INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyrightmaterial had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuingfrom left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographsincluded in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. U·M·I University Microfilms International A Bell & Howelt lntormatron Company 300 North Zeeb Road. Ann Arbor. M148106-1346 USA 313/761-4700 800/521-0600
146
Embed
U·M·I - scholarspace.manoa.hawaii.edu · I would like to thank Drs. N. Hal Richman, Richard S. Nishioka, Masatoshi Mita, and Lisa Helms and Profs. ... 112 116..... ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of thecopy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, ifunauthorized copyrightmaterial had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuingfrom left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproducedxerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directlyto order.
U·M·IUniversity Microfilms International
A Bell & Howelt lntormatron Company300 North Zeeb Road. Ann Arbor. M148106-1346 USA
313/761-4700 800/521-0600
Order Number 9312181
The regulation of prolactin release from the pituitary of thetilapia, Oreochromis mossambicus, by cortisol and environmentalsalinity
Borski, Russell John, Ph.D.
University of Ha.waii, 1992
U·M·!300N. ZeebRd.Ann Arbor, MI48106
~_ ....... _..~_ •.-.....-._-----
THE REGULATION OF PROLACTXN RELEASE FROM THE PXTUXTARY OF
THE TILAPIA, OREOCHROMIS MOSSAMBICUS I BY CORTISOL AND
ENVXRONMENTAL SALXNITY.
A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THEUNIVERSITY OF HAWAII IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
IN ZOOLOGY
DECEMBER 1992
By
Russell John Borski
Dissertation Committee:
E. Gordon Grau, ChairmanChristopher L. Brown
Greg A. AhearnGillian D. Bryant-Greenwood
Fred I. Kamemoto
I dedicate this dissertation to my mother and father,
Kathleen and Herman Borski, for their constant support in
all my endeavors - particularly this one, fish hormones.
iii
-----.__ ._ ...
ACKNOWLEDGEMENTS
I thank everyone of the Fish Endocrinology Laboratory
1986-1992 at the Hawaii Institute of Marine Biology,
especially Greg Weber, Joanne Yoshikawa, Luis Santana, and
Mette Hansen for all their contributions and assistance
throughout the production of this thesis.
I thank Professor Howard A. Bern for his continued
counsel throughout my undergraduate and graduate studies
and for spawning my initial and continued interest in
research.
I would like to thank Drs. N. Hal Richman, Richard S.
Nishioka, Masatoshi Mita, and Lisa Helms and Profs. Chris
Brown, Graham Young, Phillip Helfrich, Yoshitaka Nagahama,
and Tetsuya Hirano for their valuable advice and
discussions which led to the maturation of this
dissertation topic and the continued development of my
science career.
Finally, special gratitude goes out to my advisor,
Professor E. Gordon Grau for teaching me science, for
sharing the successes, and for helping me through the
hurdles of my thesis research. Most of all, I thank
Gordon for the opportunity to have worked and learned
under his guidance.
iv
ABSTRACT
Prolactin is an essential hormone in the freshwater
osmoregulation of the euryhaline teleost fish, tilapia,
Oreochromis mossambicus. Cortisol, on the other hand, is
important in seawater adaptation in the tilapia. The
present studies address how cortisol and environmental
salinity regulate prolactin cell function in the tilapia.
During in vitro incubations, prolactin release is
inhibited in a dose-related manner by cortisol. This
action is mimicked by the synthetic glucocorticoid agonist
dexamethasone but not by other classes of steroids tested.
Perifusion studies indicate that physiological
concentrations of cortisol inhibit prolactin release
within 20 min. cortisol reduces cAMP and ca2+
accumulation in the tilapia pituitary within 15 min, a
time-course similar to the one over which cortisol
inhibits prolactin release. These studies suggest that
the rapid inhibition of prolactin release by cortisol is a
specific glucocorticoid action that is mediated, in part,
by the cAMP and Ca2+ second-messenger systems.
Previous studies have shown that reductions in medium
osmotic pressure, which reflect the blood osmotic pressure
v
--------------
of a tilapia adapting to fresh water, rapidly stimulate
prolactin release, while elevations in medium osmotic
pressure inhibit prolactin release. The present studies
clearly indicate that exposure to reduced osmotic pressure
increases intracellular free ca2+ in single, tilapia
prolactin cells. This hyposmolar-induced elevation in
intracellular free ca2+ occurs within 30 seconds, is
sustained as long as the cells are exposed to hyposmotic
medium, and can be reduced to prestimulated levels by
exposure to hyperosmotic medium. Taken together, these data
suggest a mediatory role for ca2+ in the induction of
prolactin secretion by osmotic pressure.
Consistent with prolactin's role in freshwater
osmoregulation, the quantity of the two tilapia prolactins
(tPRL188 and tPRL177) and their release are greater from the
pituitaries of freshwater tilapia compared with that from
seawater fish. Nonetheless, the relative content of the two
tilapia prolactins (tPRL188/tPRL177) in the pituitaries of
freshwater tilapia was significantly higher (1.5:1) than
that seen in seawater fish (0.75:1). These studies indicate
that the processing of the tPRL188 and tPRL177 may be
differentially sensitive to environmental salinity and that
1. Effects of cortisol on cAMP and 45Ca2+accumulation in RPDs incubated for 15 min ••••••.•• 65
2. Effects of various steroids on PRL releaseduring 18-20 hr incubations in hyposmoticmedia (300 mOsmolal) ••••.•••••.••••••••••••••••••• 67
. ix
LIST OF FIGURES
1. The content of tPRL188 and tPRL177from the RPD of tilapia reared for 7 monthsin fresh water and seawater ••••••••••••.•••.....•• 21
2. The relative content of the larger to smallerPRL expressed as a ratio from the RPD ofFW- and SW-reared tilapia ••.••••••••••.••••.••.•.• 23
3. The effect of osmotic pressure during 18-20 hrstatic incubation on the release of tPRL188and tPRL177 from the RPD of tilapia rearedin fresh water and seawater for 7 months ••.•.•.••• 26
4. The effects of osmotic pressure on the relativequantity of tissue and total tPRL188 andtPRL177 expressed as a ratio during 18-20hr static incubation .....•........................ 28
5. The relative quantity of tPRL188 and tPRL17 7from RPD of tilapia reared in FW for 7 monthsand transferred to SW for an additional 49 daysor transferred to SW for 49 days ••••••••.•.•••..•. 31
6. The relative quantity of tPRL188 and tPRL17 7from RPD of tilapia acclimated to fresh wateror seawater for 21 days and 35 days •••.•.•........ 33
7. The effect of osmotic pressure on the releaseof tPRL188 and tPRL17 7 during 18-20 hrstatic incubation of RPD from tilapiaacclimated to fresh water and seawater for21 and 35 days .. . . . . . . . . . . . . . . . . . . .. 37
8. Effect of cortisol and dexamethasone on PRLrelease from the tilapia RPD during 18-20 hrof static incubation in hyposmotic medium •........ 56
x
9. Effect of cortisol on the release of [3H]PRLfrom the RPD during perifusion incubation ••.•••.. 58
10. Effect of ca2+ ionophore A23187 on theinhibition of PRL release by 50 rtM cortisolin hyposmotic and hyperosmotic media during18-20 hr of static incubation •..•.••••••••••••••. 61
11. Effect of dbcAMP and IBMX on the inhibition ofPRL release by 50 nM cortisol in hyposmoticmedium during 18-20 hr of static incubation •..••• 63
12. cortisol concentrations from the plasma oftilapia reared in fresh water and seawaterfor 4 months 69
13. Effect of hyperosmotic, hyposmotic, andhyposmotic medium containing 300 nMsomatostatin on PRL release from dispersedPRL cells and from the rostral pars distalis ••••• 90
14. Effect of Ca2+ concentrations on therelative intensity of fura-2 fluorescenceexpressed as the ratio of fura-2 bound to Ca2+
to free fura-2 92
15. Effect of hyposmotic medium on the relativeintensity of fura-2 fluorescence in a typicalsilent and spontaneously active PRL cellduring perifusion 95
16. Effect of hyposmotic medium on [ca2+]i
expressed as the relative intensity of fura-2fluorescence 97
17. Effect of hyperosmotic medium on the relativeintensity of fura-2 fluorescence in singlePRL cells e 0 0 0 a 0 0 0 0 0 a 0 CI 0 0 0 CI 0 a _ • • • • • • • • • • •• 100
xi
~-_._.._-_.
18. Effect of depleting ca2+ from the mediumon the hyposmotic-induced increase in relativeintensity of the fura-2 fluorescence in atypical cell 102
systems W-385 sonicator, Heat systems-Ultrasonics, Inc.;
Farmingdale, NY), boiled in a water bath for 3 min, and
frozen at -80°C (Laemmli, 1970; Specker et al., 1985a;
Kelly et al., 1988).
The in vivo RPO content of the PRLs was determined
from freshly dissected RPO transferred directly to the
SOS-buffer. As above, the tissues were sonicated, boiled
and frozen prior to hormone separation by gel
electrophoresis.
