Fish Physiology and Biochemistry vol. 12 no. 3 pp 193-202 (1993) Kugler Publications, Amsterdam/New York
Changes in intestinal fluid transport in Atlantic salmon (Salmo salar L) during parr-smolt transformation Philip A. Veillettel, Ronald J. White 2 , and Jennifer L. Speckerl 1 Department of Zoology, University of Rhode Island, Kingston, RI 02881, U.S.A.; 2 Department of Biology, Eastern Washington University, Cheney, WA 99004, U.S.A.
Accepted: April 23, 1993 Keywords: intestinal fluid transport, parr-smolt transformation, seawater adaptation, Atlantic salmon
Abstract We examined changes in fluid transport by the intestine of Atlantic salmon (Salmo salarL) undergoing parrsmolt transformation during springtime. In vitro measurements of fluid transport rate (J,) across noneverted middle and posterior intestinal sac preparations were made in late April and early June 1990 and from February through June 1991 for juvenile smolting fish. Intestinal J, was also compared between parr- and smolt-stage salmon in both years. To evaluate the osmoregulatory role of the intestine, Jv was measured for smolts adapted to seawater and their cohorts remaining in fresh water. The middle intestine of smolting fish underwent a significant decrease in fluid transport during the springtime, while posterior intestinal Jv significantly increased. Parr-stage fish decreased Jv in the middle intestine during springtime similar to smolts. However, the posterior intestinal J, of smolts showed a significant increase over the parr around the peak smolt period in both years. Seawater-adapted smolts generally exhibited posterior intestinal Jv approximately double that of freshwater cohorts. A decrease over time shown for the middle intestine, together with the increased Jv in the posterior intestine preceding and after seawater entry, suggests the development of a functional regionalization during parr-smolt transformation, with the posterior intestine taking on increased importance in osmoregulation in seawater.
Introduction
(Brown 1957). Glomerular filtration rate decreases, gill Na4' , K+-ATPase activity increases, and intes-
Smith (1932) demonstrated that teleosts in seawater survive dehydration by drinking water and reducing urine production. In anadromous salmonids, many preadaptive physiological changes occur during the parr-smolt transformation which prepare the fish for seawater while they are still in fresh water (McCormick and Saunders 1987; Hoar 1988; Specker 1988). Salinity tolerance develops during a fast-growing, freshwater phase of the smolting parr
tinal fluid uptake increases in the springtime prior to seawater entry (Holmes and Stainier 1966; Zaugg and McLain 1970, 1972; Collie and Bern 1982). Finally, in response to seawater entry, drinking rate increases (Usher et al. 1988). The intestine is a major site of salt and water balance in vertebrates, but much remains to be known of its role in seawater adaptation of fishes. In euryhaline fishes, intestinal absorption increases
Correspondence to: Philip A. Veillette, Department of Zoology, University of Rhode Island, Kingston, RI 02881, USA; Fax (401) 792-4256
194 during adaptation to seawater and this process is mediated by corticosteroids (Collie and Hirano 1987). In coho salmon (Oncorhynchus kisutch), the posterior intestine attains a rate of fluid absorption during the smolt-period in fresh water which is as high as the rate attained when they are adapted to seawater (Collie and Bern 1982). Atlantic salmon (Salmo salar) show a marked bimodal distribution in growth in their first year and only some become 1 + smolts (Thorpe 1977; Thorpe et al. 1980; Saunders et al. 1982). We were interested in whether the posterior intestine is part of the hypoosmoregulatory mechanisms that preadapt Atlantic salmon to seawater as has been shown for coho salmon (Collie and Bern 1982) and if so whether parr and smolts differed during the springtime. We examined concurrent changes in the middle intestine and identified regional differences in the pattern of changes that occur in parr and smolts during springtime. Furthermore, we followed changes in smolts after entry into seawater.
Materials and methods Developmental changes Juvenile Atlantic salmon were used to monitor developmental changes in parr and smolts. The fish were reared at the North Attleboro National Fish Hatchery (NFH) in 2.5 m wide by 1.5 m deep raceways with flow-through water at 100 C. They were fed 2% of their body weight each day over several feedings. In November they were separated into two size classes and subsequently maintained in adjacent sections of the same raceway separated by a wire screen. Fish used for measuring intestinal fluid transport were transported to the University of Rhode Island salmon facility at the Biological Sciences Center in late April and early June 1990 and in February, March, early April, late April, May and June 1991. Fish were held in flow-through well water at 8-15°C for 2-5 days without food prior to sampling. Lower mode parr have the tendency to enter the faster growing phase and become upper mode smolts (Kristinsson et al. 1985). Because of this,
parr were selected by taking smaller, less silvery, lower mode fish which exhibited parr marks and yellowing of the pectoral fins and the opercular and ventral regions. Smolting fish were chosen from the upper growth mode. Rate of fluid transport (J) across the middle and posterior intestine was measured for 12 smolting fish at sample times in February through May (except late April posterior intestine, n = 11) and for 8 smolting fish in June 1991. Intestinal Jv was also measured for middle and posterior regions for 12 parr in May 1991. In 1990, measurements were made on the middle and posterior intestine of 11 - 12 parr and 11- 12 smolting fish at both sample times. We have chosen to use the term middle intestine as used by Colin et al. (1985), rather than anterior intestine, so that differentiation may be made between the intestinal pyloric ceca area and the non-pyloric ceca area anterior to the ileorectal valve. The fish were anesthetized in 2-phenoxyethanol and killed by a blow to the head or by a dorsal cut through the spinal cord. Body weight, fork length, and appearance were recorded. Fish were then placed on ice until removal of the intestine. The maximum duration between these procedures and the removal of the intestine from all fish and their placement in aerated physiological saline was 1.5h. The body cavity was opened and the posterior intestine was cut free of the body wall. The intestine was removed by gently tearing the mesenteries until they hung free from the peritoneal cavity. An effort was made to prevent the intestine from becoming contaminated with external body secretions. The middle intestine was cut at the junction just posterior to the attachment of the pyloric ceca. Adhering mesenteries and blood vessels were then removed and the middle and posterior segments were separated at the prominent ileorectal valve. The lumen was rinsed with Ringer (140 mM NaCl, 15 mM NaHCO 3, 1.5 mM CaC12, 1 mM KH 2PO 4 , 0.8 mM MgSO 4, 2.5 mM KCI, 10 mM glucose and 5 mM N-2-Hydroxyethylpiperazine-N ' -2-ethanesulfonic acid (HEPES) buffer, pKa = 7.5), adjusted to pH 7.8 and 300 + 10 mmol-kg - ' at 250 C. Osmolality of buffer solutions was measured on a vapor pressure osmometer (Wescor, model 5100).
195 Measurement of Jv After rinsing out the contents of the intestine with buffered Ringer, the most distal end was tied off with light waxed or unwaxed, unflavored, flossing nylon. A blunted hypodermic needle attached to a 20 or 30 ml plastic syringe was then inserted into the other end of the intestine and a single tie of nylon was made about the intestine and needle. The intestine was then inflated with Ringer until it began to slip off the shank of the needle. At this point the filling was stopped, the intestine was slid off the needle and the knot was pulled tight. Despite possible variability in the pressures among the intestinal sacs, Collie and Bern (1982) have shown that noneverted intestinal sacs from coho salmon do not have significant differences in the slope of sac weight loss with time for hydrostatic pressures between 0-20 cm H 20. The sac was then suspended in 35-50 ml of Ringer at 13 0C (1990) or 14°C (1991) and bubbled vigorously with 95%°7 02/5% CO 2. The sac was pre-incubated lh prior to measuring Jv. After pre-incubation the sac was weighed to the nearest mg at 10 or 15 min intervals for at least lh. Collie and Bern (1982) have shown with histological examination that no gross morphological changes are apparent after 7h incubation of the non-everted intestinal sac preparation. The slope of the regression line for sac weight loss plotted against time represents rate of water loss. This was normalized by dividing by the surface area of the sac to give a Jv in plcm-2-h 1. Surface area was determined by spreading the sac serosal side down on Truwax Baseplate wax (Dentsply International Inc.), outlining it with #7 jewelers forceps then filling the groove with neutral decolorizing charcoal (Norit Co., Inc.). The outline of the sac was then traced with a Scriptel SPD Graphics Tablet digitizer (Jandel Corp.).
Seawater adaptation In July 1990, 1 + freshwater Atlantic salmon (reared as previously described) were either kept in fresh water at the salmon facility or acclimated to seawater at the University of Rhode Island's
Graduate School of Oceanography (GSO). Freshwater fish were maintained in aerated, flowthrough well water with temperature varying seasonally from 8-16°C. Salmon at the GSO were transferred to 9%0 seawater. Immediately after transfer, we began adding full strength, Narragansett Bay seawater. The tank was maintained with aeration and flow-through seawater at a salinity of 30 + 2%0. Water temperature varied seasonally from 5--18°C. Prior to sampling, fish were starved for 2-7 days. Posterior intestinal J, was measured on 12 freshwater- and 12 seawater-adapted fish in late August 1990. Middle and posterior intestinal Jv were measured on either 12 freshwater fish or 6 or 12 seawater-adapted fish at selected times from November 1990 through March 1991. In addition, measurements were made on 6 seawater stunts in late March 1991. Sampling of fish and preparation of intestinal sacs were as previously described; with the exception that an alternate physiological Ringer (composed of 150 mM NaCl, 2.5 mM KCI, 3.5 mM CaCI2, 1 mM MgC1 2, 7 mM NaHCO 3, 0.7 mM NaHPO 4, 5 mM HEPES and 10 mM glucose, adjusted to pH 7.8 and 310-316 mmol.kg - 1) was used for intestinal sac preparations from seawateradapted salmon at all times and freshwater salmon in late August 1990. Fluid transport was measured at 16°C in August 1990 and 14°C at all other times.
