| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada N1G 2W1
Address all correspondence and requests for reprints to: Office of the Dean of Graduate Studies, University Center, University of Guelph, Guelph, Ontario, Canada N1G 2W1. E-mail: alastair{at}exec.admin.uoguelph.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Pregnant rats were injected daily through a chronically implanted intracerebroventricular cannula either with a specific monoclonal antibody raised against rat relaxin from days 12–22 of gestation or with an antibody raised against fluorescein as a control. The rats were watched closely from the afternoon of day 20 of pregnancy, and the process of birth was observed. No sign of dystocia was observed in any of the rats in the experiment. Neutralization of endogenous relaxin caused a significant decrease in the length of gestation (505.4 ± 3.1 h) compared with that in rats treated with PBS (524.6 ± 0.5 h) or that in rats treated with a nonspecific antibody (525.9 ± 0.7 h). The time to the onset of delivery was also shorter in the relaxin-neutralized group (507.8 ± 1.1 h) compared with that in either PBS-treated (526.5 ± 0.6 h) or fluorescein antibody-treated (525.3 ± 0.7 h) animals. In contrast, there was no significant effect of the relaxin antibody on length of straining, duration of parturition, delivery interval, live birth rate, or body weight of the neonates. Premature delivery in the relaxin-neutralized group was accompanied by a 24-h advance in the fall in plasma progesterone.
These data support the hypothesis that there may be a central relaxin system that is independent of the peripheral relaxin system. Central relaxin may have a significant physiological role on the timing of pregnancy in the rat, but does not affect the course of labor once it has started.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At the same time, there have been several reports that relaxin acts centrally to affect the release of oxytocin and vasopressin from the brain (4), and Jones and Summerlee (5) suggested that relaxin might act centrally to control oxytocin release and thereby affect the timing of the onset of birth. In the experiments in which rats were passively immunized against endogenous relaxin (2), there was no effect on the timing of birth, only an effect on the process of delivery once the birth process had started.
These two views are not necessarily contradictory. The brain is protected, in many circumstances, from the peripheral circulation by the blood-brain barrier, and peripheral injection of antibody may not have affected the central actions of the hormone. Experiments were performed, therefore, to test whether neutralizing the central effects of relaxin, although not neutralizing the effects of relaxin in the periphery, affects the onset of the timing and the course of delivery in pregnant rats.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of monoclonal antibody
Purified monoclonal antibody to rat relaxin was produced
according to the methodology described by Lao Guico-Lamm et
al. (6). Briefly, rat relaxin, purified by the method of Walsh and
Niall (7), was injected into male BALB/c mice to stimulate antibody
production. At the end of 1 yr, spleen cells were harvested and fused
with NSO mouse myeloma cells. After approximately 2 weeks, the culture
medium was tested for the presence of antirat relaxin activity using
the hybridoma-screening kit enzyme-linked immunosorbent assay (ELISA)
supplied by BRL (Gaithersburg, MD). Nunc immuoplates (Vanguard
International, Neptune, NJ) were coated with rat relaxin (250 ng
relaxin/50 µl PBS/well). Fifty-microliter samples of mouse serum and
hybridoma culture medium were added to each well.
ß-Galactosidase-linked sheep antimouse IgG was used as the second
antibody, with 
-nitro-phenyl-ß-D-galactosidase as the
chromogenic substrate. The second antibody and substrate were used
according to the manufacturer’s directions (BRL). Absorbance at 410 nm
was determined using a Minireader II (Dynatech Laboratories,
Alexandria, VA).
Hybridomas positive for antirat relaxin activity were cloned and grown as ascites tumors in BALB/c mice (8). Antirelaxin antibodies were enriched in the ascites fluid from these mice according to the method described by Voss, Jr., and Watt (9).