SDS Polyacrylamide Gel Electrophoresis
The tilapia PRLs were separated in tissue and medium
samples by SOS-polyacrylamide gel electrophoresis (PAGE)
as described by Laemmli (1970) and Specker et al.,
(1985a). A vertical slab electrophoresis apparatus (Bio
Rad; Richmond, CAl was used. The samples containing the
two tilapia PRLs were stacked in a 4% 37.5:1
17
acrylamide:bis-acrylamide gel and separated in a 15%
37.5:1 acrylamide:bis-acrylamide gel (12 cm long, 0.15 cm
thick). All samples were electrophoresed at 30 rnA
constant current/gel for 4-5 hr using a voltage- and
current-regulated power supply (ISCO; Lincoln, Nebraska).
The gels were stained in 1 liter of Coomassie blue R-250
(10% methanol, 5% acetic acid) on a gyratory platform at
room temperature for 15-18 hr. The gels were destained
(10% methanol, 7% acetic acid, 83% dH20 solution) until
clearly discernable bands were observed and then placed in
a storage solution of 7% acetic acid. Bands of both PRLs
were quantified by using a densitometer and proprietary
software (Hoefer Scientific, San Francisco, CA). All
values are expressed in optical density units (ODU)
measured by the densitometer. The optical densities of
the stained PRL bands were linearly related to the amounts
of the PRLs loaded onto the gel over a range extending
from 0.5 to 3 times the amount of PRL we typically load.
Prolactin release is expressed as a percentage of PRL re
leased into the incubation medium divided by the total
amount of the PRL in the incubation medium plus tissue.
18
statistical Analysis
Differences among means were evaluated using analysis
of variance (Crunch Software; San Francisco, CAl followed
by the least-significant difference test for predetermined
comparisons (Steele and Torrie, 1980). Experiments
employing only two groups were analyzed using the
Student's t-test. statistical differences between the two
tilapia PRLs within a treatment were analyzed using the
paired Student's t-test.
RESULTS
In Vivo Content and In Vitro Release of tPRL 188 and
tPRL177 from the RPD of FW and SW Raised Tilapia
The RPD content of tPRL188 was almost 5X higher in
FW-reared tilapia than in SW-reared fish (Fig. 1; P <
0.001). In a similar way, the RPD content of tPRL177 was
greater in FW-reared fish compared to SW-reared fish (Fig.
1; P < O.OOl). In FW, the RPD contained significantly
more tPRL188 than tPRL177, whereas in sw the reverse was
true (Fig. 1; P < 0.01, P < 0.05).
In order to better illustrate the relative content of
the two tilapia PRLs from FW- and SW-reared fish, we ex-
19
Figure 1. The content of tPRL188 and tPRL177 from the RPDof tilapia reared for 7 months in fresh water andseawater, measured in optical densitometric units (O.D.U.)(mean ± SEMi n = 5-6 RPD). ***, P < 0.001, **, P < 0.01,*, P < 0.05.
20
**
tPRL177
*,
I
***r-,
0....._ ......-
6
5
6'00
4~
x::) 3ciQ...J 2a:Q.
1
21
---------- --------- ._.,-- --
Figure 2. The relative content of the larger and smallerPRL expressed as a ratio (tPRL1SS/tPRL177) from the RPD ofFW- and SW-reared tilapia (mean ± SEMi n = 5-6 RPD). ***,P < 0.001.
22
2.0
1.5r-,,........
..Ja:Q....,....... 1.0
CX)CX)....
..Ja:Q....,
0.5
0.0
-----_.. _-_. ---- .-
·
·***
•
•
•
· II
•
·
FW SW
23
pressed the data as a ratio (tPRL188/tPRL177). This ratio
was found to be strongly dependent on rearing salinity
(Fig 2). The tPRL188/tPRL177 ratio was significantly
higher (~1.5:1) in the RPD of FW-reared tilapia than that
seen in the RPD of SW fish (~O.75:1; p < 0.001).
The tPRL188/tPRL177 ratio shifts in the tilapia RPD
with environmental salinity. Previous studies have shown
that a small reduction in medium osmotic pressure in vitro
stimulates the release of PRL (Grau et al., 1982). This
suggested that the shift in the tPRL188/tPRL177 ratio with
environmental salinity in vivo might result from the
differential effects of osmotic pressure on the
production, release, and/or metabolism of the two PRLs.
I, therefore, undertook in vitro studies directed at
determining whether exposure to physiological changes of
medium osmotic pressure might differentially alter the
pituitary content and/or release of the two tilapia PRLs
from the RPD of FW- and SW-reared tilapia.
As in previous studies, hyposmotic medium stimulated
the release of tPRL177 from RPD of FW fish over levels
observed in hyperosmotic medium during 18-20 hr
incubations (Nagahama et al., 1975; Wigham et al., 1977;
Grau et al., 1982). Here I report that reduced osmotic
pressure also increases the release of tPRL188 in vitro.
Reduced osmotic pressure also increases the release of
24
Figure 3. The effect of osmotic pressure (355 and 300mosm) during 18-20 hr static incubation on the release oftPRL188 and tPRL177 (mean ± SEM) from the RPD (n = 5-6RPD) of tilapia reared in fresh water and seawater for 7months. ***, p < 0.001 and *, p < 0.05.
25
*
GOGO'P'"
..Ja:a.-'P'"
..Ja:a.-
***
eSBelel::l l~d %
26
o
Figure 4. The effects of osmotic pressure (355 and 300mosm) on the relative quantity of tissue (A) and total(medium + tissuei B) tPRL188 and tPRL177 expressed as aratio (tPRL188/tPRL177) during 18-20 hr static incubation.RPD from tilapia reared for 7 months in FW and SW wereutilized for the incubations (mean ± SEMi n = 5-6 RPD).***, P < 0.001.
?7
:{3i 3i0 0
88 ~E E * {IR 8 *(") Cf.) *
co DIIJ ~
0 &q ~ II) C!CIi .... .... 0 C)
.u.~ll:ldll 88~11:IcA
!{3i ]J
~ i ~
~ ~!{-e DIiJ ~
, , , , I
~ &q ~ ~ C)
N .... .... C) 0
.u.~ll:Idll 88~ll:ldl
28
_-_-_00. . .. 0.· ._ . _
both PRLs from RPD of SW fish (Fig. 3). Nevertheless,
during exposure to either hyposmotic or hyperosmotic
medium, the RPD of FW-reared tilapia released
significantly more of both PRLs than did the RPD of SW
reared fish (Fig. 3i P < 0.001, P < 0.05).
We also found that while variations in medium osmotic
pressure did alter PRL release, the tPRL1SS/tPRL177 ratio
of the incubations remained unaltered from values
established in vivo in the rearing salinity. This was
true whether considering the tissue content of the two
PRLs alone (Fig. 4Ai P < 0.001), their medium content
alone (not shown), or both in combination (Fig 4Bi P <
0.001) •
Long-term Acclimation study
After it was determined that the tPRL1SS/tPRL177
ratio shifts in the RPD of FW- compared to SW-reared
tilapia (Fig. 2), my next objective was to determine
whether the tPRL188/tPRL177 ratio reverses when the
rearing salinity is reversed. When tilapia were reared
for 7 months in FW and transferred to SW and held for 49
days, the tPRL188/tPRL177 ratio was reduced (~0.4:1) from
that seen in tilapia that were retained in FW (~1.5:1i
Fig. 5; P < 0.05). The tPRL188/tPRL177 ratio in fish
-_.-. __.._-_.
Figure 5. The relative quantity (tPRL188/tPRL177) oftPRL1 8 8 and tPRL17 7 from RPD of tilapia reared in FW for 7months and transferred to FW for an additional 49 days(FW-FW) or transferred to SW for 49 days (FW-SW). RPD oftilapia reared in SW for 7 months and transferred to SWfor an additional 49 days (SW-SW) or transferred to FW for49 days (SW-FW) were also analyzed for relative quantitiesof the two PRLs. (mean ± SEMi n = 5-6 RPD). **, P <0.01.
30
.~~~------
II
I
I 1
I
2.0
1.5
r--.r--...-..III:Q.-- 1.0CDCD..-
..III:Q.-
0.5
0.0FW·FW
**
FW·SW SW-SW
**
SW-FW
_._-_._.._-----_.- .._"-_. -'---
31
Figure 6. The relative quantity (tPRL188/tPRL177) oftPRL188 and tPRL177 from RPD of tilapia acclimated tofresh water (solid line) or seawater (dashed line) for 21
days and 35 days (mean ± SEMi n = 6 RPD). ***, P < 0.001,**, P < 0.01.
--_._.---_._---- _..._----_. _....--._-
33
transferred to SW (~0.4:1) for 49 days is similar to that
seen in SW-reared fish (~0.5:1). By contrast, the
tPRL188/tPRL177 ratio of fish transferred from SW to FW
for 7 weeks was increased (~1.3:1) compared to that seen
in fish retained in SW (~0.5:1; p < 0.01) and was very
close to the ratio seen in fish that were retained in FW.
Short-term Acclimation Study
The foregoing evidence shows that the tPRL188/tPRL177
ratio in the tilapia RPD depends on the rearing salinity.
This pattern is retained even after 18-20 hr incubations.
In addition, the tPRL188/tPRL177 ratio can be reversed, at
least within 49 days, by reversing the rearing salinity.
