinking induced by icv injection of AII or water deprivation, as well as the pressor action of AII (26, 27, 31, 32). When ANP was locally administered into the SFO of rats, it inhibited AII-induced water drinking (33). This antidipsogenic action of ANP seems to be mediated via its receptors in the CVOs. We showed that centrally and peripherally administered ghrelin induced a large increase in Fos-LI neurons in the AP/NTS and did not induce Fos-LI neurons in the OVLT, MnPO, or SFO. These sites are known to be involved in drinking behavior and body fluid homeostasis. Previous studies showed that the expression of Fos was observed in the PVN, Arc, AP, and NTS after icv or ip injection of ghrelin (18, 34, 35, 36). In the present study, we showed many Fos-LI neurons were observed in the PVN and Arc as the positive control after icv and iv injection of ghrelin. Rats with AP lesions are known to drink more than intact control rats after overnight water deprivation and after systemic injection of AII (37, 38, 39, 40). It is probable that centrally and peripherally administered ghrelin inhibit water intake partially via acting on the AP/NTS.
In the present study, although water intake increased 120–180 min after icv injection of ghrelin, these increases may have been caused by the feeding-associated drinking (prandial drinking) because food intake markedly increased 60 min after injection (Fig. 3, A and C). The prandial drinking was also observed 30–60 min after iv injection of ghrelin (10 nmol/kg) in the time-course study (Fig. 3, B and D). Furthermore, we showed that the antidipsogenic effects of ghrelin were more potent than those of ANP when food was withdrawn to remove the effects of prandial drinking (Fig. 4, A and C). It is interesting that the antidipsogenic effect of ghrelin precedes the orexigenic effect. Recently Pulman et al. (41) reported that ghrelin increased intracellular calcium in the SFO neurons and had an excitatory effect on them in vitro. The SFO neurons project their axons to the hypothalamic feeding center, including the Arc, SON, PVN, and lateral hypothalamic area (42, 43, 44). We showed that only a few Fos-LI were observed in the SFO after icv and iv injection of ghrelin (Figs. 5, E and F, and 6, E and F). There is a discrepancy between these results and a previous report (41), in which ghrelin had an excitatory effect on SFO neurons. Honda et al. (45) reported that spontaneously active and intermittently burst-firing 
-aminobutyric acid interneurons affect other SFO neurons in rats. One possibility is that the icv and iv injected ghrelin probably excite inhibitory neurons, or a high concentration of ghrelin might have an excitatory effect on the SFO neurons and have a more potent effect on feeding than drinking. Another possibility is that ghrelin may have a direct inhibitory effect on neurons in the SFO and other CVOs such as OVLT because Fos study cannot detect inhibitory effects on neurons.
Intravenous injection of ghrelin (1 nmol/kg) caused a decrease in water intake and an increase in food intake. The icv injection (1 nmol/rat) also caused a significant decrease in water intake and an increase in food intake. The dose of peripherally administered ghrelin in the circulation was very small, compared with that of centrally administered ghrelin in the cerebrospinal fluid because the volume of cerebrospinal fluid is approximately 300 µl in rats. The effects of peripherally administered ghrelin (0.1, 1, and 10 nmol/kg) on water intake and food intake should be more potent than those of centrally administered ghrelin (0.1, 1, and 10 nmol/rat).
Intravenous injection of ANP (2 nmol/rat) significantly decreased water intake in rats deprived of water (26). In the present study, the concentration of peripherally administered ghrelin (1 nmol/kg) is only 0.2–0.3 nmol in each rat. Thus, this dose is very small, compared with ANP (2 nmol/rat). Although we did not compare the effects of iv injection of ghrelin and those of ANP on water drinking, ghrelin is probably a more potent antidipsogen than ANP on water intake. This notion is supported by the present results showing that ghrelin is more potent than ANP when the effect of prandial drinking is removed.
In the present study, icv injection of ghrelin decreased water intake and urine volume. It is possible that decreased water drinking caused an antidiuretic effect; however, the antidiuresis may have been induced by the increased AVP caused by icv injection of ghrelin (20). On the other hand, iv injection of ghrelin decreased water intake but had no effect on urine volume. Ishizaki et al. (20) reported that iv injection of ghrelin (10 nmol/rat) significantly increased the plasma AVP level, and iv injection of ghrelin (1 nmol/rat) did not change the plasma AVP level. In the present study, the concentration of peripherally administered ghrelin (1 and 10 nmol/kg) is only 0.2–0.3 and 2–3 nmol in each rat. Thus, the current doses may be too small for AVP release (20). It is apparent that ghrelin has more potent antidipsogenic effects than renal effects when injected peripherally. We showed that icv injection of ANP had no effect on urine volume in dehydrated rats, which coincides well with the results in sheep that icv injection of ANP had no obvious effects on renal function (46).