Inhibition by ouabain and N 2 The effect of Ringer containing 10 - 4 M ouabain or of N2 in place of 0 2/CO 2 to the serosal side of posterior intestinal sac preparations was determined. Intestinal sacs were preincubated for 60 min; then fluid transport was measured for 50 min. Following this, either ouabain was added to the Ringer or N2 treatment was started. The intestinal sacs were allowed to incubate under these conditions for 20 min, followed by a second J, determination over 50 min. The N 2 inhibition experiments 1.7 were performed on April 10, 1990 using 21.7 g (mean SEM; n = 8) parr. The ouabain inhibi-
196 Table 1. Size and stage characteristics of Atlantic salmon used for measuring intestinal fluid transport in winter and spring 1991 Month
Growth mode
n
Body weight (g)
Fork length (cm)
Condition factor*
Silvering**
Parr marks***
Yellowing****
February March Early April Late April May
Upper Upper Upper Upper Upper Lower Upper
12 12 12 12 12 12 8
90.4+ 6.3 119.0+4.3 137.4+4.8 112.3+4.6 142.2+8.1 62.8 4.8 136.1 +9.1
20.2 + 0.4 22.3+0.3 23.6+0.2 22.4+0.3 24.8+0.4 17.9 + 0.4 24.3 + 0.6
1.08+0.02 1.08+0.01 1.04+ 0.02 0.99+0.02 0.92+0.01 1.07 +0.02 0.94 +0.02
1.5+0.04 2.5+0.16 3.0+ 0.00 2.8+0.10 2.9+0.06 1.2+0.07 2.8 +0.09
2.0+ 0.14 2.6+0.11 2.9 +0.06 2.9+0.07 3.0+0.00 1.3 +0.10 2.9 + 0.06
1.8+0.18 2.2+0.11 2.4+ 0.06 2.4+0.08 2.8+0.07 1.5 0.13 2.9±0.08
June
Data are shown as mean + SEM; * body weight/fork length 3 x 100; **ranked 1 (no silvering) to 3 (silvering); *** ranked 1 (parr marks) to 3 (no parr marks); **** ranked 1 (yellowing) to 3 (no yellowing).
tion experiments were done May 11, 1990 using 56.1 + 3.9 g (n = 9) fish which were all quite silvery with no yellowing and no parr marks.
+
+
CortisolRIA and gill Na , K -A TPase assay Cortisol was assayed in unextracted plasma using a radioimmunoassay (RIA) as described by Young (1986) and validated in our hands (Bisbal and Specker 1991). The gill Na+, K+-ATPase activity was determined using semipurified homogenates (Zaugg and McLain 1972; Bisbal and Specker 1991).
Statistics Intestinal Jv data from smolting fish in 1991 were analyzed using a one-way analysis of variance, followed by the Tukey HSD procedure to examine changes over time. Student's t-test was used to compare means between parr and smolt data from May 1991. Analysis of the 1990 parr and smolt data was performed using analysis of variance for two factor experimental design: 2 Jv determinations x 2 developmental stages. Comparisons were made using Tukey HSD procedure when appropriate. Data from the two intestinal regions were analyzed separately in all cases. Significance was accepted at p < 0.05. SAS (SAS Institute, Cary, NC) was used for all statistical procedures.
Results Developmental changes Upper growth mode Atlantic salmon used for this study in winter and spring 1991 appeared to undergo parr-smolt transformation: they grew considerably between February and early April and body markings increasingly appeared more like that of smolts (Table 1). Body weight of these fish increased from 90.4 + 6.3 g (mean + SEM) in February to 137.4 + 4.8 g in early April and fork length increased from 20.2 + 0.4 cm to 23.6 + 0.2 cm over this same period. Condition factor (body weight/fork length3 x 100) decreased from 1.08 + 0.02 at the beginning of sampling to 0.92 ± 0.01 in May. In February, the upper growth mode fish had faint silvering, and moderate parr marks and yellowing of the opercular, ventral, and fin margin regions. By early April the fish had heavy silvering and faint parr marks and yellowing, indicative of smolting Atlantic salmon. Parr selected from the lower growth mode in May 1991 were smaller than smolts. The mean body weight and fork length of parr were less than that of all smolting fish used in 1991 (Table 1). Condition factor of parr (1.07 + 0.02) was considerably larger than condition factor of smolts in May. This information, taken together with the appearance of the parr (Table 1), indicated these fish were not smolting. In 1990, smolts were bigger than parr (Table 2). Mean body weight of smolts was 93.5 + 3.5 g and
197 Table 2. Size and stage characteristics of Atlantic salmon used for measuring intestinal fluid transport in spring 1990 Month
Growth mode
n
Body weight (g)
Fork length (cm)
Parr marks*
Yellowing*
Late April
Lower Upper Lower Upper
12 12 12 12
25.4+3.2 93.5+3.5 35.4+3.2 114.0+ 7.7
13.4 +0.6 20.6 + 0.3 14.7 +0.4 22.2+0.5
1.8 +0.3 2.9 +0.1 1.2+0.1 2.9+0.1
1.9+0.3 3.0 + 0.0 1.4+0.1 2.9+0.1
Early June
Data are shown as mean + SEM; * ranked as in Table 1.
Middle a
20
15
]
a,b
ab
of fish April, ng-ml smolts. tivities 40.1 +
a,b
5
c'7 0
E.
-a
Posterior 30
a
a a,b
25
20
b,c c
15
February
arch
Early
February
March
Early April
_
Late April
_
_
May
_
June
Fig. 1. Fluid transport rate (Jv) across non-everted sacs prepared from the middle and posterior intestine of smolting Atlantic salmon during winter and spring 1991. Data are shown as mean + SEM; n = 8 in June; n = 11 for late April posterior Jv; n = 12 all other times. Middle and posterior intestine were analyzed separately. Letters indicate groupings from Tukey HSD procedure. Bars not sharing the same letters are significantly (p < 0.05) different.
114.0 + 7.7 g in late April and early June, respectively; whereas parr were 25.4 3.2 g and 35.4 + 3.2 g. Differences in body weight between fish classified as parr and smolts correspond to differences in fork length. In addition to the information presented in Table 2, we have several other pieces of data that argue for a parr and smolt status used for salmon taken from the lower and upper modes
at North Attleboro NFH in 1990. In early plasma cortisol levels were 13.8 3.0 1 for parr and 34.1 3.6 ng-ml- for the In early June, gill Na + , K+-ATPase acin parr and smolts used in this study were 5.7 tAmol-mg- 1 .h - 1 and 99.0 + 1.0 Amol.mg- '.h- 1, respectively. Fluid transport in both the middle and posterior intestine of smolting salmon in 1991 changed over time as indicated by the Tukey groupings in Figure 1. The middle intestine significantly decreased water uptake over the period from March to June. Water uptake in the posterior intestine increased significantly from 14.2 0.9 tz1-cm - 2 .h - I in early April to 23.0 + 2.0 zl-cm-2-h-' in late April. Fluid transport remained elevated and was significantly higher in May and June than in February, March, or early April. The timing of increased water uptake in the posterior intestine coincided with the strong appearance of smolt-like characteristics of the fish (Table 1). Figure 2 shows water uptake for the middle (top panel) and posterior (lower panel) sections of the intestine of parr and smolts in spring 1990 and 1991. In 1990, water uptake across the middle intestine changed over time, but there was no significant main effect due to stage nor a significant interaction. On the other hand, for J, across the posterior intestine there was a significant interaction between stage and month. Examining each month, we found a difference in Jv between parr and smolts in April but not in June. In May 1991, water uptake across the middle intestine did not differ between developmental stages. However, posterior intestinal Jv of smolts was significantly greater than that of parr. From Figure 2 it was apparent that for both parr and smolts, the middle intestine decreased water
198 Middle
[] Pal rr
a
20
[
15
Smolt
15
b
5
3 iz
'C
10 O0
0
E O
Posterior
(.,
-a
30
30 l
a
Ad i.
20
water fish in March when posterior J was relative-
1
ly low. In August, when freshwater- and seawateradapted fish were compared simultaneously, posterior water uptake was more than double in seawater fish than in freshwater fish. The difference in transport rates across the posterior intestine was clearly due to environment.