This observation led me to question whether the ratio of
the two PRLs is altered during short-term acclimation «
49 days) of fish to different salinities (i.e., FWand
SW).
Figure 6 shows the in vivo tPRL188/tPRL177 ratio in
the RPD from fish that were transferred from FW to SW for
either 21 or 35 days. The tPRL188/tPRL177 ratio in the
RPD of fish transferred from FW to SW (~0.75:1) for 21
days was significantly reduced compared to the ratio
observed in RPD of control fish that were transferred from
FW to FW (~1.2:1; p < 0.01). This pattern was similar to
34
_.._-----_._-
that in fish transferred from FW to SW and held for 35
days (Fig. 6; P < 0.001).
The RPD of tilapia acclimated to FW for either 21 or
35 days released significantly more of both forms of PRL
during exposure to either hyperosmotic or hyposmotic
medium for 18-20 hr than did the RPD from fish acclimated
to SW (Fig. 7B,C; P < 0.01, P < 0.001). Similar to the
pattern seen in incubations of RPD from fish raised for 7
months in FW and SW (Fig. 3), RPD from fish acclimated to
FW and SW for 21 and 35 days released more tPRL188 and
tPRL177 when incubated in hyposmotic medium than when held
in hyperosmotic medium (Fig. 7A-C). There were no
differences between tPRL188 and tPRL177 release from the
RPD under our incubation conditions, regardless of the
acclimation salinity of the fish. As previously observed
with rearing salinity, we also found that while variations
in medium osmotic pressure did alter PRL release, the
tPRL188/tPRL177 ratio of the incubations remained
unaltered from values established from RPD of FW- and SW
acclimated tilapia in vivo. This was true whether
considering the tissue content of the two PRLs alone,
their medium content alone, or both in combination (data
not shown).
Figure 7. The effect of osmotic pressure (355 and 300mosm) on the release of tPRL188 and tPRL177 during 18-20hr static incubation of RPD from tilapia acclimated tofresh water (Day 0, A) and seawater for 21 (B) and 35 days(Ci mean ± SEMi n = 5-6 RPD). ***, P < 0.001, **, P <0.01.
36
-~- ----- - ------ -"---
iI
~
Ij
!I
A Day 0
70, D lPRL188
8O-IlIJtPRL177
j:ii!3O0.
#20
10 IW...:J
355 300Freeh Water
355 300Seawalar
Day 35c
o I ' ......
355 300Fresh Water
O· ***tPRL188 I I
'"jIllIPRL'f Oro I80 I
10
J:-'«:0.30
1/.20
355 300seawater
Day 21B
o I '
355 300Fresh Water
10
70
80
50
I 40
ii! 300.
# 20
DISCUSSION
Clarifying the functional significance of tilapia
having two PRLs has attracted considerable interest since
they were first described by Specker et al., (1984, 1985a,
b). Other than a preliminary observation that tPRL188
alone may promote an increase in the weight and length of
juvenile FW tilapia, no physiological dinstinction between
the two PRL molecules has yet been reported (cf. Specker
et al., 1985a, b). My findings suggest, however, that the
processing of the two tilapia PRLs may be differentially
sensitive to environmental salinity. The relative RPD
content or ratio (tPRL188/tPRL177) of the larger to
smaller PRL molecule shifts from a higher (> 1) value in
FW fish to a lower « 1) value in SW tilapia when fish are
reared from the stage of yolk-sac absorption for 7 months
(Figs. 1, 2). The tPRL188/tPRL177 ratio can be reversed
within 49 days when tilapia reared in FW and SW are later
acclimated to SW and FW, respectively (Fig. 5). Moreover,
this alteration of the ratio occurs not only in tilapia
reared for 7 months in FW and SW, but also appears in
tilapia reared in brackish water and acclimated for 21
days in FW and SW (Fig. 6).
Prolactin cell function has been shown repeatedly to
be augmented in FW tilapia compared with SW tilapia.
38
Dharmamba and Nishioka (1968) showed that both the area of
the RPD and the size of its individual PRL cells, are
greater in the pituitaries of FW-acclimated tilapia
compared with SW-acclimated tilapia. Clarke (1973) also
showed that the RPD of FW tilapia contain more tPRL177
than the RPD of SW tilapia. In addition, details of fine
structure and changes in the rate of 3H-leucine
incorporation suggest that PRL cell activity is enhanced
in FW tilapia compared with SW tilapia (Nagahama et al.,
1975). I report here, for the first time, that the
content of tPRL188, like that of tPRL177, is higher in the
RPD of FW tilapia compared with the RPD of SW tilapia. My_
data are in agreement with the idea that both PRLs play
essential roles in FW adaptation in tilapia and a variety
of other teleost fishes (cf. Clarke and Bern, 1980;
Specker, 1985a).
Both PRLs are present in greater quantitities in RPD
from FW tilapia than from those of SW fish. However, when
comparing FW tilapia (reared and acclimated) with SW
tilapia, I found that the change in RPD content of tPRL188
(~78%) always exceeded that of tPRL177 (~53%; data not
shown; Chi square test, p < 0.05). This suggests that
there is a significantly more pronounced shift in the
quantity of tPRL188 than tPRL177 in the RPD of the tilapia
exposed to alterations in environmental salinities.
39
----- --.-._.
Whether this shift in the tPRL1SS/tPRL177 ratio reflects
modifications in the production, secretion or possibly the
degradation of both or one of the PRLs has been determined
recently. SUbsequent to my investigations, studies show
that more tPRL188 than tPRL177 is synthesized in RPD of
FW-acclimated tilapia compared with RPD of SW-acclimated
fish, whereas the reverse is true for SW fish (personal
communication, Yoshikawa and Grau) •
Not only was the content of the two PRLs in the RPD
of FW tilapia higher ,than that observed in SW fish, but so
too was the ability of FW RPD to release the two hormones
during exposure to either hyperosmotic or hyposmotic
medium in vitro. Nevertheless, exposure to reduced
osmotic pressure augmented the release of both PRLs from
RPD of FW and SW fish over levels observed during exposure
to hyperosmotic pressure. Overall, release of both PRLs
in vitro appeared to be equally sensitive to medium
osmotic pressure (Fig. 3).
The tPRL188/tPRL177 ratio established in vivo in FW
and in SW, respectively, was not altered by changes in
medium osmotic pressure after the RPD were incubated for
18-20 hr in vitro. I found this to be true whether
examining RPD content of the two PRLs alone, their medium
content alone, or both in combination. It would appear
then that the shift in the tPRL18S/tPRL177 ratio seen with
40
a change in environmental salinity does not result from
the differential release of the two tilapia PRLs elicited
directly by physiological changes in osmotic pressure, at
least under conditions described herein.
Studies in our laboratory have shown that SW-reared
tilapia grew almost 2X faster than FW-reared tilapia
(Kuwaye et al., 1991, in press). I considered the
possibility that factors associated with increased growth
rate or fish size might account for the shift in
tPRL1SS/tPRL177 rati~. However, I found no correlation
between the size of individual fish and the
tPRL1SS/tPRL177 ratio, at least when comparing individuals
from a particular salinity (i.e., FWor SW; data not
shown). Furthermore, the tPRL1SS/tPRL177 ratio in the RPD
of fish acclimated to SW for 21 days was 0.75:1, much
lower than its FW controls (ratio = 1.2:1), even though
the average weight of fish in the two groups was similar
(223 and 220 g, respectively). Therefore, I conclude that
the shift in the tPRL1SS/tPRL177 ratios in FW and SW fish
is not likely to be related to differences in fish size or
growth rate. Differences in age also do not appear to be
a critical factor, since all FW- and SW-reared tilapia
were of the same age. Clearly then, it appears that the
alterations in the relative content of the two PRLs in
vitro and in vivo are a result of variations in
41
environmental salinity and are not due to differences in
the age or growth rate of the tilapia being compared.
I was unable to detect any differences between the
release of the two tilapia PRLs under my culture condi
tions. This finding is in agreement with that of Specker
and colleagues (1985a), who showed no variations in the
release or synthesis of the two PRLs during exposure to
reduced osmotic pressure. In vitro studies have shown
that the release of both tilapia PRLs responded similarily
to cortisol, estradiol-17B, urotensin II, vasoactive
sigma) and 3-Isobutyl-1-Methyl-Xanthine (IBMX; sigma) were
added directly to the incubation media at 1 roM and 0.1 mM,
respectively. Media in control experiments received equal
volumes of ethanol or dimethyl sulfoxide.
49
Perifusion Incubation
For perifusion, RPD were preincubated individually
under conditions described above for 48 hours in
hyperosmotic medium containing [3 H]leucine (New England
Nuclear) at 6 ~Ci/ml (1 ~Ci = kBq). The perifusion
apparatus has been described previously (Grau et al.,
1986). Eighteen RPD containing [3 H]leucine-labeled PRL
were transferred to each incubation chamber. Perifusion
medium (28 ± 1°C) was identical to the preincubation
medium, but without [3H]leucine. Before each experiment,
RPD were perifused for 2-3 hr in hyperosmotic medium
until the spontaneous release of [3H]-PRL was stable.
Samples of perifusate were collected at 10 min intervals.