The secretion of AVP from pituitary induced by ghrelin may cause the antidipsogenic effect. However, many literatures showed that AVP itself has little influence on thirst. For instance, AVP infused iv (47) or injected into the preoptic area (48) of rats in water balance does not modulate drinking, although AVP inhibits drinking indirectly through water retention in a long run. In the dog, AVP weakly stimulates drinking in water-satiated animals or those with hyperosmotic stimulus (49). Because antidipsogenic action of ghrelin is far greater in terms of potency and efficacy, compared with AVP, it is unlikely that the ghrelin effect is mediated by modulation of AVP secretion.
In the present study, there was a difference in the inhibitory pattern of water intake by ghrelin between the dose-response study and the time-course study. We measured water drinking after injection of ghrelin in other studies; as a result, the effects of ghrelin lasted for 60–90 min in the majority of rats and 120–180 min in the minority of rats. Consequently, this different pattern on water intake perhaps might occur due to individual variation. However, we could suggest that the inhibition of water intake by ghrelin lasted at least 60 min.
Because it is well known that increased arterial pressure inhibits water drinking in several circumstances (50, 51, 52), ghrelin may increase arterial blood pressure and cause the inhibition of water intake. However, centrally and peripherally administered ghrelin decreases mean arterial pressure (MAP) in conscious rabbits (53). Furthermore, centrally administered ghrelin did not change MAP, and peripherally administered ghrelin decreased MAP in rats (20). We also measured MAP in conscious rats after icv administration of ghrelin, but there was no change in the MAP (data not shown).
Wren et al. (54) reported that high doses of ghrelin stimulated the release of various kinds of hormones, including GH, GHRH, ACTH, CRH, neuropeptide Y, AVP, and corticosterone, from the pituitary in rats. Thus, there is a possibility that centrally and peripherally administered ghrelin may cause antidipsogenic effects indirectly on water intake through some hormone, including ANP. Although we could not exclude the effects of some hormones after icv and iv injection of ghrelin, we consider that it is important to show our findings that ghrelin, an orexigenic peptide, may inhibit water intake. In the future, the possible involvement of some hormones, including ANP, on ghrelin-inducing suppression of drinking should be made clear.
It is difficult to reveal the physiological relevance of this responsiveness of centrally and peripherally administered ghrelin on antidipsogenic effects. It is quite puzzling that a hormone like ghrelin with a strong effect on food intake at the same time will inhibit water intake because drinking is strongly associated with feeding. Most species show a close relationship between drinking and feeding (6, 7, 8). About 80% of spontaneous daily water intake is temporally associated with feeding in rats (9, 10). However, why is there a quite unique hormone like ghrelin that has a strong acceleration effect on feeding and at the same time an inhibitory effect on drinking? In the present study, the inhibitory effects on water intake appeared earlier than those on feeding. Further studies are required to demonstrate the relationship between the roles of ghrelin in controlling fluid balance and feeding.
Very recently Samson et al. (55) demonstrated that centrally administered obestatin inhibited water intake in rats. Obestatin has been derived from the same prohormone with ghrelin and reported to exert effects on food intake that oppose those of ghrelin (56). In the future, the central mechanism of the action of ghrelin and obestatin should be clarified in relation with water intake in rats.
The present study showed that centrally and peripherally administered ghrelin potently inhibited water intake in dehydrated rats. In particular, centrally administered ghrelin was more potent than ANP, which is recognized as the most potent antidipsogen known thus far. We showed that centrally and peripherally administered ghrelin activated the neurons in the AP and NTS. Recently Tachibana et al. (57) reported that centrally administered ghrelin acted as an antidipsogenic peptide in chicks. These results suggest that central and peripheral ghrelin might have an important regulatory role in the body fluid balance through the regulation of drinking behavior in rats, as previously reported for the regulation of feeding behaviors. In the future, the central mechanism of the action of ghrelin for inhibiting drinking should be clarified in relation to the induction of feeding in rats.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: All authors have nothing to disclose.
First Published Online January 25, 2007
Abbreviations: AII, Angiotensin II; ANP, atrial natriuretic peptide; AP, area postrema; Arc, arcuate hypothalamic nucleus; AVP, arginine vasopressin; CVO, circumventricular organ; Fos, c-fos protein; icv, intracerebroventricular; LI, like immunoreactivity; MAP, mean arterial pressure; MnPO, median preoptic nucleus; NTS, nucleus of the tractus solitarius; OVLT, organum vasculosum of the lamina terminalis; PB, phosphate buffer; PVN, paraventricular nucleus; SFO, subfornical organ; SON, supraoptic nucleus.
Received July 25, 2006.
Accepted for publication January 12, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Fry and A. V. Ferguson Ghrelin modulates electrical activity of area postrema neurons Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2009; 296(3): R485 - R492. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.05 and 