20
1
b 15
a
:r-
I
e
+
10
15
iiog10 5
0
water uptake by the intestine of freshwater postsmolts and seawater smolts and stunts. Rate of water uptake by the middle intestine was similar in freshwater- and seawater-adapted fish in November and December. However, in March both the seawater smolts and stunts exhibited higher rates of water uptake across the middle intestine. Water transport across the posterior intestine was generally higher in seawater-adapted fish than in their freshwater cohorts. The exception was in the sea-
0
Late April
Early June 1990
May 1991
Fig. 2. Fluid transport rate (Jv) across non-everted sacs prepared from the middle and posterior intestine of parr- and smolt-stage Atlantic salmon in spring 1990 and 1991. Jv measurements in 1990 were done at 130C and in 1991 at 14 0C. Smolt data for 1991 is from Figure 1. Data are shown as mean ± SEM; n = 11-12. Middle and posterior intestine were analyzed separately. Data from 1990 and 1991 were analyzed separately. Bars not sharing the same letters are significantly (p < 0.05) different.
and solute transport during late spring 1990. The stage difference shown for the posterior intestine in late April 1990 and May 1991 was clearly an increase in smolt Jv,.The low posterior Jv of smolts in early June 1990, despite not being significantly different from late April, may indicate a loss of this smolt characteristic.
Seawater adaptation Freshwater post-smolts and seawater smolts used in this study were of similar size through the fall (Table 3). Seawater stunts (late March) were smaller than post-smolts in fresh water and smolts in seawater. Differences in fork-length correspond to differences in body weight. Included in Table 3 are the measurements for
Inhibition by ouabain and N2 Fluid uptake was inhibited by both ouabain (10- 4 M) and by displacement of 02 with N2 (Table 4). The inhibition was significant (paired t-test). The difference in uptake rates of the controls (Table 4) is attributed to the stage of development. The fish used for N2 inhibition experiments were in the parr stage, whereas the fish used for ouabain inhibition experiments were undergoing smolt transformation.
Discussion Atlantic salmon show a distinct bimodality in growth rates during the fall of their first year which results in smolting of some but not all of the population (Thorpe 1977; Kristinsson et al. 1985). By capitalizing on this bimodal developmental pattern we have shown that increased fluid transport across the non-everted posterior intestine sac preparation is a developmental and not a seasonal correlate. Atlantic salmon remaining in the parr stage during the springtime do not appear to exhibit increased intestinal fluid transport rates, whereas their cohorts that are undergoing smolt transformation show concurrently higher fluid transport rates across the
199 2 Table 3. Size characteristics and in vitro fluid transport rate (Jv, tl -cm- -h 1) across middle and posterior intestine of Atlantic salmon post-smolts kept in fresh water and smolts transferred to seawater
Month (1990-91)
Environment
n
Body weight (g)
Fork length (cm)
Middle intestine J
Posterior intestine Jv
Late August
Fresh water Seawater Fresh water Seawater Fresh water Seawater Seawater
12 12 12 12 12 6 6
105.1 ±+ 7.0 101.0+ 10.2 156.4+ 9.5 167.0+ 17.3 214.0 11.8 179.8 13.3 66.8 6.5
23.4+ 0.6 22.1+0.8 25.4+0.5 26.9+0.7 27.8+0.5 26.5 0.7 21.4+0.7
13.0+ 1.9 12.9+± 1.6 13.6+± 1.5 26.6+3.5 20.4±2.0
19.7 + 4.2 44.5 + 7.6 21.5+2.2 40.5 2.3 11.1 + 1.8 20.9+3.6 42.4+5.5
Early November Late November Late December Middle March Late March Data are shown as mean
SEM; - = no data.
Table 4. Ouabain (10-4 M) and N2 inhibition of fluid transport rate (Jv, jlI-cm -2h-1) across the posterior intestine of Atlantic salmon Treatment
n
Control Jv
Treatment Jv,
70decrease
Ouabain N2
9 8
30.1 ±2 .4 16.8+3.7
10.1 + 1.7 5.7±0.9
67+ 5 51 +12
The salmon used for ouabain treatment were smolting; whereas salmon used for N2 treatment were in the parr stage. Data are shown as mean + SEM.
posterior intestine in vitro. The timing of the increase in fluid transport across the posterior intestine of smolts occurs when these fish are typically tolerant of seawater. We have also shown that the posterior intestine of salmon transferred to seawater exhibits a higher fluid transport rate. This suggests that the posterior intestine exhibits functional changes which are part of the complex of preadaptive changes anadromous salmonids undergo during the parr-smolt transformation. Our measurements of fluid transport rates for the posterior intestine are very similar to those obtained for coho salmon by Collie and Bern (1982). However, in coho salmon the fluid transport rate of the posterior intestine was consistently higher than that of the middle intestine and this was not the case for Atlantic salmon. In our study in March 1991, the transport rate of the middle intestine was higher than the J of the posterior intestine in fish from the upper growth mode. Thereafter the J of the middle intestine decreased significantly. Similarly, in late April 1990, middle intestine J, was higher than the posterior Jv in parr-stage Atlantic salmon
and the middle gut decreased fluid uptake significantly thereafter. The 2-fold decrease in both parr and smolts in 1990 was not development-dependent and can only be viewed at this time as a seasonal phenomenon. This springtime decrease in middle intestinal Jv seen in this study differs from the report by Usher et al. (1991) which showed that the Jv of the middle intestine increased during spring in Atlantic salmon. The increase they observed may be a phenomenon occurring prior to the springtime decrease in fluid uptake we found. Jv levels were within a similar range in both studies. We demonstrate that the changes in fluid transport across the middle intestine are not development-dependent whereas, because Usher et al. (1991) did not distinguish between parr and smolt stages, no distinction could be made between the consequences of season or development. Few studies have been conducted on the regionalization of intestinal function during the parrsmolt transformation. Collie (1985) examined proline absorption in coho salmon and found the middle intestine to have a higher rate than the posterior intestine. The highest rate of proline absorption was in pyloric ceca. In Atlantic salmon there seems to be functional regionalization of the middle compared to the posterior intestine during the springtime with perhaps the middle intestine participating more in nutrient uptake and less in salt and water balance. This possibility warrants further investigation. The difference in fluid transport rate of the posterior intestine between parr- and smolt-stage
200 Atlantic salmon was not due to the difference in their size. In early June 1990 the smolts were more than 3-times heavier than parr and their fluid transport weights were similar. Smolts increased in body weight between late April and early June 1990, but during this same period the J,'s decreased. Furthermore, post-smolts in fresh water increased body weight from early November to late December 1990, while posterior Jv decreased almost in half. Our findings from seawater-adapted Atlantic salmon suggest intestinal fluid transport across the posterior intestine is increased beyond freshwater pre-adaptive levels. Fluid transport rate in the posterior intestine of seawater-adapted salmon was typically much higher than the largest measurement from freshwater smolting fish. This was not the case with coho salmon in which posterior fluid transport rates were similar in freshwater smolts and seawater smolts (Collie and Bern 1982). In a similar study, J, across the middle intestine was unchanged by transfer of Atlantic salmon smolts to seawater (Usher et al. 1991). However, the observation of a decrease in sodium concentration in the digesta of Atlantic salmon smolts adapted to seawater indicates uptake of salt by the gut (Usher et al. 1990). In our study, middle intestinal J, was either unchanged or elevated in seawater-adapted smolts compared to their freshwater cohorts, although the middle intestine exhibited no preadaptation to seawater. The functional regionalization of middle and posterior intestine occurring in smolts in the springtime appears to be maintained after seawater entry. Seawater smolts in late November and seawater stunts in March had posterior Jv considerably higher than in the middle intestine, similar to the smolt-stage salmon in fresh water. The exception to this was the seawater smolts in March when middle Jv was higher than the posterior Jv. The significance of this is unclear. However, the generally high posterior transport rates for seawater-adapted fish over their freshwater cohorts suggests that the posterior intestine of Atlantic salmon maintains an important role in hypoosmoregulation in a marine environment. The mechanisms of fluid transport across the
salmonid intestine are unresolved. However, our findings are in agreement with Usher et al. (1991) for Atlantic salmon: fluid absorption is linked to active solute transport. We show that ouabain and N2 inhibit transport. This suggests that, as in the eel and coho salmon intestine (Oide 1967; Hirano et al. 1976; Collie and Bern 1982), fluid transport is caused in part by ouabain sensitive Na+, K + -
ATPases that are partially dependent on oxidative metabolism. Thus weight loss from the non-everted intestinal sac preparation results from net fluid flux from mucosal to serosal surfaces and is secondarily linked to this salt pump. In our hands, ouabain inhibited fluid transport across the posterior intestine 67% compared to 68% for coho salmon, and N2 inhibited fluid transport 51 %1compared to 59% for coho salmon (Collie and Bern 1982). The factors regulating these changes in intestinal function during the parr-smolt transformation are unknown. Cortisol is a likely endocrine signal as increased cortisol levels are correlated with smolting in coho salmon and Atlantic salmon (Specker and Schreck 1982; Young 1986; Langhorne and Simpson 1986) and exogenous cortisol can improve seawater tolerance and increase gill and gut Na+, K+-ATPase in Pacific salmon, trout, and Atlantic salmon (McCormick and Bern 1989; Madsen 1990a,b; Bisbal and Specker 1991; Specker et al. 1993).In the present study, plasma cortisol levels were elevated in smolts compared to parr in early April 1990. However, in a recent study (Usher et al. 1991) fluid transport across the middle intestine of Atlantic salmon was unaffected by cortisol treatment in vitro. The lack of effect may be because there is a long latent period between the cortisol stimulus and increased fluid transport. Thyroid hormones may also regulate changes in intestinal function since elevated thyroxine levels accompany parr-smolt transformation, thyroxine increases the whole animal transepithelial potential of coho salmon transferred to seawater in April, and because they may act by preparing the gut for the transition to seawater (Dickhoff and Sullivan 1987; Iwata et al. 1987; Specker 1988). Anadromous salmon acquire the ability to maintain salt and water balance in the marine environment during the developmental process of smolting
201
which occurs in fresh water. Preadaptive changes for life at sea have been noted in the gills, kidney, urinary bladder, and gut (McCormick and Saunders 1987). The developmental changes in the gut are of particular interest because of the importance of the intestinal epithelium's other role in nutrient absorption. Known functional changes in the salmonid intestine during parr-smolt transformation are limited to descriptions of increased fluid transport in the posterior intestine of coho salmon in the smolt stage, increased proline influx across the intestine of coho salmon in the smolt stage, regional differences in proline influx in the coho salmon (Collie and Bern 1982; Collie 1985), and increased fluid transport in the middle intestine of smolt-stage Atlantic salmon (Usher et al. 1991). Because smolting is a metabolically demanding, springtime phenomenon, these observations could have been explained as consequences of either development or season. We have shown that in Atlantic salmon the springtime increase in fluid transport across the posterior intestine is developmentdependent and a functional component of preparation for life at sea. The middle intestine of both parr and smolt stage Atlantic salmon undergoes a considerable decrease in fluid transport rate during the spring, perhaps because it is more engaged in nutrient uptake during this season when growth increases.