Prolactin release was quantified by a direct counting
method previously described and validated (Grau et al.,
1987). We have found that the magnitude of the [3H]PRL
response to experimental manipulation during perifusion
incubations is proportional to the level of [3H]PRL
release that is established in hyperosmotic medium during
the period prior to stimulation. For this reason, the
average activity of [3 H]PRL in the last 3 fractions
collected from each perifusion chamber immediately before
each experiment was used to normalize [3 H]PRL release.
This resulted in a considerable reduction in the
50
variability of the responses observed among replicate
perifusion incubations.
45Ca2+ Accumulation into the La3+-Resistant Pool of the
RPD
For studies of Ca2+ accumulation, six RPD were loaded
into each perifusion chamber in hyperosmotic medium for 2
hr. The rate of ca2+ accumulation was characterized using
a method previously described and validated (Richman et
al., 1990). In brief, RPD were exposed to 45Ca2+ (12
~Ci/ml) for 15 min, then rinsed with 2 ml of ice-cold
saline (355 mOsm) for 30 sec, and finally, perifused with
ice-cold 4.2 mM LaCl3 for 7 min to displace extracellular
45 Ca3+. Each RPD was then placed in 250 ~l of 1 M NaOH,
sonicated, and neutralized with 250 ~l of 1 N HCl. 45Ca2+
activity was normalized to tissue protein (bicinchoninic
acid protein assay, Pierce Chemical Co.).
cAMP Accumulation in the RPD
Individual tilapia RPD were incubated for 2 hr in 500
~l of hyperosmotic medium. Experimental media (500 ~l)
were then introduced to RPD for 15 min along with IBMX,
which was added to suppress the breakdown of cAMP.
51
._._-_.-,--~-- ----- - .. _-------_.--
Following the experiment, tissues were fixed with 250 ~l
of ice-cold 6% trichloroacetic acid in distilled water.
Tissues were prepared for radioimmunoassay (RIA) by
sonicating in trichloroacetic acid, centrifuging for 3 min
(Beckman microcentrifuge B), and extracting the
supernatant 4 times with 1 ml of water-saturated ethyl
ether. Samples were dried in a 70-80 0C water bath under
an atmosphere of 99.9% N2• All samples were acetylated
and cAMP levels were determined according to the protocol
for the New England Nuclear cAMP RIA kit, previously
validated for use with tilapia RPD (Helms et al., 1991).
Direct measurements of cAMP were closely correlated with
levels normalized to tissue protein (p < 0.001); hence
only direct measurements have been reported (Helms et al.,
1991).
Plasma Cortisol Determinations
Male tilapia were raised from the period of yolk-sac
absorption for 4 months in 700 I oval tanks containing FW
or SW. In order to minimize stress-induced elevations in
plasma cortisol levels I used the blood sampling method of
Young (1986). Tilapia were not approached for at least 12
hours prior to blood sampling. Fish were netted in one
sweep and exposed to a 300 mgjliter solution of tricaine
52
methanesulfonate (Sigma) buffered with sodium bicarbonate.
Blood samples from the severed caudal vein were collected
in heparinized microcapillary tubes. Plasma was isolated
by centrifugation and stored at -80°C. Plasma cortisol
concentrations were determined by radioimmunassay
according to the method of Young (1986), modified for
tilapia.
statistical Analysis
Differences among means were determined using
analysis of variance (Crunch Software, San Francisco)
followed by the least significant difference test for
predetermined comparisons (Steele and Torrie, 1980).
Experiments with only two groups were analyzed using the
unpaired Student's t test.
RESULTS
Effect of Cortisol and Dexamethasone on PRL Release from
the RPD During static Incubation
Cortisol produced a dose-related inhibition of PRL
release from the tilapia RPD during 18-20 hr of in vitro
incubation in hyposmotic medium (Fig. 8A). This effect
53
was significant for doses ranging from 10 nM (p < 0.05) to
1 ~M (p < 0.001), with maximum inhibition occurring at 50
nM (p < 0.001). In an otherwise identical experiment, we
wanted to determine whether cortisol's actions are shared
by its synthetic agonist, dexamethasone. Dexamethasone
was more effective than cortisol, inhibiting PRL release
in a dose-related manner starting at 1 nM (p < 0.001; Fig.
8B).
Time-Course of the Inhibition of PRL Release by Cortisol
The time required for cortisol to reduce PRL release
was investigated. As in previous studies, [3H]PRL release
increased within 10-20 min after introduction of
hyposmotic medium. This response was rapidly suppressed
by 200 nM cortisol (p < 0.001). Release diverged from
control levels immediately and was significantly reduced
within 20 min (p < 0.05; Fig. 9A). cortisol was also
effective at 50 nM, although to a lesser degree and after
a longer delay (50 min; p < 0.05; Fig. 9B).
54
Figure 8. Effect of cortisol (A) and dexamethasone (B) onPRL release from the tilapia RPD during 18-20 hr of staticincubation in hyposmotic medium. Values are expressed asmean ± SEM (for A, n = 12-18 RPDs, except 20 nM dose, n =6; for B, n = 8 RPDs). Asterisks denote valuessignificantly different from control (C) values (*, p <0.05; ***, p < 0.001).
Figure 9. Effect of cortisol on the release of [3H]PRLfrom RPD during perifusion incubation. The last 30 min ofthe initial 2- to 3-hr pe~ifusion in hyperosmotic mediumare shown between -30 and 0 min. Hyposmotic medium wasintroduced to all chambers immediately following time 0maintained until the end of the experiment. cortisol wasintroduced (arrow) into half of the perifusion chamberseither immediately after the 30-min fraction (A) or 50 minfraction (B). The increase in [3 H]PRL release in responseto hyposmotic medium was reduced within 20 min by 200 nMcortisol (A) and within 50 min by 50 nM cortisol (B).Closed squares, hyposmotic medium; closed circles,cortisol. Asterisks denote significant differences (*, p< 0.05; **, P < 0.01). Each point represents the mean ±SEM of 5 perifusion chambers with 18 RPDs per chamber.
Ca2+ Ionophore A23187 Blocks the Inhibition of PRL Release
by cortisol
During 18-20 hr of static incubation, 50 nM cortisol
reduced PRL release in both hyposmotic (p < 0.001) and
hyperosmotic (p < 0.05) media (Fig. 10). This inhibition
of PRL release was blocked by the ca2+ ionophore A23187 (p
< 0.001).
Effect of dbcAMP and IBMX on the Inhibition of PRL Release
by cortisol
Previous studies (Grau et al., 1982) have shown that
the addition of the membrane-permeant derivative of cAMP
(dbcAMP) and the phosphodiesterase inhibitor IBMX, sub
stances which stimulate the cAMP messenger system, can
also increase PRL release from the tilapia RPD. During
18-20 hr of static incubation in hyposmotic medium, 50 nM
cortisol significantly reduced PRL release (p < 0.01; Fig.
11). Dibutyryl cAMP and IBMX completely blocked this
inhibition (p < 0.01).
59
Figure 10. Effect of Ca++ ionophore A23187 (1 ~M) on theinhibition of PRL release by 50 nM cortisol (F) inhyposmotic (300 mosm) and hyperosmotic (355 mosm) mediaduring 18-20 hr of static incubation (mean ± SEM; n = 1112 RPDs). C, control. Asterisks denote significantdifferences (*, P < 0.05; **, P < 0.01; ***, p < 0.001).
60
iI
* I* ""1
* iI
L
IoII')
- ----* i* I
* 1!L
iT
* I* J* L
(
~I
oI't')
.{i
oN
i i
o....
I"-Lo..CX)""-.-at"')ONt"')<{
Lo..<,aa~
aat"')
I"-Lo..CX)""-.-lOt"')lON~<{
Lo..I <,
T 1O1O~
1O1Ot"')
( i0
3S'V3l3t1 ltjd %
61
Figure 11. Effect of dbcAMP (1 mM) and IBMX (0.1 mM) onthe inhibition of PRL release by 50 nM cortisol (F) inhyposmotic medium during 18-20 hr of static incubation (mean± SEMi n = 7 RPDs). Asterisks denote significant differences (**, p < 0.01).
62
SOl •• ••I I
II70
60 i - 11-w 50Ul<J:w-J
40w
+a:::-Ja::: 30a..
~
20
10
0300 300/F 300/F
dbcAMP
63
cortisol Decreases the Accumulation of 45Ca2+ into the
La3+-Resistant ca++ Pool and Reduces the Accumulation of
cAMP in the RPD
The following experiments were directed toward deter
mining whether cortisol might alter cellular Ca2+ metabo
lism. To this end, we investigated the possible effects
of cortisol on the accumulation of extracellular 45ca2+
into the La3+-resistant Ca++ pool of the RPD. The RPD
tissues were perifused in hyperosmotic medium alone or
with 50 nM cortisol. After 15 min, the osmotic pressure
of all media were reduced and 45Ca2+ radiotracer was
added. Cortisol significantly reduced the accumulation of
45c a2+ into the RPD in 15 min (p < 0.01; Table 1). The
rapid actions of cortisol on PRL release and on 45ca2+
accumulation suggested that it might also act on cAMP
metabolism. with IBMX added to suppress phosphodiesterase
activity, exposure of RPD to reduced osmotic pressure for
15 min significantly increased cAMP accumulation in the
RPD over levels in hyperosmotic medium (p < 0.001, data
not shown). This increase was sUbstantially reduced by 50
nM cortisol (p < 0.01; Table 1).