Antibody efficacy and purification
The mouse pelvic ligament bioassay, modified by Steinetz
et al. (10), was used to screen the preparations of enriched
antirelaxin antibody. Young virgin female mice (18–20 g) of an outbred
CD1 strain of mice were used. The mice were primed with an injection of
5 µg/0.1 ml estradiol cyprionate in sesame oil sc. Seven days after
priming, mice were divided into four experimental groups. Group I
(relaxin vehicle control) was injected with 0.2 ml 1% benzo-purpurine
(BZP; 1% in distilled water as azo dye repository vehicle), sc. Group
II (relaxin treatment group) was injected with 0.5 µg rat relaxin in
0.2 ml BZP, sc. The dose of relaxin was chosen on the basis of the
experiments carried out by Lao Guico-Lamm et al. (2), who
showed that this dose was sufficient to cause relaxation of mouse
pelvic ligaments. Group III (monoclonal antibody control group) was
treated with 1 mg enriched monoclonal antibody ip, which was shown to
be negative for antirat relaxin activity in the ELISA, followed 30 min
later by sc injection of 0.5 µg rat relaxin in 0.2 ml BZP, sc. Group
IV (antirat relaxin monoclonal group) was treated with 1 mg, ip,
enriched monoclonal antibody that was positive for antirat relaxin
activity, followed 30 min later by sc injection of 0.5 µg relaxin in
0.2 ml BZP, sc. The effects of five different enriched antirat relaxin
monoclonal antibodies were tested. Twenty mice were used in each
treatment group. Twenty-four hours after treatment the mice were killed
by CO2 asphyxiation, and the lengths of their pelvic
ligaments were measured. The mean lengths of the pelvic ligaments were
compared among groups. Logarithmic transformation of the data were
used, and the data were compared using ANOVA. Significant differences
were determined using Bonferroni’s test (11) at the 5% level.
Of the five monoclonal antibodies that showed antirat relaxin activity in the ELISA assay, one showed greater ability to neutralize the effect of rat relaxin in the bioassay. This monoclonal was designated MCA-3. It was purified by hydroxylapatite chromatography, and the protein peaks were analyzed by SDS-PAGE. The monoclonal was freeze-dried and stored at 4 C until use. It was reconstituted in PBS.
Animals
Primiparous Sprague-Dawley rats (Harlan Sprague-Dawley,
Indianapolis, IN; 230–280 g), bred at approximately 90 days, were used
in these studies. The day on which sperm were found in the vagina was
designated day 1 of pregnancy, and 0300 h on day 1 was arbitrarily
chosen to represent the estimated time of insemination. The rats
arrived at the University of Guelph on day 7 of pregnancy and were
caged individually in the Central Animal Facility of the University of
Guelph. Food and water were available ad libitum. The rats
were maintained on a 14-h light, 10-h dark lighting regimen, with
lights on at 0600 h. The ambient room temperature was 18 C.
Surgical preparation of the animals
Pregnant rats were fitted with a chronic indwelling cannula,
placed in the right lateral ventricle of the brain, for injection of
antirat relaxin monoclonal antibody. The technique has been described
in detailed previously (12). In brief, rats were anesthetized on the
morning of day 10 of pregnancy with diazepam (Valium 10, Hoffmann La
Roche, Etobicoke, Canada; 0.8 mg, ip), xylaxine (Rompun, Bayer,
Agriculture Division, Animal Health, Etobicoke, Canada; 4 mg, ip), and
ketamine hydrochloride (Rogarsetic, Rogar STC, London, Canada; 4 mg,
ip).
Anesthetized rats were placed in a stereotaxic frame (SR6, Narishige, Tokyo, Japan) using nontraumatic ear bars to ensure that the ear drums were not ruptured by the procedure. Under aseptic conditions, a small hole was drilled in the skull overlying the right lateral ventricle, and an ethylene oxide-sterilized microcannula (220A, Kopf Instruments, San Francisco, CA) was lowered into position so that the tip lay within the ventricle according to the coordinates: bregma, 2.5 mm lateral to the midline, and 3.0 mm ventral to the surface of the cortex (13). The cannula was sealed in place using cold curing dental acrylic. The scalp was sutured, and the rat was treated with postoperative analgesic (Meperidine HCl, demerol, 2.5 mg every 4 h, im) for 24 h. No postoperative antibiotic was administered. The operations took 2–3 h to complete, and rats were given 15 ml physiological saline, sc, to compensate for fluid loss during the experiments. Rats were nursed for 24 h postoperatively.
Effect of neutralizing endogenous relaxin with monoclonal
antibodies to relaxin on birth
The effect of neutralizing endogenous relaxin within the
ventricular system was tested on the process of birth in rats.