Acknowledgements We wish to thank Mr. W. Booker and Mr. L.
Lofton at the North Attleboro National Fish Hatchery for their help in maintaining and supplying the Atlantic salmon. Facilities for maintaining salmon in seawater were generously provided by Dr. Ted Durbin. Sean Cornell provided skillful technical assistance. Dr. Allen T. Scholz provided numerous suggestions with the manuscript. This publication is the result of research funded by a URI Undergraduate Research Award (to P.A.V.), NOAA office of Sea Grant, U.S. Department of Commerce, under grant #NA89AA-D-SG082, and by U.S. Department of Interior, Fish and Wildlife Service, Cooperative Research Program Agree-
ment, Project #F-52R, with Rhode Island Division of Fish and Wildlife. The U.S. government is authorized to produce and distribute reprints for governmental purposes notwithstanding any copyright notation that may appear hereon. References cited Bisbal, G.A. and Specker, J.L. 1991. Cortisol stimulates hypoosmoregulatory ability in Atlantic salmon, Salmo salar L. J. Fish Biol. 39: 421-432. Brown, M.E. 1957. Experimental studies on growth. In Physiology of Fishes. Vol. 1, pp. 361-400. Edited by M.E. Brown. Academic Press, New York. Colin, D.A., Nonnotte, G., LeRay, C. and Nonnotte, L. 1985. Na transport and enzyme activities in the intestine of the freshwater and sea-water adapted trout (Salmo gairdneriiR.). Comp. Biochem. Physiol. 81A: 695-698. Collie, N.L. 1985. Intestinal nutrient transport in coho salmon (Oncorhynchus kisutch) and the effects of development, starvation, and seawater adaptation. J. Comp. Physiol. B. 156: 163-174. Collie, N.L. and Bern, H.A. 1982. Changes in intestinal fluid transport associated with smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch (Walbaum). J. Fish Biol. 21: 337-348. Collie, N.L. and Hirano, T. 1987. Mechanisms of hormone actions on intestinal transport. In Vertebrate Endocrinology: Fundamentals and Biomedical Implicatons. Vol. 2, pp. 239-244. Academic Press, New York. Dickhoff, W.W. and Sullivan, C.V. 1987. The thyroid gland in smoltification. Am. Fish. Soc. Symp. 1: 197-210. Hirano, T., Morisawa, M., Ando, M. and Utida, S. 1976. Adaptive changes in ion and water transport mechanism in the eel intestine. In Intestinal Ion Transport. pp. 301-317. Edited by J.W.L. Robinson. MTP Press, Lancaster. Hoar, W.S. 1988. The physiology of smolting salmonids. In Fish Physiology. Vol. 2, part B, pp. 275-343. Edited by W.S. Hoar and D.J. Randall. Academic Press, New York. Holmes, W.N. and Stainer, I.M. 1966. Studies on the renal excretion of electrolytes by the trout Salmo gairdneri'(Richardson). J. Exp. Biol. 44: 33-46. Iwata, M., Nishioka, R.S. and Bern, H.A. 1987. Whole animal transepithelial potential (TEP) of coho salmon during the parr-smolt transformation and effects of thyroxine, prolactin and hypophysectomy. Fish Physiol. Biochem. 3: 25-38. Kristinsson, J.B., Saunders, R.L. and Wiggs, A.J. 1985. Growth dynamics during the development of bimodal lengthfrequency distribution in juvenile Atlantic salmon (Salmo salar L.). Aquaculture 45: 1-20. Langhorne, P. and Simpson, T.H. 1986. The interrelationship of cortisol, gill (Na+K+)ATPase, and homeostasis during the parr-smolt transformation of Atlantic salmon (Salmo salar L.). Gen. Comp. Endocrinol. 61: 203-213.
202 Madsen, S.S. 1990a. Cortisol treatment improves the development of hypoosmoregulatory mechanisms in the euryhaline rainbow trout, Salmo gairdneri. Fish Physiol. Biochem. 8: 45-52. Madsen, S.S. 1990b. The role of cortisol and growth hormone in seawater adaptation and development of hypoosmoregulatory mechanisms in sea trout parr (Salmo trutta trutta). Gen. Comp. Endocrinol. 79: 1-11. McCormick, S.D. and Bern, H.A. 1989. In vitro stimulation of Na+ -K +-ATPase activity and ouabain binding by cortisol in coho salmon gill. Am. J. Physiol. 256: R707-715. McCormick, S.D. and Saunders, R.L. 1987. Preparatory physiological adaptations for marine life of salmonids: Osmoregulation, growth, and metabolism. Am. Fish. Soc. Symp. 1: 211-229. Oide, M. 1967. Effects of inhibitors on transport of water and ion in isolated intestine and Na+-K+ ATPase in intestinal mucosa of the eel. Ann. Zool. Jap. 40: 130-135. Saunders, R.L., Henderson, E.B. and Glebe, B.D. 1982. Precocious sexual maturation and smoltification in male Atlantic salmon (Salmo salar). Aquaculture 28: 211-229. Smith, H.W. 1932. Water regulation and its evolution in the fishes. Quart. Rev. Biol. 7: 1-26. Specker, J.L. 1988. Preadaptive role of thyroid hormones in larval and juvenile salmon: Growth, the gut and evolutionary considerations. Am. Zool. 28: 337-349. Specker, J.L. and Schreck, C.B. 1982. Changes in plasma corticosteroids during smoltification of coho salmon, Oncorhynchus kisutch. Gen. Comp. Endocrinol. 46: 53-58. Specker, J.L., Portesi, D.M., Cornell, S.C. and Veillette, P.A. 1993. Methodology for implanting cortisol in Atlantic salmon
and effects of chronically elevated cortisol on osmoregulatory physiology. Aquaculture (In press). Thorpe, J.E. 1977. Bimodal distribution of length of juvenile Atlantic salmon (Salmo salar L.) under artificial rearing conditions. J. Fish Biol. 11: 175-184. Thorpe, J.E., Morgan, R.I.G., Ottaway, E.M. and Miles, M.S. 1980. Time of divergence of growth groups between potential 1 + and 2 + smolts among sibling Atlantic salmon. J. Fish Biol. 17: 13-21. Usher, M.L., Talbot, C. and Eddy, F.B. 1988. Drinking in Atlantic salmon smolts transferred to seawater and the relationship between drinking and feeding. Aquaculture 73: 237-246. Usher, M.L., Talbot, C. and Eddy, F.B. 1990. Effects of transfer to seawater on digestion and gut function in Atlantic salmon smolts (Salmo salar L.). Aquaculture 90: 85-96. Usher, M.L., Talbot, C. and Eddy, F.B. 1991. Intestinal water transport in juvenile Atlantic salmon (Salmo salar L.) during smolting and following transfer to seawater. Comp. Biochem. Physiol. 100A: 813-818. Young, G. 1986. Cortisol secretion in vivo by the interrenal of coho salmon (Oncorhynchus kisutch) during smoltification: Relationship with plasma thyroxine and plasma cortisol. Gen. Comp. Endocrinol. 63: 191-200. Zaugg, W.S. and McLain, L.R. 1970. Adenosinetriphosphatase activity in gills of salmonids: seasonal variation and salt water influence in coho salmon, Oncorhynchus kisutch. Comp. Biochem. Physiol. 35: 587-596. Zaugg, W.S. and McLain, L.R. 1972. Changes in gill adenosinetriphosphatase activity associated with parr-smolt transformation in steelhead trout, coho, and spring chinook salmon. J. Fish. Res. Bd. Can. 29: 167-171.