64
Table 1
Effects of cortisol on cAMP and 45Ca2+ accumulationin RPDs incubated for 15 min
Controlcortisol
cAMP,pmol/RPD
0.17 ± 0.01 (5)0.12 ± 0.01*** (5)
45 ca2+,dpm/ILg protein
20.36 ± 1.30 (6)14.05 ± 1.42** (6)
RPDs were incubated in hyposmotic (300 mosm) media in theabsence or presence of cortisol (50 nM). Values representmean ± SEM with the number of RPDs in parentheses. cyclicAMP and 45Ca++ accumulation data were obtained fromseparate experiments. For cAMP, incubation mixturescontained 0.1 mM IBMX. 45 Ca2+ data represent accumulationinto the La3+-resistant Ca++ pool. Double asterisksdenote p < 0.01 for differences from control.
65
specificity
It was of interest to determine whether the
inhibition of PRL release by cortisol was specific to this
SW-adapting hormone or whether the inhibitory response
might be a general steroid hormone effect. Cortisol (50
nM) significantly reduced PRL release from the RPD during
18-20 hr of static incubation in hyposmotic medium (p <
0.001; P < 0.05; Table 2). In contrast, estradiol-17B and
testosterone stimulated PRL release, while cholesterol,
17a,20B-dihydroxy-4-pregnen-3-one, progesterone,
aldosterone, and 11-deoxycorticosterone were without
effect (50 nM; P < 0.05; Table 2).
Plasma Cortisol Concentrations in FW and SW Tilapia
To be sure that the concentrations of cortisol used
in my in vitro studies are physiological I measured
circulating cortisol levels in FW and SW tilapia. The
concentration of cortisol in the plasma of SW tilapia (154
± 23.7 nM) was significantly higher than levels measured
in the plasma of FW tilapia (63 ± 9.9 nM; P < 0.01; Fig.
12) •
66
Table 2
Effects of various steroids on PRL release during 18-20hr incubations in hyposmotic media (300 mOsmolal)
RPD were incubated in media in the absence or presenceof steroids (50 nM). Values represent mean ± SEM with thenumber of RPDs in parentheses. Asterisks indicatesignificant differences from control (*, p < 0.05; **, P <0.01; ***, P < 0.001).
67
----_.----_._--_..... _---
Figure 12. Cortisol concentrations from the plasma oftilapia reared in fresh water (FW) and seawater (SW) for 4months (mean ± SEMi n = 4) .
68
I150 --i
..-..... .1
~ ic 125 J
'-"" :!
200..,
!
175 •
sw
**
HIIiIi
FW
,II
100 ...;II
!75-1
!50~
I
....JoC/1I0::oU
69
-
DISCUSSION
This work shows that physiological concentrations of
cortisol rapidly inhibit the response of the tilapia PRL
cell to reduced osmotic pressure. Prolactin release de
clined immediately from control levels after the introduc
tion of cortisol, becoming significantly reduced within 20
min. This inhibition was accompanied by reductions in
tissue cAMP levels and in 45Ca2+ accumulation, actions
that are compatible with a mediating role for these two
Reno, NV) via one-way stop cocks and polyethylene tUbing.
The manifold output is connected to the input port of the
perifusion chamber by another piece of polyethylene
tubing. The rate of perifusion through the chamber was
maintained at 360 ~l/min by keeping the height of the
85
syringes and volume of all solutions in the syringes
constant throughout the experiment.
single cell fura-2 ratio measurements were made with
a dual excitation spectrofluorometer (ARCM-MIC-N, Spex
Industries, Edison, NJ) interfaced with a Oiaphot-TMO
inverted microscope (CF 40 X oil immersion fluorite
objective, Nikon). The microscope was equipped with
fluorescence optics, a 50 watt halogen illuminator,
epifluorescence illumination, and a quartz nosepiece (for
UV). Excitation light alternated between 340 and 380 nM
(narrow bandpass filters, SPEX) by a computer-controlled
chopper mirror. Fluorescent emission intensity was
transduced every 2 sec by a photomultiplier tUbe focused
on a single PRL cell after it had passed through a 500 nM
emission filter. A pinhole (1 mm) placed in the
epiillumination path restricted the UV illumination to
only the cell of interest.
All data are expressed as the relative intensity of
the ratio of fura-2 fluorescence excited at 340 nm (fura-2
bound to ca2+) to that excited by 380 nm (free fura-2)
from which background (autofluorescence) was subtracted.
Shifts in this ratio (340/380) result directly from
changes in [ca2+]i which are independent of dye
concentration, cell thickness, and absolute optical
86
efficiency of the instrument (Tsien et al., 1985;
Grynkiewicz et al., 1985; cf. Poenie et al., 1986).
Linearity between the relative intensity of the fura
2 fluorescence and [ca2+) in my set-up was examined in a
cell-free system in which fura-2 was added directly to
hyperosmotic medium containing different [ca2+) (Fig. 14).
statistical Analysis.
statistical analyses were performed using PC-SAS (SAS
Institute, Cary, NC).. Prior to analysis, raw data were
log10 or loge transformed to decrease heteroscedasticity.
Measurements of intracellular changes in ca2+ were
analyzed in 10 sec increments by repeated measures ANOVA
(PC-SAS; Huynh and Feldt, 1970). Regression analysis
using the least-squares method was utilized to determine
whether alterations in [ca2+ ) are linearly correlated with
changes in relative fluorescent intensity (340j380) (Steele
and Torrie, 1980).
87
RESULTS
Isolation of single PRL Cells by Dissociation of RPD
Exposure to reduced osmotic pressure augmented PRL
release from single PRL cells and from the RPD (Fig. 13).
Likewise, during exposure to hyposmotic medium, SRIF, a
potent inhibitor of PRL release in tilapia (Grau et al.,
1982; Rivas et al., 1986), dramatically reduced release of
PRL from dissociated PRL cells in a similar manner to that
from intact tissue (Fig. 13).
Effect of [ca2+) on the Relative Intensity (340/380) of
Fura-2 Fluorescence
The linearity of the relation between variations in
[ca2+] and shifts in fura-2 fluorescence (e.g., [ca2+]i)
was confirmed in my set-up by measuring the relative
fluorescence of fura-2 in media containing different
[ca2+]. Increasing loge[Ca2+] ranging from 1-100 ~M
correlated linearly with increases in the fluorescence of
fura-2 (R2=O.99, Fig. 14).
88
Figure 13. Effect of hyperosmotic (355 mOsm), hyposmotic(300 mOsm), and hyposmotic medium containing 300 nMsomatostatin (300/SRIF) on PRL release from dispersed PRLcells and from the rostral pars distalis (RPD; tissue).static incubations were 18-20 hr. (mean ± SEMi n = 9 RPD).
89
25
-o~ 20~C')-~ 15'mc:Q)-.5 10
~~Q) 5a:
f(x) = 4.19 • In (x) + 5.672R = 0.99
10010
[Ca 2+] (uM)
o-+--~-r--r-r-T"I~---,..-,..-r-T"""I'"'''''''''''
1
90
Figure 14. Effect of Ca2+ concentrations (expressed asloge) on the relative intensity of fura-2 fluorescenceexpressed as the ratio of fura-2 bound to Ca2+ to freefura-2 (340/380).
Effects of Hyposmolarity and Hyperosmolarity on [ca2+]i in
single PRL Cells
Previous studies have shown that reductions in medium
osmotic pressure rapidly stimul~te PRL release from the
tilapia RPD within 10-20 min (Grau et al., 1987). In
order to determine whether hyposmotic pressure alters
[ca2+]i in PRL cells within a similar time course over
which it stimulates PRL release we continuously measured
changes in [ca2+]i in dispersed PRL cells during e~~osure
to hyposmotic medium. These measurements show two
distinct types of PRL cells: silent and spontaneously
active (Fig. 15). During exposure to hyperosmotic medium
there is approximately a 2-fold change between minimum and
maximum [ca2+]i for each oscillation in the active cell.
These fluctuations occur repeatedly every 45-60 sec (Fig.
15B). During exposure to hyperosmotic medium the silent
cells showed relatively constant levels of [ca2+]i (Fig.
15A). Of the cells we have measured, 75% were silent and
25% were spontaneously active. Fura-2 measurements of
relative changes in [ca2+]i in both types of cell reveal
that [ca2+]i rapidly rises during exposure to hyposmotic
medium (Figs. 15, 16). In the silent PRL cell, exposure
to hyposmotic medium causes a rapid increase in [ca2+]i
93
Figure 15. Effect of hyposmotic medium on the relativeintensity of fura-2 fluorescence (340/380) in a typicalsilent (A) and spontaneously active (B) PRL cell duringperifusion. Prolactin cells were perifused withhyperosmotic medium (355 mOsm) and then were exposed tohyposmotic medium (300 mOsm). Increases in the relativeintensity of fura-2 fluorescence (340/380) indicateincreases in [ca2+]i.