Twenty-two rats were fitted with an indwelling microcannula placed in
the ventricular system, as described above, and left for 48 h to
recover. The rats were divided, at random, into three experimental
groups. Group A rats were injected each morning (between 0900–1000 h)
from days 12–22 of pregnancy with 1 µl monoclonal antibody to rat
relaxin (MCA-3) in PBS, intracerebroventricularly (icv; 1 mg/ml: n
= 8). Group B rats were injected with 1 µl monoclonal antibody to
fluorescein in PBS, icv (n = 7). Group C animals were injected
with 1 µl PBS alone, icv (n = 7).
Observations on birth
Rats were observed continuously from 0600 h on day 21 of
pregnancy until birth occurred. The observers were not aware of the
treatment given to each rat. During the dark phase of the photoperiod,
rats were observed using a 25-watt red light bulb and a flashlight with
a red filter. Observers recorded the time of onset of marked abdominal
straining and the time of delivery of the young and placentas. The
following parameters were determined: duration of pregnancy [time from
presumed insemination (0300 h on day 1) to the time of onset of
straining], duration of straining (from onset straining to the onset
of delivery), onset of delivery (time from insemination to the delivery
of the first pup), duration of parturition (from the onset of straining
to delivery of the last pup or placenta), delivery interval (the
interval between successive deliveries), live birth rate, and birth
weight of the pups (4 h after completed delivery). The postpartum
survival rate of the litters was determined 10 days after birth.
Analysis of data
The mean values of all birth parameters, with the exception of
the mean length of delivery of individual pups and the mean time
interval between deliveries, were compared using ANOVA. Significant
differences (5% level) were determined using Bonferroni’s test (13).
Logarithmic transformation of all birth data were carried out before
analysis. The mean length of delivery of individual pups and the mean
interval between deliveries were compared using a split plot
analysis.
Effects of neutralizing the effects of central relaxin on serum
progesterone concentrations
The effects of neutralizing the central effects of endogenous
relaxin were determined on the antepartum decline in progesterone,
which is an essential component of the normal progression of
parturition (14). Twelve day 10 pregnant rats were fitted with a
chronic indwelling microcannula in the right lateral cerebral
ventricle, as described above, and an indwelling jugular cannula
(polyethylene tubing; id, 0.63 mm; od, 1.4 mm). The free end of the
jugular cannula was routed sc to emerge on the dorsal surface of the
neck close to the head. The cannula was filled with a solution of
heparinized saline (100 U/ml) and sealed with a small sterilized gold
plug. The rats were randomly assigned to receive daily injections of
either 1 µl monoclonal antibody to rat relaxin (MCA-3) in PBS (n
= 4), 1 µl monoclonal to fluorescein in PBS (n = 4), or PBS (1
µl) alone (n = 4). From day 12 of pregnancy, blood samples (200
µl) were collected once a day at the time of icv injection of
monoclonal antibody from manually restrained, conscious rats. On day
20, additional blood samples were collected every 4 h from
0800 h until 0400 h on day 23 (except in animals in which
birth occurred before completion of the sampling regimen). Blood
sampling in the dark phase of the 24-h cycle was carried out under red
illumination as described. After sampling, the cannula was refilled
with heparinized saline and sealed. Blood samples were centrifuged at
15,000 x g for 15 min at 4 C, and the sera were stored
at -75 C until assayed.
Serum progesterone levels were measured using a specific RIA (14) with antiserum (UCB Bioproducts, Liege, Belgium) that cross-reacted 20% with pregnenolone and less than 0.05% with other steroids. The mean inter- and intraassay coefficients of variation were 10.3% and 4.6%, respectively, with a sensitivity of 12.5 pg/ml. The mean serum progesterone levels between the groups were compared using a split plot analysis. Significance was taken at the 5% level.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Injection of relaxin into estrogen-primed mice resulted in a 50–55%
increase in the length of the pelvic ligament compared with that in
controls. This growth-promoting action of rat relaxin was not inhibited
by pretreatment with monoclonal antibody that was negative for antirat
relaxin activity. In contrast, pretreatment with four of the five
monoclonal antibodies that were positive for antirat relaxin activity
negated the action of rat relaxin on the pelvic ligaments. There was no
significant difference between these groups and mice injected with BZP
alone. The results of these experiments are shown in Fig. 1
.
|
. The titration curve for sample 3
(MCA-3) exhibited the most marked slope, indicating that this sample
had the greatest affinity for rat relaxin. Cross-reactivity of this
sample was tested against rat insulin, rat IGF-I, porcine relaxin, and
recombinant human relaxin (Fig. 3). There
was little or no cross-reactivity between MCA-3 and any of the
compounds tested, except rat relaxin.