Changes in intestinal fluid transport in Atlantic salmon (Salmo salar L) during parr-smolt transformation Philip A. Veillettel, Ronald J. White 2 , and Jennifer L. Speckerl 1 Department of Zoology, University of Rhode Island, Kingston, RI 02881, U.S.A.; 2 Department of Biology, Eastern Washington University, Cheney, WA 99004, U.S.A.
Accepted: April 23, 1993 Keywords: intestinal fluid transport, parr-smolt transformation, seawater adaptation, Atlantic salmon
Abstract We examined changes in fluid transport by the intestine of Atlantic salmon (Salmo salarL) undergoing parrsmolt transformation during springtime. In vitro measurements of fluid transport rate (J,) across noneverted middle and posterior intestinal sac preparations were made in late April and early June 1990 and from February through June 1991 for juvenile smolting fish. Intestinal J, was also compared between parr- and smolt-stage salmon in both years. To evaluate the osmoregulatory role of the intestine, Jv was measured for smolts adapted to seawater and their cohorts remaining in fresh water. The middle intestine of smolting fish underwent a significant decrease in fluid transport during the springtime, while posterior intestinal Jv significantly increased. Parr-stage fish decreased Jv in the middle intestine during springtime similar to smolts. However, the posterior intestinal J, of smolts showed a significant increase over the parr around the peak smolt period in both years. Seawater-adapted smolts generally exhibited posterior intestinal Jv approximately double that of freshwater cohorts. A decrease over time shown for the middle intestine, together with the increased Jv in the posterior intestine preceding and after seawater entry, suggests the development of a functional regionalization during parr-smolt transformation, with the posterior intestine taking on increased importance in osmoregulation in seawater.
Introduction
(Brown 1957). Glomerular filtration rate decreases, gill Na4' , K+-ATPase activity increases, and intes-
Smith (1932) demonstrated that teleosts in seawater survive dehydration by drinking water and reducing urine production. In anadromous salmonids, many preadaptive physiological changes occur during the parr-smolt transformation which prepare the fish for seawater while they are still in fresh water (McCormick and Saunders 1987; Hoar 1988; Specker 1988). Salinity tolerance develops during a fast-growing, freshwater phase of the smolting parr
tinal fluid uptake increases in the springtime prior to seawater entry (Holmes and Stainier 1966; Zaugg and McLain 1970, 1972; Collie and Bern 1982). Finally, in response to seawater entry, drinking rate increases (Usher et al. 1988). The intestine is a major site of salt and water balance in vertebrates, but much remains to be known of its role in seawater adaptation of fishes. In euryhaline fishes, intestinal absorption increases
Correspondence to: Philip A. Veillette, Department of Zoology, University of Rhode Island, Kingston, RI 02881, USA; Fax (401) 792-4256
194 during adaptation to seawater and this process is mediated by corticosteroids (Collie and Hirano 1987). In coho salmon (Oncorhynchus kisutch), the posterior intestine attains a rate of fluid absorption during the smolt-period in fresh water which is as high as the rate attained when they are adapted to seawater (Collie and Bern 1982). Atlantic salmon (Salmo salar) show a marked bimodal distribution in growth in their first year and only some become 1 + smolts (Thorpe 1977; Thorpe et al. 1980; Saunders et al. 1982). We were interested in whether the posterior intestine is part of the hypoosmoregulatory mechanisms that preadapt Atlantic salmon to seawater as has been shown for coho salmon (Collie and Bern 1982) and if so whether parr and smolts differed during the springtime. We examined concurrent changes in the middle intestine and identified regional differences in the pattern of changes that occur in parr and smolts during springtime. Furthermore, we followed changes in smolts after entry into seawater.
Materials and methods Developmental changes Juvenile Atlantic salmon were used to monitor developmental changes in parr and smolts. The fish were reared at the North Attleboro National Fish Hatchery (NFH) in 2.5 m wide by 1.5 m deep raceways with flow-through water at 100 C. They were fed 2% of their body weight each day over several feedings. In November they were separated into two size classes and subsequently maintained in adjacent sections of the same raceway separated by a wire screen. Fish used for measuring intestinal fluid transport were transported to the University of Rhode Island salmon facility at the Biological Sciences Center in late April and early June 1990 and in February, March, early April, late April, May and June 1991. Fish were held in flow-through well water at 8-15°C for 2-5 days without food prior to sampling. Lower mode parr have the tendency to enter the faster growing phase and become upper mode smolts (Kristinsson et al. 1985). Because of this,
parr were selected by taking smaller, less silvery, lower mode fish which exhibited parr marks and yellowing of the pectoral fins and the opercular and ventral regions. Smolting fish were chosen from the upper growth mode. Rate of fluid transport (J) across the middle and posterior intestine was measured for 12 smolting fish at sample times in February through May (except late April posterior intestine, n = 11) and for 8 smolting fish in June 1991. Intestinal Jv was also measured for middle and posterior regions for 12 parr in May 1991. In 1990, measurements were made on the middle and posterior intestine of 11 - 12 parr and 11- 12 smolting fish at both sample times. We have chosen to use the term middle intestine as used by Colin et al. (1985), rather than anterior intestine, so that differentiation may be made between the intestinal pyloric ceca area and the non-pyloric ceca area anterior to the ileorectal valve. The fish were anesthetized in 2-phenoxyethanol and killed by a blow to the head or by a dorsal cut through the spinal cord. Body weight, fork length, and appearance were recorded. Fish were then placed on ice until removal of the intestine. The maximum duration between these procedures and the removal of the intestine from all fish and their placement in aerated physiological saline was 1.5h. The body cavity was opened and the posterior intestine was cut free of the body wall. The intestine was removed by gently tearing the mesenteries until they hung free from the peritoneal cavity. An effort was made to prevent the intestine from becoming contaminated with external body secretions. The middle intestine was cut at the junction just posterior to the attachment of the pyloric ceca. Adhering mesenteries and blood vessels were then removed and the middle and posterior segments were separated at the prominent ileorectal valve. The lumen was rinsed with Ringer (140 mM NaCl, 15 mM NaHCO 3, 1.5 mM CaC12, 1 mM KH 2PO 4 , 0.8 mM MgSO 4, 2.5 mM KCI, 10 mM glucose and 5 mM N-2-Hydroxyethylpiperazine-N ' -2-ethanesulfonic acid (HEPES) buffer, pKa = 7.5), adjusted to pH 7.8 and 300 + 10 mmol-kg - ' at 250 C. Osmolality of buffer solutions was measured on a vapor pressure osmometer (Wescor, model 5100).
195 Measurement of Jv After rinsing out the contents of the intestine with buffered Ringer, the most distal end was tied off with light waxed or unwaxed, unflavored, flossing nylon. A blunted hypodermic needle attached to a 20 or 30 ml plastic syringe was then inserted into the other end of the intestine and a single tie of nylon was made about the intestine and needle. The intestine was then inflated with Ringer until it began to slip off the shank of the needle. At this point the filling was stopped, the intestine was slid off the needle and the knot was pulled tight. Despite possible variability in the pressures among the intestinal sacs, Collie and Bern (1982) have shown that noneverted intestinal sacs from coho salmon do not have significant differences in the slope of sac weight loss with time for hydrostatic pressures between 0-20 cm H 20. The sac was then suspended in 35-50 ml of Ringer at 13 0C (1990) or 14°C (1991) and bubbled vigorously with 95%°7 02/5% CO 2. The sac was pre-incubated lh prior to measuring Jv. After pre-incubation the sac was weighed to the nearest mg at 10 or 15 min intervals for at least lh. Collie and Bern (1982) have shown with histological examination that no gross morphological changes are apparent after 7h incubation of the non-everted intestinal sac preparation. The slope of the regression line for sac weight loss plotted against time represents rate of water loss. This was normalized by dividing by the surface area of the sac to give a Jv in plcm-2-h 1. Surface area was determined by spreading the sac serosal side down on Truwax Baseplate wax (Dentsply International Inc.), outlining it with #7 jewelers forceps then filling the groove with neutral decolorizing charcoal (Norit Co., Inc.). The outline of the sac was then traced with a Scriptel SPD Graphics Tablet digitizer (Jandel Corp.).
Seawater adaptation In July 1990, 1 + freshwater Atlantic salmon (reared as previously described) were either kept in fresh water at the salmon facility or acclimated to seawater at the University of Rhode Island's
Graduate School of Oceanography (GSO). Freshwater fish were maintained in aerated, flowthrough well water with temperature varying seasonally from 8-16°C. Salmon at the GSO were transferred to 9%0 seawater. Immediately after transfer, we began adding full strength, Narragansett Bay seawater. The tank was maintained with aeration and flow-through seawater at a salinity of 30 + 2%0. Water temperature varied seasonally from 5--18°C. Prior to sampling, fish were starved for 2-7 days. Posterior intestinal J, was measured on 12 freshwater- and 12 seawater-adapted fish in late August 1990. Middle and posterior intestinal Jv were measured on either 12 freshwater fish or 6 or 12 seawater-adapted fish at selected times from November 1990 through March 1991. In addition, measurements were made on 6 seawater stunts in late March 1991. Sampling of fish and preparation of intestinal sacs were as previously described; with the exception that an alternate physiological Ringer (composed of 150 mM NaCl, 2.5 mM KCI, 3.5 mM CaCI2, 1 mM MgC1 2, 7 mM NaHCO 3, 0.7 mM NaHPO 4, 5 mM HEPES and 10 mM glucose, adjusted to pH 7.8 and 310-316 mmol.kg - 1) was used for intestinal sac preparations from seawateradapted salmon at all times and freshwater salmon in late August 1990. Fluid transport was measured at 16°C in August 1990 and 14°C at all other times.