A4
-0co 3.5('l')-0~('l') 3->-.-:=(I)
2.5eSe-Q) 2>.--caQi 1.5a:
10 120 240 360 480 600 720 840 960
Time (sec)
B4
S' 3.5CDa~ 3('f)->-.-:= 2.5(I)cSc 2-~.-S 1.5Q) .-----..".-------,
expressed as the relative intensity of fura-2 fluorescence(340/380, log transformed). The last minute of theinitial 5-10 min perifusion in hyperosmotic medium (355mOsm) are shown between -60 and -10 sec. Hyposmoticmedium (300 mOsm; closed squares) or hyperosmotic medium(open circles) was introduced to half of the cellsimmediately after the -10 sec point marked by the arrow.Asteriks denote differences between effects of hyposmoticand hyperosmotic medium (**, p < 0.01). Each point represents the mean ± SEM of 9 cells.
that peaks to almost 2 X over levels observed during
incubations in hyperosmotic medium (Figs. 15A, 17A, 18).
Upon exposure of spontaneously active cells to hyposmotic
medium, the amplitudes of the oscillations in [ca2+]i seen
during exposure to hyperosmotic medium are greatly reduced
(Fig. 15A). Compared with exposure to hyperosmotic
medium, however, the average [ca2+]i is higher after
exposure of active cells to hyposmotic medium. On
average, the rise in [ca2+]i induced by hyposmotic
medium in both types of PRL cells becomes statistically
significant within 30 sec (Fig. 16).
The rise in [ca2+]i observed in PRL cells exposed to
hyposmotic medium can last up to at least 20 min (Fig.
17). This elevation in [ca2+]i seen during exposure of
PRL cells to hyposmotic medium can be reversed by
elevating the osmotic pressure of the incubation medium
(Fig. 17). Exposure of PRL cells to hyperosmotic medium
significantly reduced [ca2+]i' within 10 sec, from levels
observed in hyposmotic medium (Figs. 17, 18). These
changes in [ca2+]i in PRL cells exposed to alterations in
osmotic pressure can be replicated in a single recording
of a cell (Fig. 18).
98
Figure 17. Effect of hyperosmotic medium on the relativeintensity of fura-2 fluorescence (340/380; logtransformed) in single PRL cells. Graph A shows a typicalresponse of a PRL cell to hyperosmotic (355 mOsm) andhyposmotic medium (300 mOsm). Two cells were initiallyexposed to hyperosmotic medium and then to hyposmoticmedium. Then one cell was exposed to hyperosmotic medium(A, open arrow) and the other to hyposmotic medium (A,closed arrow). Decreases in the fura-2 fluorescence(340/380) indicate decreases in [ca2+]i. Graph B showsthe average response to hyperosmotic medium of 9 separatemeasurements of single cells. The last minute of theinitial 5-10 min perifusion in hyposmotic medium are shownbetween -60 and -10 sec. Hyperosmotic medium (opencircles) or hyposmotic medium (closed squares) was introduced to half of the cells immediately after the -10 secpoint marked by the arrow. Asteriks denote differencesbetween effects of hyposmotic and hyperosmotic medium (**,p < 0.01). Each point represents the mean + SEM of 9cells.
99
----_._~-- .._-_.
6 A
-~5M-o~
~4
~ene 3S..5~ 2'.as
G)a:1
355
o 200 400 600 800 1000 1200 1400
Time (sec)
**-I I
-- II.. ...It-t-' -.~
....411 , r~
..,""""~- ~~
- t-
I I I I I I I I I
0.7 BS"coM 0.6Q~
~0.5
~
'~ 0.4Sca; 0.3
=;:.! 0.2Q)
a:.9 0
.1
a·60 ·40 ·20 a 20 40 60 80 100 120
Time (see)
-.
100
Figure 18. Effect of depleting Ca2+ from the medium onthe hyposmolar-induced increase in relative intensity offura-2 fluorescence (340/380) in a typical cell. The cellwas initially perifused with hyperosmotic (355 mOsm)medium and then was introduced to hyposmotic (300 mOsm),then to hyperosmotic, then to hyposmotic, and finally tohyposmotic medium absent of Ca2+ and containing 10 mM EGTA(arrow).
elevations in salinity, possess regulatory factors that
are capable of rapidly inhibiting PRL secretion while
simultaneously promoting tolerance to SW. Cortisol is one
such factor. As previously discussed, the content of PRL
in the pituitaries of SW tilapia is markedly reduced
compared with the content seen in pituitaries of FW fish.
This reduction of PRL cell activity may result, in part,
from the actions of cortisol and elevated osmotic
pressure.
To conclude, I have described the interactions
between environmental salinity, osmotic pressure, and
cortisol and their abilities to alter PRL cell function in
the tilapia. During this process I have reported: 1) a
novel mechanism of steroid hormone action, 2) the
differential processing of the two tilapia PRLs, and 3)
the first direct measurements of changes in [ca2+]i in
osmosensitive endocrine cells of known osmoregulatory
output. These studies have layed the groundwork for
future research in the mechanisms that mediate osmotic and
cortisol induced changes in PRL secretion.
115
----------------
REFERENCES
Abraham, M. (1971) The ultrastructure of the cell typesand the neurosecretory innervation in the pituitary ofMugil cephalus L. from fresh water, the sea, and ahypersaline lagoon. Gen. Compo Endocrinol. 17:324-350.
Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K.,and Watson, J.D. (1983) Molecular biology of the cell.Garland PUblishing, New York, NY. pp. 1-1146.
Barry, T.P., and Grau, E.G. (1986) Estradiol-17B andthyrotropin-releasing hormone stimulate prolactin releasefrom the pituitary gland of a teleost fish in vitro. Gen.Compo Endocrinol. 62:306-314.
Berridge, M.J. (1985) The molecular basis of communicationwithin the cell. Sci. Amer. 253: 142-152.
Ben-Jonathan, N., Arbogast, L.A., and Hyde, J.F. (1989).Neuroendocrine regulation of prolactin release. Prog.Neurobiology. 33:399-447.
Biales, G., Dichter, M.A., and Tischler, A. (1977) Sodiumand calcium action potential in pituitary cells. Nature.267:172-174.
Borski, R.J., Helms, L.M.H., Richman, N.H. III, and Grau,E.G. (1991) cortisol rapidly reduces prolactin release andcAMP and 45ca2+ accumulation in the cichlid fish pituitaryin vitro. Proc. Natl. Acad. Sci. 88:2758-2762.
116
Clarke, W.C. (1973) Disc-electophoretic identification ofprolactin in the cichlid teleosts Tilapia and cichlasomaand densitometric measurement of its concentration inTilapia pituitaries during salinity transfer experiments.Can. J. Zool. 51:687-695.
Clarke, W.C., and Bern, H.A. (1980) comparativeendocrinology of prolactin. In: _ Hormonal Proteins andPeptides. C. H. Li, ed. Academic Press, New York, Vol. 8,pp. 105-197.
Cooke, I., Graf, R., Grau, S., Haylett, B., Meyers, D.,and RUben, P. (1989) Crustacean peptidergic neurons inculture show immediate outgrowth in simple medium. Proc.Natl. Acad. Sci. USA. 86:402-406.
Croxton, T.L., Ben-Jonathan, N., and Armstrong, W. McD.(1988) Gonadotropin-releasing hormone induced oscillatorymembrane currents in rat gonadotropes. Endocrinology.123: 1783-1791.
Dauder, S., Young, G., and Bern, H.A. (1990) Effect ofhypophysectomy, replacement therapy with ovine prolactin,and cortisol and triiodothyronine treatment on prolactinreceptors of the tilapia (Oreochromis mossambicus). Gen.Compo Endocrinol. 77:378-385.
Dharmamba, M., and Nishioka, R.S. (1968) Response of"prolactin-secreting" cells of Tilapia mossambica toenvironmental salinity. Gen. Comp. Endocrinol. 6:295-302.
Duval, D., Durant, S., and Homo-Delarche, F. (1983) Nongenomic effects of steroids; interactions of steroid molecules with membrane structures and functions. Biochimicaand Biophysica Acta. 737:409-442.
117
----------_._---- ._----------- --
Dufy, B., Vincent, J.D., Fleury, H., Pasquier, P.O.,Gourdji, D., and Tixier-Vidal, A. (1979) Membrane effectsof thyrotropin-releasing hormone and estrogen shown byintracellular recording from pituitary cells. Science.204:509-510.
Durant, S., Duval, D., and Homo-Delarche, F. (1986)Factors involved in the control of fibroblastproliferation by glucocorticoids: A review. Endoc. Rev.7:254-269.
Foskett, J.K., and Scheffey, C. (1982) The chloride cell:Definitive identification as the salt-secretory cell inteleosts. science. 215:164-166.
Foskett, J.K., Bern, R.A., Machen, T.E., and Connor, M.(1983) Chloride cells and the hormonal control of teleostfish osmoregulation. J. Exp. BioI. 106:255-281.
Foskett, J.K., Gunter-Smith, P.J., Melvin, J., and Turner,R.J. (1989) Physiolo~ical localization of an agonistsensitive pool of Ca + in parotid acinar cells. Proc.Natl. Acad. Sci. USA. 86:167-171.
Foskett, J.K., and Melvin, J.E. (1989) Activation of salivary secretion: coupling of cell volume and [ca2+]i insingle cells. Science. 244:1582-1585.