|
|
Effects of neutralizing central relaxin on the timing of birth
Mean litter size (12.9 ± 0.4 pups) and live birth rate
(98%) were not significantly different among groups. No dystocia was
noted for any of the rats in any of the groups tested. Only one rat in
the PBS control group did not appear to deliver the same number of
placentas as the number of fetuses, and data from this animal were
excluded from the analysis. Complete data for the birth parameters of
the three groups are summarized in Table 1.
|
There were no significant differences in pup survival rate at 10 days postpartum among the three groups. Although a systematic study of nipple morphology was not undertaken, there appeared to be no gross differences in the presence of mammary nipples in the three groups of animals tested.
Effects of neutralizing central relaxin on plasma progesterone
Plasma progesterone levels in pregnant rats are shown in Fig. 4. In rats treated with PBS alone or the
monoclonal antibody to fluorescein, progesterone concentrations ranged
from 90–120 pg/ml on day 20 of pregnancy. Progesterone started to
decline in the dark phase (evening) on day 21 and declined to baseline
progesterone levels (
20 pg/ml) in the light phase (morning) of day
22. These animals delivered on day 23. In contrast, rats in the group
treated with monoclonal antibodies to relaxin had plasma progesterone
levels on day 20 ranging from 90–110 pg/ml, and concentrations fell
over the next 24 h to reach baseline in the light phase (morning)
on day 21. These animals delivered on day 22.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The difference in results between the experiments neutralizing relaxin in the periphery and within the brain may be due to the presence of the blood-brain barrier. The brain is invested, and to some extent protected, by this barrier. It prevents molecules the size of relaxin from passively gaining access to the brain, although there is some evidence that exogenous relaxin can act on hypothalamic centers via the subfornical organ (16, 17) and nucleus medianus (Geddes, B. J., R. S. Poterski, and A. J. S. Summerlee, manuscript submitted). This barrier would also prevent the monoclonal antibody from reaching the brain. In addition, cerebrospinal fluid is separated from the blood and the extracellular fluid space of the brain by an ependymal barrier that also limits the passive passage of molecules as large as relaxin or the antibody to relaxin. Therefore, it is possible in the series of experiments by Sherwood and colleagues (1, 2, 15) that treatment with neutralizing antibodies in the periphery only affected circulating endogenous relaxin and did not cross into the brain to neutralize that actions of central relaxin. In the current experiments the reverse might hold true; central injection of antibody neutralizes central relaxin, but does not affect the action of endogenous hormone in the periphery. It is possible that monoclonal antibody injected centrally leaked out into the peripheral circulation. However, the lack of an observed effect of the antibody treatment on the process of birth and the survival rate of the pups makes this suggestion unlikely. Moreover, the lack of observed effect on nipple morphology supports this idea further. Zhao et al. (19) observed a modest effect on drinking behavior during pregnancy in relaxin-neutralized rats. They argued that this was probably an effect of relaxin at the circumventricular organs; the interface between the blood, the brain, and the cerebrospinal fluid where there are known to be relaxin-binding sites (20, 21). However, Summerlee and colleagues (Summerlee A. J. S., D. J. Hornsby, and D. G. Ramsey, manuscript submitted) reported significant effects on nocturnal drinking behavior in rats passively immunized in the cerebrospinal fluid against endogenous relaxin, which further substantiates a separate role for relaxin in the brain.
In 1984, Summerlee et al. (23) reported that injection of exogenous porcine relaxin suspended reflex milk ejection. Since that time there have been a number of reports that exogenous relaxin acts on the brain. For example, injection of porcine relaxin into the cerebrospinal fluid affects the release of oxytocin and vasopressin (24) and LH (25), and induces a significant dipsogenic response (12). In addition to relaxin-binding sites in the brain (20, 21), there is evidence for relaxin message within specific regions of the brain (26, 27, 28). Recent evidence indicates that relaxin treatment is associated with an increase in the expression of c-fos protein suggestive of a direct action on the brain (29). However, until the current experiments there were no data to confirm whether the central actions of relaxin were of physiological relevance.