Inhibition by ouabain and N 2 The effect of Ringer containing 10 - 4 M ouabain or of N2 in place of 0 2/CO 2 to the serosal side of posterior intestinal sac preparations was determined. Intestinal sacs were preincubated for 60 min; then fluid transport was measured for 50 min. Following this, either ouabain was added to the Ringer or N2 treatment was started. The intestinal sacs were allowed to incubate under these conditions for 20 min, followed by a second J, determination over 50 min. The N 2 inhibition experiments 1.7 were performed on April 10, 1990 using 21.7 g (mean SEM; n = 8) parr. The ouabain inhibi-
196 Table 1. Size and stage characteristics of Atlantic salmon used for measuring intestinal fluid transport in winter and spring 1991 Month
Growth mode
n
Body weight (g)
Fork length (cm)
Condition factor*
Silvering**
Parr marks***
Yellowing****
February March Early April Late April May
Upper Upper Upper Upper Upper Lower Upper
12 12 12 12 12 12 8
90.4+ 6.3 119.0+4.3 137.4+4.8 112.3+4.6 142.2+8.1 62.8 4.8 136.1 +9.1
20.2 + 0.4 22.3+0.3 23.6+0.2 22.4+0.3 24.8+0.4 17.9 + 0.4 24.3 + 0.6
1.08+0.02 1.08+0.01 1.04+ 0.02 0.99+0.02 0.92+0.01 1.07 +0.02 0.94 +0.02
1.5+0.04 2.5+0.16 3.0+ 0.00 2.8+0.10 2.9+0.06 1.2+0.07 2.8 +0.09
2.0+ 0.14 2.6+0.11 2.9 +0.06 2.9+0.07 3.0+0.00 1.3 +0.10 2.9 + 0.06
1.8+0.18 2.2+0.11 2.4+ 0.06 2.4+0.08 2.8+0.07 1.5 0.13 2.9±0.08
June
Data are shown as mean + SEM; * body weight/fork length 3 x 100; **ranked 1 (no silvering) to 3 (silvering); *** ranked 1 (parr marks) to 3 (no parr marks); **** ranked 1 (yellowing) to 3 (no yellowing).
tion experiments were done May 11, 1990 using 56.1 + 3.9 g (n = 9) fish which were all quite silvery with no yellowing and no parr marks.
+
+
CortisolRIA and gill Na , K -A TPase assay Cortisol was assayed in unextracted plasma using a radioimmunoassay (RIA) as described by Young (1986) and validated in our hands (Bisbal and Specker 1991). The gill Na+, K+-ATPase activity was determined using semipurified homogenates (Zaugg and McLain 1972; Bisbal and Specker 1991).
Statistics Intestinal Jv data from smolting fish in 1991 were analyzed using a one-way analysis of variance, followed by the Tukey HSD procedure to examine changes over time. Student's t-test was used to compare means between parr and smolt data from May 1991. Analysis of the 1990 parr and smolt data was performed using analysis of variance for two factor experimental design: 2 Jv determinations x 2 developmental stages. Comparisons were made using Tukey HSD procedure when appropriate. Data from the two intestinal regions were analyzed separately in all cases. Significance was accepted at p < 0.05. SAS (SAS Institute, Cary, NC) was used for all statistical procedures.
Results Developmental changes Upper growth mode Atlantic salmon used for this study in winter and spring 1991 appeared to undergo parr-smolt transformation: they grew considerably between February and early April and body markings increasingly appeared more like that of smolts (Table 1). Body weight of these fish increased from 90.4 + 6.3 g (mean + SEM) in February to 137.4 + 4.8 g in early April and fork length increased from 20.2 + 0.4 cm to 23.6 + 0.2 cm over this same period. Condition factor (body weight/fork length3 x 100) decreased from 1.08 + 0.02 at the beginning of sampling to 0.92 ± 0.01 in May. In February, the upper growth mode fish had faint silvering, and moderate parr marks and yellowing of the opercular, ventral, and fin margin regions. By early April the fish had heavy silvering and faint parr marks and yellowing, indicative of smolting Atlantic salmon. Parr selected from the lower growth mode in May 1991 were smaller than smolts. The mean body weight and fork length of parr were less than that of all smolting fish used in 1991 (Table 1). Condition factor of parr (1.07 + 0.02) was considerably larger than condition factor of smolts in May. This information, taken together with the appearance of the parr (Table 1), indicated these fish were not smolting. In 1990, smolts were bigger than parr (Table 2). Mean body weight of smolts was 93.5 + 3.5 g and
197 Table 2. Size and stage characteristics of Atlantic salmon used for measuring intestinal fluid transport in spring 1990 Month
Growth mode
n
Body weight (g)
Fork length (cm)
Parr marks*
Yellowing*
Late April
Lower Upper Lower Upper
12 12 12 12
25.4+3.2 93.5+3.5 35.4+3.2 114.0+ 7.7
13.4 +0.6 20.6 + 0.3 14.7 +0.4 22.2+0.5
1.8 +0.3 2.9 +0.1 1.2+0.1 2.9+0.1
1.9+0.3 3.0 + 0.0 1.4+0.1 2.9+0.1
Early June
Data are shown as mean + SEM; * ranked as in Table 1.
Middle a
20
15
]
a,b
ab
of fish April, ng-ml smolts. tivities 40.1 +
a,b
5
c'7 0
E.
-a
Posterior 30
a
a a,b
25
20
b,c c
15
February
arch
Early
February
March
Early April
_
Late April
_
_
May
_
June
Fig. 1. Fluid transport rate (Jv) across non-everted sacs prepared from the middle and posterior intestine of smolting Atlantic salmon during winter and spring 1991. Data are shown as mean + SEM; n = 8 in June; n = 11 for late April posterior Jv; n = 12 all other times. Middle and posterior intestine were analyzed separately. Letters indicate groupings from Tukey HSD procedure. Bars not sharing the same letters are significantly (p < 0.05) different.
114.0 + 7.7 g in late April and early June, respectively; whereas parr were 25.4 3.2 g and 35.4 + 3.2 g. Differences in body weight between fish classified as parr and smolts correspond to differences in fork length. In addition to the information presented in Table 2, we have several other pieces of data that argue for a parr and smolt status used for salmon taken from the lower and upper modes
at North Attleboro NFH in 1990. In early plasma cortisol levels were 13.8 3.0 1 for parr and 34.1 3.6 ng-ml- for the In early June, gill Na + , K+-ATPase acin parr and smolts used in this study were 5.7 tAmol-mg- 1 .h - 1 and 99.0 + 1.0 Amol.mg- '.h- 1, respectively. Fluid transport in both the middle and posterior intestine of smolting salmon in 1991 changed over time as indicated by the Tukey groupings in Figure 1. The middle intestine significantly decreased water uptake over the period from March to June. Water uptake in the posterior intestine increased significantly from 14.2 0.9 tz1-cm - 2 .h - I in early April to 23.0 + 2.0 zl-cm-2-h-' in late April. Fluid transport remained elevated and was significantly higher in May and June than in February, March, or early April. The timing of increased water uptake in the posterior intestine coincided with the strong appearance of smolt-like characteristics of the fish (Table 1). Figure 2 shows water uptake for the middle (top panel) and posterior (lower panel) sections of the intestine of parr and smolts in spring 1990 and 1991. In 1990, water uptake across the middle intestine changed over time, but there was no significant main effect due to stage nor a significant interaction. On the other hand, for J, across the posterior intestine there was a significant interaction between stage and month. Examining each month, we found a difference in Jv between parr and smolts in April but not in June. In May 1991, water uptake across the middle intestine did not differ between developmental stages. However, posterior intestinal Jv of smolts was significantly greater than that of parr. From Figure 2 it was apparent that for both parr and smolts, the middle intestine decreased water
198 Middle
[] Pal rr
a
20
[
15
Smolt
15
b
5
3 iz
'C
10 O0
0
E O
Posterior
(.,
-a
30
30 l
a
Ad i.
20
water fish in March when posterior J was relative-
1
ly low. In August, when freshwater- and seawateradapted fish were compared simultaneously, posterior water uptake was more than double in seawater fish than in freshwater fish. The difference in transport rates across the posterior intestine was clearly due to environment.