Gardner, E.A. (1989) The possible role of the thyroidhormones triiodothyronine (T3) and thyroxine (T4> in theosmoregulation of the tilapia, Oreocbromis mossambicus.Masters Thesis, University of Hawaii, USA.
118
Gershengorn, M.C., and Thaw, C. (1985) Thyrotropinreleasing hormone (TRH) stimulates biphasic elevation ofcytoplasmic free calcium in GH3 cells. Further evidencethat TRH mobilizes cellular and extracellular Ca2+ .Endocrinology. 116:591-596.
Gershengorn, M.C. (1986) Mechanism of thyrotropinreleasing hormone stimulation of pituitary hormonesecretion. Ann. Rev. Physiol. 48:515-526.
Gorski, J., and Gannon, F. (1976) Current models ofsteroid hormone action: A critique. Ann. Rev. Physiol.38:425-450.
Grau, E.G., Nishioka, R.S., and Bern, H.A. (1981) Effectsof osmotic pressure and calcium ion on prolactin releasein vitro from the rostral pars distalis of the tilapia,Sarotherodon mossambicus. Gen. Compo Endocrinol. 45: 406408.
Grau, E.G., Nishioka, R.S., and Bern, H.A. (1982) Effectsof somatostatin and urotensin lIon tilapia pituitaryprolactin release and interactions between somatostatin,osmotic pressure, ca2+, and adenosine 3',5'-monophosphatein prolactin release in vitro. Endocrinology. 110:910-915.
Grau, E.G., Shimoda, S.K., Ford, C.A., Helms, L.M.H.,Cooke, I.M., and Pang, P.K.T. (1986) The role of calciumin prolactin release from the pituitary of a teleost fishin vitro. Endocrinology. 119:2848-2855.
Grau, E.G., Ford, C.A., Helms, L.M.H., Shimoda, S.K., andCooke, I.M. (1987) Somatostatin and altered medium osmoticpressure elicit rapid changes in prolactin release fromthe rostral pars distalis of the tilapia, Oreochromismossambicus in vitro. Gen. Compo Endocrinol. 65:12-18.
119
Grau, E.G., and Helms, L.M.H. (1989) The tilapia prolactincell: A model for stimulus-secretion coupling. Fish Phys.Biochem. 7:11-19.
Grynckiewicz, G., Poenie, M., and Tsien, R. Y. (1985) A newgeneration of ca++ indicators with greatly improvedfluorescence properties. J. BioI. Chem. 260: 3440-3450.
Hazama, A. and Okada, Y. (1990a) Involvement of Ca2+induced ca2+ in the volume regulation of human epithelialcells exposed to a hypotonic medium. Biochem. Biophys.Res. Corom. 167:287-293.
Hazama, A., and Okada, Y. (1990b) Biphasic rises incytosolic free Ca2+ in association with activation of K+and Cl- conductance during regulatory volume decrease incultured human epithelial cells. Europ. J. Physio1416:710-714.
Helms, L.M.H. (1988) The role of the cAMP and Ca2+messenger systems in regulating prolactin release from thepituitary of the tilapia, Oreochromis mossambicus. Ph.D.Thesis. University of Hawaii, U.S.A.
Helms, L.M.H., Grau, E.G., and Borski, R.J. (1991)Effects of osmotic pressure and somatostatin on the cAMPmessenger system of the osmosensitive prolactin cell of ateleost fish, the tilapia (Oreochromis mossambicus). Gen.Compo Endocrinol. 83:111-117.
Hirano, T., and Mayer-Gostan, N. (1978) in comparativeEndocrinology, eds. Gaillard, P.J. & Boer, H.H.(Elsevier/North, Holland, Amsterdam), pp. 209-212.
120
---_._..._---_.__.-...._---- ..~._...
Hirano, T. (1980) in Endocrinology 1980, eds. Cumming,I.A., Funder, J.W., Mendelsohn, F.A.O. (Australian Acad.sci., Canberra), pp. 186-189.
Hirano, T. (1986) The Spectrum of Prolactin Action inTeleosts. In: Comparative Endocrinology: Developments andDirections. C. L. Ralph, ed. Alan R. Liss, Inc., New York,pp. 53-74.
Holl, R.W., Thorner, M.O., Mandell, G.L., SUllivan, J.A.,Sinha, Y.A., and Leong, D.A. (1988) Spontaneousoscillations of intracellular calcium and growth hormonesecretion. J. Biol. Chem. 263:9682-9685.
Hua, S.Y., and Chen, ·Y.Z. (1989) Membrane receptormediated electrophysiological effects of glucocorticoid onmammalian neurons. Endocrinology 124:687-691.
Huynh, H., and Feldt, L.S. (1970) Conditions under whichmean square ratios in repeated measurements designs haveexact F-distributions. J. Amer. statistical Assoc.65: 1582-1589.
Hwang, P.P., Sun, C.M., and Wu, S.M. (1989) Changes ofplasma osmolality, chloride concentration and gill Na-KATPase activity in tilapia Oreochromis mossambicus duringseawater acclimation. Mar. Biol. 100:295-299.
Katzenellenbogen, B.S. (1980) Dynamics of steroid hormonereceptor action. Ann. Rev. Physiol. 42:17-35.
Ingram, C.D., Bicknell, R.J., and Mason, W.T. (1986)Intracellular recordings from bovine anterior pituitarycells: modulation of spontaneous activity by regulators ofprolactin secretion. Endocrinology. 119:2508-2515.
121
Kelley, K.M., Nishioka, R.S., and Bern, H.A. (1988) Noveleffects of vasoactive intestinal polypeptide and peptidehistidine isoleucine: inhibition of in vitro secretion ofprolactin in the tilapia (Oreochromis mossambicus). Gen.compo Endocrinol. 72:97-106.
Kelley, K.M., Nishioka, R.S., and Bern, H.A. (1990) Invitro effect of osmotic pressure and cortisol on prolactincell physiology in the coho salmo (Oncorhynchus kisutch)during parr-smolt transformation. J. Exp. Zool. 254: 7282.
Koch, B., Lutz-Bucher, B., Briaud, B., and Mialhe, P.(1978) Specific interaction of corticosteroids withbinding sites in the plasma membranes of rat anteriorpituitary gland. J. ·Endocrinol. 79:261-273.
Kuwaye, T.T., Okimoto, D.K., Shimoda, S.K., Howerton,R.D., Lin, H.R., Peter, K.T., and Grau, E.G. Effect of17a-methyltestosterone on the growth of the euryhalinetilapia, Oreochromis mossambicus, in fresh water and inseawater. Aquaculture. In Press.
Laemmli, U.K. (1970) Cleavage of structural proteinsduring the assembly of the head of bacteriophage T4.Nature. 227:680-685.
Lamberts, S.W.J., and MacLeod, R.M. (1990) Regulation ofprolactin secretion at the level of the lactotroph.Physiol. Rev. 70:279-318.
Leng, G., Mason, W.T., and Dyer, R.G. (1982) Thesupraoptic nucleu as an osmoreceptor. Neuroendocrinology.34:75-82.
Letourneau, P.C. (1975) possible roles for cell-tosubstratum adhesion in neuronal morphogenesis. Dev. Biol.44: 77-91.
Leung, K., and Munck, A. (1975) Peripheral actions ofglucocorticoids. Ann. Rev. Physiol. 37:245-272.
Levitzki, A. (1986) B-Adrenergic receptors and their modeof coupling to adenylate cyclase. Physiol. Rev. 66:819854.
Lewis, D.L., Goodman, M.B., John, P.A.S., and Barker, J.L.(1988) Calcium current and fura-2 signals in fluorescenceactivated cell sorted lactotrophs and somatotrophs of ratanterior pituitary. Endocrinolgy. 123:611-621.
MacDonald, D.J., and McKeown, B.A. (1985) Compo Biochem.Physiol. 80A:553-557.
Margaroli, A., Vallar, L., Elahi, F.R., Pozzan, T., Spada,A.~ and Meldolesi, J. (1987) Dopamine inhibits cytosolicCa + increases in rat lactotroph cells. J. BioI. Chern.262:13920-13927.
Melmed, S. (1984) Insulin suppresses growth hormone secretion by rat pituitary cells. J. Clin. Invest. 73:14251433.
Nagahama, Y., Nishioka, R.S., and Bern, H.A. (1973)Responses of prolactin cells of two euryhaline marinefishes, Gillichthys mirabilis and Platichthys stellatus toenvironmental salinity. Z. Zellforsch. 136:153-157.
123
Nagahama, Y., Nishioka, R.S., Bern, H.A., and Gunther,R.L. (1975) Control of prolactin secretion in teleosts,with special reference to Gillichthys mirabilis andTilapia mossambica. Gen. Compo Endocrinol. 25:166-188.
Negulescu, P.A., and Machen, T.E. (1990) Intracellular ionactivites and membrane transport in parietal cellsmeasured with fluorescent dyes. Methods in Enzymology.192:38-81.
Nicoll, C.S., Wilson, S.W., Nishioka, R.S., and Bern, H.A.(1981) Blood and pituitary prolactin levels in tilapia(Sarotherodon mossambicus; Teleostei) from differentsalinities as measured by a homologous radioimmunoassay.Gen. Compo Endocrinol. 44:365-373.