The mechanism of the central action of relaxin is not clear. Although there are binding sites for relaxin (20, 21), there have been reports that blocking the central actions of angiotensin II negates the central actions of relaxin on blood pressure (30), drinking (12), and the release of LH (25). Several lines of evidence suggest that the subfornical organ may be important in mediating the effects of relaxin (see review in Ref.4). Summerlee and Wilson (17) reported that lesion of the subfornical organ in pregnant rats resulted in earlier delivery of rats and negated the relaxin-induced delay of birth in animals with artificially elevated plasma relaxin levels at the end of pregnancy. The subfornical organ and related circumventricular organs are critically involved in angiotensin II-mediated actions within the brain (31). Whether relaxin has a direct or an indirect action through angiotensin II on the brain, or both, remains to be resolved, for at least one report (32) suggests that relaxin may have a direct effect on neurosecretory terminals.
Relaxin-neutralized rats not only showed a 24-h advance in the time of delivery, they also showed a 24-h phase shift in the pattern of progesterone secretion at the end of pregnancy. This phase shift needs to be investigated further. In retrospect, it would have been appropriate to have measured plasma relaxin levels in the relaxin-neutralized rats to determine whether there was any change in the profile of relaxin secretion in the peripheral circulation. However, volumes of blood taken from the rats during the experiments were insufficient to be able to measure plasma relaxin as well as progesterone immunoreactivity. In addition, it would also have been appropriate to monitor the peripheral actions of relaxin in other ways, for example by an effect on interpelvic ligament length, but this was not done. The experimental protocol (injection of monoclonal antibody daily from day 12 of pregnancy to the time of delivery) was chosen to replicate the experiments of Hwang et al. (15). In our experiments there was a 24-h phase shift of delivery. As parturition in the rat is closely linked with photoperiod (33), and the fall in progesterone and the rise in relaxin immediately before birth are also linked to photoperiod (34, 35), it would be interesting to know whether the effects observed in our experiments were related to the length of time the action of endogenous rat relaxin in the cerebrospinal fluid was neutralized. For example, would treatment with the monoclonal antibody from earlier in gestation result in a greater advance in the timing of birth or, conversely, would treatment with the antibody from later in gestation not advance delivery? Sherwood and colleagues (35) also reported that litter size influenced serum progesterone and relaxin immunoreactivity levels and the timing of birth; small litter size (<5 pups) resulted in a less clear link between photoperiod and the timing of birth. They concluded that small litter size affected the process of luteolysis, which, in turn, was linked to photoperiod. Litter size was unlikely to affect the results in the current set of experiments, as litters ranged in size from 9–14 pups (mean ± SEM, 12.9 ± 0.4).
These experiments have shown that there is a role for brain relaxin on the timing of the onset of birth, but not on the processes of labor and delivery in the rat. They provide evidence that there may be a central relaxin system that is separate and independent from the classic peripheral actions of relaxin. Together with evidence of other central effects of relaxin, the demonstrated presence of relaxin-binding sites, and synthetic machinery for relaxin within the brain, these data support the idea that relaxin has a series of specific actions on the brain.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 25, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. E Bullesbach, F. R Boockfor, G. Fullbright, and C. Schwabe Cryptorchidism induced in normal rats by the relaxin-like factor inhibitor Reproduction, March 1, 2008; 135(3): 351 - 355. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Feng, N. V. Bogatcheva, A. A. Kamat, A. Truong, and A. I. Agoulnik Endocrine Effects of Relaxin Overexpression in Mice Endocrinology, January 1, 2006; 147(1): 407 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. D. Sherwood Relaxin's Physiological Roles and Other Diverse Actions Endocr. Rev., April 1, 2004; 25(2): 205 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Y. Hsu Cloning of Two Novel Mammalian Paralogs of Relaxin/Insulin Family Proteins and Their Expression in Testis and Kidney Mol. Endocrinol., December 1, 1999; 13(12): 2163 - 2174. [Abstract] [Full Text]
|
|
|
|
|
|
A. J. S. Summerlee, D. J. Hornsby, and D. G. Ramsey The Dipsogenic Effects of Rat Relaxin: The Effect of Photoperiod and the Potential Role of Relaxin on Drinking in Pregnancy Endocrinology, May 1, 1998; 139(5): 2322 - 2328. [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 |
), rat IGF (
),
porcine relaxin (•), and rat relaxin (
). Each data
point represents the mean of at least five determinations. The
data are expressed as a mean percentage of the control value
(absorbance in the presence of hormone divided by absorbance in the
absence of hormone).