20
1
b 15
a
:r-
I
e
+
10
15
iiog10 5
0
water uptake by the intestine of freshwater postsmolts and seawater smolts and stunts. Rate of water uptake by the middle intestine was similar in freshwater- and seawater-adapted fish in November and December. However, in March both the seawater smolts and stunts exhibited higher rates of water uptake across the middle intestine. Water transport across the posterior intestine was generally higher in seawater-adapted fish than in their freshwater cohorts. The exception was in the sea-
0
Late April
Early June 1990
May 1991
Fig. 2. Fluid transport rate (Jv) across non-everted sacs prepared from the middle and posterior intestine of parr- and smolt-stage Atlantic salmon in spring 1990 and 1991. Jv measurements in 1990 were done at 130C and in 1991 at 14 0C. Smolt data for 1991 is from Figure 1. Data are shown as mean ± SEM; n = 11-12. Middle and posterior intestine were analyzed separately. Data from 1990 and 1991 were analyzed separately. Bars not sharing the same letters are significantly (p < 0.05) different.
and solute transport during late spring 1990. The stage difference shown for the posterior intestine in late April 1990 and May 1991 was clearly an increase in smolt Jv,.The low posterior Jv of smolts in early June 1990, despite not being significantly different from late April, may indicate a loss of this smolt characteristic.
Seawater adaptation Freshwater post-smolts and seawater smolts used in this study were of similar size through the fall (Table 3). Seawater stunts (late March) were smaller than post-smolts in fresh water and smolts in seawater. Differences in fork-length correspond to differences in body weight. Included in Table 3 are the measurements for
Inhibition by ouabain and N2 Fluid uptake was inhibited by both ouabain (10- 4 M) and by displacement of 02 with N2 (Table 4). The inhibition was significant (paired t-test). The difference in uptake rates of the controls (Table 4) is attributed to the stage of development. The fish used for N2 inhibition experiments were in the parr stage, whereas the fish used for ouabain inhibition experiments were undergoing smolt transformation.
Discussion Atlantic salmon show a distinct bimodality in growth rates during the fall of their first year which results in smolting of some but not all of the population (Thorpe 1977; Kristinsson et al. 1985). By capitalizing on this bimodal developmental pattern we have shown that increased fluid transport across the non-everted posterior intestine sac preparation is a developmental and not a seasonal correlate. Atlantic salmon remaining in the parr stage during the springtime do not appear to exhibit increased intestinal fluid transport rates, whereas their cohorts that are undergoing smolt transformation show concurrently higher fluid transport rates across the
199 2 Table 3. Size characteristics and in vitro fluid transport rate (Jv, tl -cm- -h 1) across middle and posterior intestine of Atlantic salmon post-smolts kept in fresh water and smolts transferred to seawater
Month (1990-91)
Environment
n
Body weight (g)
Fork length (cm)
Middle intestine J
Posterior intestine Jv
Late August
Fresh water Seawater Fresh water Seawater Fresh water Seawater Seawater
12 12 12 12 12 6 6
105.1 ±+ 7.0 101.0+ 10.2 156.4+ 9.5 167.0+ 17.3 214.0 11.8 179.8 13.3 66.8 6.5
23.4+ 0.6 22.1+0.8 25.4+0.5 26.9+0.7 27.8+0.5 26.5 0.7 21.4+0.7
13.0+ 1.9 12.9+± 1.6 13.6+± 1.5 26.6+3.5 20.4±2.0
19.7 + 4.2 44.5 + 7.6 21.5+2.2 40.5 2.3 11.1 + 1.8 20.9+3.6 42.4+5.5
Early November Late November Late December Middle March Late March Data are shown as mean
SEM; - = no data.
Table 4. Ouabain (10-4 M) and N2 inhibition of fluid transport rate (Jv, jlI-cm -2h-1) across the posterior intestine of Atlantic salmon Treatment
n
Control Jv
Treatment Jv,
70decrease
Ouabain N2
9 8
30.1 ±2 .4 16.8+3.7
10.1 + 1.7 5.7±0.9
67+ 5 51 +12
The salmon used for ouabain treatment were smolting; whereas salmon used for N2 treatment were in the parr stage. Data are shown as mean + SEM.
posterior intestine in vitro. The timing of the increase in fluid transport across the posterior intestine of smolts occurs when these fish are typically tolerant of seawater. We have also shown that the posterior intestine of salmon transferred to seawater exhibits a higher fluid transport rate. This suggests that the posterior intestine exhibits functional changes which are part of the complex of preadaptive changes anadromous salmonids undergo during the parr-smolt transformation. Our measurements of fluid transport rates for the posterior intestine are very similar to those obtained for coho salmon by Collie and Bern (1982). However, in coho salmon the fluid transport rate of the posterior intestine was consistently higher than that of the middle intestine and this was not the case for Atlantic salmon. In our study in March 1991, the transport rate of the middle intestine was higher than the J of the posterior intestine in fish from the upper growth mode. Thereafter the J of the middle intestine decreased significantly. Similarly, in late April 1990, middle intestine J, was higher than the posterior Jv in parr-stage Atlantic salmon
and the middle gut decreased fluid uptake significantly thereafter. The 2-fold decrease in both parr and smolts in 1990 was not development-dependent and can only be viewed at this time as a seasonal phenomenon. This springtime decrease in middle intestinal Jv seen in this study differs from the report by Usher et al. (1991) which showed that the Jv of the middle intestine increased during spring in Atlantic salmon. The increase they observed may be a phenomenon occurring prior to the springtime decrease in fluid uptake we found. Jv levels were within a similar range in both studies. We demonstrate that the changes in fluid transport across the middle intestine are not development-dependent whereas, because Usher et al. (1991) did not distinguish between parr and smolt stages, no distinction could be made between the consequences of season or development. Few studies have been conducted on the regionalization of intestinal function during the parrsmolt transformation. Collie (1985) examined proline absorption in coho salmon and found the middle intestine to have a higher rate than the posterior intestine. The highest rate of proline absorption was in pyloric ceca. In Atlantic salmon there seems to be functional regionalization of the middle compared to the posterior intestine during the springtime with perhaps the middle intestine participating more in nutrient uptake and less in salt and water balance. This possibility warrants further investigation. The difference in fluid transport rate of the posterior intestine between parr- and smolt-stage
200 Atlantic salmon was not due to the difference in their size. In early June 1990 the smolts were more than 3-times heavier than parr and their fluid transport weights were similar. Smolts increased in body weight between late April and early June 1990, but during this same period the J,'s decreased. Furthermore, post-smolts in fresh water increased body weight from early November to late December 1990, while posterior Jv decreased almost in half. Our findings from seawater-adapted Atlantic salmon suggest intestinal fluid transport across the posterior intestine is increased beyond freshwater pre-adaptive levels. Fluid transport rate in the posterior intestine of seawater-adapted salmon was typically much higher than the largest measurement from freshwater smolting fish. This was not the case with coho salmon in which posterior fluid transport rates were similar in freshwater smolts and seawater smolts (Collie and Bern 1982). In a similar study, J, across the middle intestine was unchanged by transfer of Atlantic salmon smolts to seawater (Usher et al. 1991). However, the observation of a decrease in sodium concentration in the digesta of Atlantic salmon smolts adapted to seawater indicates uptake of salt by the gut (Usher et al. 1990). In our study, middle intestinal J, was either unchanged or elevated in seawater-adapted smolts compared to their freshwater cohorts, although the middle intestine exhibited no preadaptation to seawater. The functional regionalization of middle and posterior intestine occurring in smolts in the springtime appears to be maintained after seawater entry. Seawater smolts in late November and seawater stunts in March had posterior Jv considerably higher than in the middle intestine, similar to the smolt-stage salmon in fresh water. The exception to this was the seawater smolts in March when middle Jv was higher than the posterior Jv. The significance of this is unclear. However, the generally high posterior transport rates for seawater-adapted fish over their freshwater cohorts suggests that the posterior intestine of Atlantic salmon maintains an important role in hypoosmoregulation in a marine environment. The mechanisms of fluid transport across the
salmonid intestine are unresolved. However, our findings are in agreement with Usher et al. (1991) for Atlantic salmon: fluid absorption is linked to active solute transport. We show that ouabain and N2 inhibit transport. This suggests that, as in the eel and coho salmon intestine (Oide 1967; Hirano et al. 1976; Collie and Bern 1982), fluid transport is caused in part by ouabain sensitive Na+, K + -
ATPases that are partially dependent on oxidative metabolism. Thus weight loss from the non-everted intestinal sac preparation results from net fluid flux from mucosal to serosal surfaces and is secondarily linked to this salt pump. In our hands, ouabain inhibited fluid transport across the posterior intestine 67% compared to 68% for coho salmon, and N2 inhibited fluid transport 51 %1compared to 59% for coho salmon (Collie and Bern 1982). The factors regulating these changes in intestinal function during the parr-smolt transformation are unknown. Cortisol is a likely endocrine signal as increased cortisol levels are correlated with smolting in coho salmon and Atlantic salmon (Specker and Schreck 1982; Young 1986; Langhorne and Simpson 1986) and exogenous cortisol can improve seawater tolerance and increase gill and gut Na+, K+-ATPase in Pacific salmon, trout, and Atlantic salmon (McCormick and Bern 1989; Madsen 1990a,b; Bisbal and Specker 1991; Specker et al. 1993).In the present study, plasma cortisol levels were elevated in smolts compared to parr in early April 1990. However, in a recent study (Usher et al. 1991) fluid transport across the middle intestine of Atlantic salmon was unaffected by cortisol treatment in vitro. The lack of effect may be because there is a long latent period between the cortisol stimulus and increased fluid transport. Thyroid hormones may also regulate changes in intestinal function since elevated thyroxine levels accompany parr-smolt transformation, thyroxine increases the whole animal transepithelial potential of coho salmon transferred to seawater in April, and because they may act by preparing the gut for the transition to seawater (Dickhoff and Sullivan 1987; Iwata et al. 1987; Specker 1988). Anadromous salmon acquire the ability to maintain salt and water balance in the marine environment during the developmental process of smolting
201
which occurs in fresh water. Preadaptive changes for life at sea have been noted in the gills, kidney, urinary bladder, and gut (McCormick and Saunders 1987). The developmental changes in the gut are of particular interest because of the importance of the intestinal epithelium's other role in nutrient absorption. Known functional changes in the salmonid intestine during parr-smolt transformation are limited to descriptions of increased fluid transport in the posterior intestine of coho salmon in the smolt stage, increased proline influx across the intestine of coho salmon in the smolt stage, regional differences in proline influx in the coho salmon (Collie and Bern 1982; Collie 1985), and increased fluid transport in the middle intestine of smolt-stage Atlantic salmon (Usher et al. 1991). Because smolting is a metabolically demanding, springtime phenomenon, these observations could have been explained as consequences of either development or season. We have shown that in Atlantic salmon the springtime increase in fluid transport across the posterior intestine is developmentdependent and a functional component of preparation for life at sea. The middle intestine of both parr and smolt stage Atlantic salmon undergoes a considerable decrease in fluid transport rate during the spring, perhaps because it is more engaged in nutrient uptake during this season when growth increases.