Nicoll, C.S. (1982) Role of prolactin in water and electrolyte balance in vertebrates. In Current Endocrinology:Prolactin. R. Jaffe, ed. Elsevier/North Holland, New York,pp. 127-166.
Nishioka, R.S., Grau, E.G., and Bern, H.A. (1985)vitro release of growth hormone from the pituitaryof tilapia (Oreochromis mossambicus). Gen. CompoEndocrinol. 60:90-94.
Ingland
Nishioka, R.S., Kelley, K.M., and Bern, H.A. (1988)Control of prolactin and growth hormone secretion inteleost fishes. Zool. Sci. 5:267-280.
Olivereau, M., Chambolle, P., and Dubourg P. (1981) Ultrastructure of prolactin cells in the eel kept in freshwater, deionized water and normal and concentrated seawater. BioI. Cell. 42:153-166.
124
--_..__.-- - ..._--- .. _.---- ..
Pickford, G.E., and Phillips, J.G. (1959) Prolactin, afactor in promoting survival of hypophysectomizedkillifish in fresh water. Science 130: 454-455.
Pietras, R.J., and Szego, C.M. (1977) Specific bindingsites for oestrogen at the outer surface of isolatedendometrial cells. Nature. 265:69-72.
Poenie, M., Alderton, J., Tsien, R. Y., and Steinhardt,R.A. (1986) Changes of free calcium levels with stages ofthe cell division cycle. Nature. 315:147-149.
Pragar, D., Yamashita, S., and Melmed, S. (1988) Insulinregulates prolactin secretion and messenger ribonucleicacid levels in pituitary cells. Endocrinol. 122: 29462952.
Rasmussen, H., and Barrett, P.Q. (1984) Calcium messengersystem: an integrated view. Physiol. Rev. 64:938-984.
Rasmussen, H., Kojima, I., Kojima, K., Zawalich, W., andApfeldorf, W. (1984) Calcium as intracellular messenger:sensitivity modulation, C-kinase pathway, and sustainedcellular response. Advances in cyclic Nucleotide Research.18: 159-193.
Richman, N.H., Helms, L.M.H., Ford, C.A., Beneshin, C.,Pang, P.K.T., Cooke, I.M., and Grau, E.G. (1990) Effectsof depolarizing concentrations of K+ and reduced osmoticpressure on calcium accumulation by the rostral parsdistalis of tilapia, Oreochromis mossambicus. Gen. CompoEndocrinol. 77:292-297.
Pang, P.K.T., and Grau, E.G. (1991) The loss of 45Ca +associated with prolactin release from the tilapia(Oreochromis mossambicus) rostral pars distalis. Gen.compo Endocrinol. 83:56-67.
Ringold, G.M., Yamamoto, K.R., Bishop, J.M., and Varmus,H.E. (1977) Glucocorticoid-stimulated accumulation ofmouse mammary tumor virus RNA: Increased rate of synthesisof viral RNA. Proc. Natl. Acad. Sci. USA. 74:2879-2883.
Rivas, R.J., Nishioka, R.S., and Bern, H.A. (1986) Invi~ro effects of somatostatin and urotensin II onprolactin and growth hormone secretion in tilapia,Oreochromis mossambicus. Gen. Compo Endocrinol. 63:245251.
Rotsztejn, W.H., Dussaillant, M., Nobou, F., and Rosselin,G. (1981) Rapid glucocorticoid inhibition by vasoactiveintestinal peptide-induced cyclic AMP accumulation andprolactin release in rat pituitary cells in culture.Proc. Natl. Acad. Sci. USA. 78:7584-7588.
Rubin, R.P. (1982) Calcium and Cellular Secretion. PlenumPress, N.Y. pp. 1-276.
Rubin, O.A., and Specker, J.L. (1988) Regulation of testisfunction in tilapia by homologous prolactins. Am. Zool.28: 55A.
Orchinik, M., Murray, T.F., and Moore, F.L. (1991) Acorticosteroid receptor in neuronal membranes. Science.252:1848-1851
126
Sadler, S.E., and Maller, J.L. (1985) in Advances inCyclic Nucleotide and Protein Phosphorylation Research,eds. Cooper, O.M.F., Seamon, K.B. (Raven, New York), Vol.19, pp. 179-194.
sato, N., Wang, X., Greer, M.A., Greer, S.E., and McAdams,S. (1990a) Evidence that ethanol induces prolactinsecretion in GH4C1 cells by producing cell swelling withresultant calcium influx. Endocrinology. 127:3079-3086.
Sato, N., Wang, X., Greer, M.A., Greer, S.E., McAdams, S.,and Oshima, T. (1990b) Medium hyposmolarity stimulatesprolactin secretion in GH4C1 cells by inducing an increasein cytosolic free ca2+ . Endocrinology. 127:957-964.
Schlegel, V., Winiger, B.P., Mollard, P., Vacher, P., Wuarin, F., Zahnd, E.R., Wolleheim, C.B., and DUfY,B. (1987)Oscillations in cytosolic ca2+ in pituitary cells due toaction potentials. Nature. 329:719-721.
Specker, J.L., Ingleton, P.M., and Bern, H.A. (1984)Comparative physiology of the prolactin cell. In:Prolactin Secrection: A MUltidisciplinary Approach. F.Mena and C. Valverde, eds. Academic Press, Orlando, FL.
Specker, J.L., King, 0.5., Nishioka, R.S., Shirahata, K.,Yamaguchi, K., and Bern, H.A. (1985a) Isolation andpartial characterization of a pair of prolactins releasedin vitro by the pituitary of a cichlid fish, Oreochromismossambicus. Proc. Natl. Acad. Sci. 82:7490-7494.
Specker, J.L., King, 0.5., Rivas, R.J., and Young, B.K.(1985b) Partial characterization of two prolactins from acichlid fish. In: Prolactin, Basic and ClinicalCorrelates. R.M. MacLeod, M.O. Thorner, U. Scapagnini,eds. Liviana Press, Padua, Vol. 1, pp.427-436.
127
Specker, J.L., Brown, P.S., and Brown, S.C. (1989) Unequalactivities of the two tilapia prolactins in the wholeanimal transepithelial potential bioassay using the redeft. Fish Physiol. Biochem. 7:119-124.
Steele, R.G.D., and Torrie, J.H. (1980) Principles andProcedures of Statistics: A Biometrical Approach. McGrawHill, New York.
Stojilkovic, S.S., and eatt, K.J. (1992) Calcium oscillations in anterior pituitary cells. Endocrine Rev. 13:256280.
Tan, C.H., Wong, L.Y, Pang, M.K., and T.J. Lam (1988)Tilapia prolactin stimulates estradiol-17B in vitellogenicoocytes of the guppy Poecilia reticulata. J. Exp. Zool.248:361-364.
Taraskevich, P.S., and Douglas, W.W. (1978) Catecholaminesof supposed hypophysiotropic function suppress actionpotentials in prolactin cells. Nature. 276:832-834.
Tashjian, A.H., Bancroft, F.e., and Levine, L. (1970)Production of both prolactin and growth hormone by clonalstrains of rat pituitary tumor cells: Differential effectsof hydrocortisone and tissure extracts. J. Cell. BioI. 47:61-70.
Towle, A., and Sze, P. (1983) Steroid binding to synapticplasma membrane: differential binding of glucocorticoidsand gonadal steroids. J. Steroid Biochem. 18:135-143.
Tsien, R.Y., Rink, T.J., and Poenie, M. (1985) Measurementof cytosolic free Ca2+ in individual small cells usingfluorescence microscopy with dual excitation wavelengths.eell Calcium. 6:145-157.
1~8
Ucker, O.S., and Yamamoto, K.R. (1984) Early events in thestimulation of mammary tumor virus RNA synthesis by glucocorticoids. J. BioI. Chem. 259:7416-7420.
Wigham, T., Nishioka, R.S., and Bern, H.A. (1977) Factorsaffecting in vitro activity of prolactin cells in theeuryhaline teleost sarotherodon mossambicus (Til apiamossambica). Gen. Compo Endocri~ol. 32:120-131.
winiger, B.P., Wuarin, F., Zahnd, G.R., Wollheim, C.B.,and Schlegel, W. (1987) Single cell monitoring ofcytosolic calcium reveals subtypes of rat piuitary cells(GH3 cells). Cell Calcium. 5:223-230.
Yamaguchi, K., Specker, J.L., King, O.S., Yokoo, Y.,Nishioka, R.S., Hirano, T., and Bern, H.A. (1988) Completeamino acid sequences of a pair of fish (tilapia)prolactins, tPRL177 and tPRL188• J. BioI. Chem. 263:91139121.
Young, G. (1986) cortisol secretion in vitro by theinterrenal of coho salmon (Oncorhynchus kisutch) duringsmoltification: Relationship with plasma thyroxine andplasma cortisol. Gen. Compo Endocrinol. 63:191-200.
Young, P.S., McCormick, S.D., Demarest, J.R., Lin, R.J.,Nishioka, R.S., and Bern, H.A. (1988) Effects of salinity,hypophysectomy, and prolactin on whole-animaltransepithelial potential in the tilapia Oreochromismossambicus. Gen. Compo Endocrinol. 71:389-397.