Acknowledgements We wish to thank Mr. W. Booker and Mr. L.
Lofton at the North Attleboro National Fish Hatchery for their help in maintaining and supplying the Atlantic salmon. Facilities for maintaining salmon in seawater were generously provided by Dr. Ted Durbin. Sean Cornell provided skillful technical assistance. Dr. Allen T. Scholz provided numerous suggestions with the manuscript. This publication is the result of research funded by a URI Undergraduate Research Award (to P.A.V.), NOAA office of Sea Grant, U.S. Department of Commerce, under grant #NA89AA-D-SG082, and by U.S. Department of Interior, Fish and Wildlife Service, Cooperative Research Program Agree-
ment, Project #F-52R, with Rhode Island Division of Fish and Wildlife. The U.S. government is authorized to produce and distribute reprints for governmental purposes notwithstanding any copyright notation that may appear hereon. References cited Bisbal, G.A. and Specker, J.L. 1991. Cortisol stimulates hypoosmoregulatory ability in Atlantic salmon, Salmo salar L. J. Fish Biol. 39: 421-432. Brown, M.E. 1957. Experimental studies on growth. In Physiology of Fishes. Vol. 1, pp. 361-400. Edited by M.E. Brown. Academic Press, New York. Colin, D.A., Nonnotte, G., LeRay, C. and Nonnotte, L. 1985. Na transport and enzyme activities in the intestine of the freshwater and sea-water adapted trout (Salmo gairdneriiR.). Comp. Biochem. Physiol. 81A: 695-698. Collie, N.L. 1985. Intestinal nutrient transport in coho salmon (Oncorhynchus kisutch) and the effects of development, starvation, and seawater adaptation. J. Comp. Physiol. B. 156: 163-174. Collie, N.L. and Bern, H.A. 1982. Changes in intestinal fluid transport associated with smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch (Walbaum). J. Fish Biol. 21: 337-348. Collie, N.L. and Hirano, T. 1987. Mechanisms of hormone actions on intestinal transport. In Vertebrate Endocrinology: Fundamentals and Biomedical Implicatons. Vol. 2, pp. 239-244. Academic Press, New York. Dickhoff, W.W. and Sullivan, C.V. 1987. The thyroid gland in smoltification. Am. Fish. Soc. Symp. 1: 197-210. Hirano, T., Morisawa, M., Ando, M. and Utida, S. 1976. Adaptive changes in ion and water transport mechanism in the eel intestine. In Intestinal Ion Transport. pp. 301-317. Edited by J.W.L. Robinson. MTP Press, Lancaster. Hoar, W.S. 1988. The physiology of smolting salmonids. In Fish Physiology. Vol. 2, part B, pp. 275-343. Edited by W.S. Hoar and D.J. Randall. Academic Press, New York. Holmes, W.N. and Stainer, I.M. 1966. Studies on the renal excretion of electrolytes by the trout Salmo gairdneri'(Richardson). J. Exp. Biol. 44: 33-46. Iwata, M., Nishioka, R.S. and Bern, H.A. 1987. Whole animal transepithelial potential (TEP) of coho salmon during the parr-smolt transformation and effects of thyroxine, prolactin and hypophysectomy. Fish Physiol. Biochem. 3: 25-38. Kristinsson, J.B., Saunders, R.L. and Wiggs, A.J. 1985. Growth dynamics during the development of bimodal lengthfrequency distribution in juvenile Atlantic salmon (Salmo salar L.). Aquaculture 45: 1-20. Langhorne, P. and Simpson, T.H. 1986. The interrelationship of cortisol, gill (Na+K+)ATPase, and homeostasis during the parr-smolt transformation of Atlantic salmon (Salmo salar L.). Gen. Comp. Endocrinol. 61: 203-213.
202 Madsen, S.S. 1990a. Cortisol treatment improves the development of hypoosmoregulatory mechanisms in the euryhaline rainbow trout, Salmo gairdneri. Fish Physiol. Biochem. 8: 45-52. Madsen, S.S. 1990b. The role of cortisol and growth hormone in seawater adaptation and development of hypoosmoregulatory mechanisms in sea trout parr (Salmo trutta trutta). Gen. Comp. Endocrinol. 79: 1-11. McCormick, S.D. and Bern, H.A. 1989. In vitro stimulation of Na+ -K +-ATPase activity and ouabain binding by cortisol in coho salmon gill. Am. J. Physiol. 256: R707-715. McCormick, S.D. and Saunders, R.L. 1987. Preparatory physiological adaptations for marine life of salmonids: Osmoregulation, growth, and metabolism. Am. Fish. Soc. Symp. 1: 211-229. Oide, M. 1967. Effects of inhibitors on transport of water and ion in isolated intestine and Na+-K+ ATPase in intestinal mucosa of the eel. Ann. Zool. Jap. 40: 130-135. Saunders, R.L., Henderson, E.B. and Glebe, B.D. 1982. Precocious sexual maturation and smoltification in male Atlantic salmon (Salmo salar). Aquaculture 28: 211-229. Smith, H.W. 1932. Water regulation and its evolution in the fishes. Quart. Rev. Biol. 7: 1-26. Specker, J.L. 1988. Preadaptive role of thyroid hormones in larval and juvenile salmon: Growth, the gut and evolutionary considerations. Am. Zool. 28: 337-349. Specker, J.L. and Schreck, C.B. 1982. Changes in plasma corticosteroids during smoltification of coho salmon, Oncorhynchus kisutch. Gen. Comp. Endocrinol. 46: 53-58. Specker, J.L., Portesi, D.M., Cornell, S.C. and Veillette, P.A. 1993. Methodology for implanting cortisol in Atlantic salmon
and effects of chronically elevated cortisol on osmoregulatory physiology. Aquaculture (In press). Thorpe, J.E. 1977. Bimodal distribution of length of juvenile Atlantic salmon (Salmo salar L.) under artificial rearing conditions. J. Fish Biol. 11: 175-184. Thorpe, J.E., Morgan, R.I.G., Ottaway, E.M. and Miles, M.S. 1980. Time of divergence of growth groups between potential 1 + and 2 + smolts among sibling Atlantic salmon. J. Fish Biol. 17: 13-21. Usher, M.L., Talbot, C. and Eddy, F.B. 1988. Drinking in Atlantic salmon smolts transferred to seawater and the relationship between drinking and feeding. Aquaculture 73: 237-246. Usher, M.L., Talbot, C. and Eddy, F.B. 1990. Effects of transfer to seawater on digestion and gut function in Atlantic salmon smolts (Salmo salar L.). Aquaculture 90: 85-96. Usher, M.L., Talbot, C. and Eddy, F.B. 1991. Intestinal water transport in juvenile Atlantic salmon (Salmo salar L.) during smolting and following transfer to seawater. Comp. Biochem. Physiol. 100A: 813-818. Young, G. 1986. Cortisol secretion in vivo by the interrenal of coho salmon (Oncorhynchus kisutch) during smoltification: Relationship with plasma thyroxine and plasma cortisol. Gen. Comp. Endocrinol. 63: 191-200. Zaugg, W.S. and McLain, L.R. 1970. Adenosinetriphosphatase activity in gills of salmonids: seasonal variation and salt water influence in coho salmon, Oncorhynchus kisutch. Comp. Biochem. Physiol. 35: 587-596. Zaugg, W.S. and McLain, L.R. 1972. Changes in gill adenosinetriphosphatase activity associated with parr-smolt transformation in steelhead trout, coho, and spring chinook salmon. J. Fish. Res. Bd. Can. 29: 167-171.
