Induction of nitric oxide synthase and associated toxicity in livers of hardhead catfish, Arms felis,
from control and epizootic sites
W. Peter Schoor
Gulf Ecology Division, U.S. Environmental Protection Agency, 1 Sabine Drive, Gulf Breeze,
Florida 32561, USA
October 10, 1996

-------
Introduction
Earlier work with a live channel catfish (Ictaluruspimciatus) pathogen, Edwardsiella ictaluri,
demonstrated the induction of nitric oxide synthase (NOS) in the head kidney, paralleling enteric
septicemia (Hawke et al. 1981; Schoor and Plumb 1994). However, another study exposing
hardhead catfish (.Ariusfelis) to microcystin-LR failed to show the induction of NOS (Schoor et
al. In prep.). Similar studies in mammalian systems have shown the involvement of the cytokines
in the induction of NOS by a mechanism which may possibly have been bypassed by the pure
toxin. In order to pursue these studies, the Florida Marine Research Institute, State of Florida
Department of Natural Resources, St. Petersburg, Florida, was contacted with a request for fish
livers from an epizootic. Dr. Jan H. Landsberg answered the request in November of 1995.

-------
Methods
Hardhead catfish livers were received on dry-ice from the State of Florida Department of Natural
Resources Laboratory in St.Petersburg, Florida, in November of 1995 and stored at -80°C. Three
of the samples were from control sites and eight were from locations where catfish epizootics
occurred during the sampling period.
AFL 10/23/95 #001
#002
#003
#004
#005
#006
AFL 10/25/95 #001
#002
AFL 10/30/95 #001
#001
AFL 11/02/95 #001
Determination of induced nitric oxide synthase (iNOS) activity was conducted by a modification
of the method reported by Schoor and Plumb (1994). Frozen tissues were weighed and minced
before homogenization in a glass tissue grinder at a 2:1 (v/w) ratio with a buffer containing 40
mM Tris (pH 7.9), 0.25 M glucose, 0.1 mM phenylmethylsulfonylfluoride, 3 mM dithiothreitol, 4
fiM flavine adenine dinucleotide (FAD), 5 mM L-arginine, 5 fig/ml aprotinin, 5 pg/ml pepstatin A,

-------
and 1 (ig/ml chymostatin (Stuehr et al. 1991). The homogenates were centrifiiged at 9,000xg for
60 min to yield a crude supernate (E-l), a small amount of which was passed through Dowex-
50W (anionic form) to remove constitutive L-citrulline and L-arginine (E-l-1). The remaining E-l
supernate was centrifuged at 100,000xg for 60 min before passage through Dowex-50W (E-2).
The induced enzyme activities were determined in a buffer containing 40 mM HEPES (pH 7.9), 1
mM nicotinamide adenine dinucleotide phosphate-reduced (NADPH), 1 mM dithiothreitol, 1 mM
L-arginine, 0.1 mM tetrahydrobiopterin, 1 (iM FAD, and 10-25 jj.1 of enzyme in a total volume of
0.6 ml, incubated at 37°C. Aliquots of 100 pi were taken at various times and reacted with 5 pi of
a pre-column derivatization mixture containing 10 mg o-phthalaldehyde, 25 pi P~
mercaptoethanol, and 0.5 ml buffer containg 0.4 M borate, 7 mM EDTA, 0.1% Brij-35, pH 9.4.
The mixture was allowed to react for 2 min in the dark before being chromatographed on a C-18
HPLC column. Elution conditions were: 0.4 ml/min flow rate at an 85%/l 5% mixture of 50 mM
sodium acetate, 4% acetonitrile, pH 5.85 and of 75%/25% acetonitrile/methanol. The
fluorescence was measured at 254 nm and compared to that of L-arginine and L-citrulline
standards. The enzymatic activity, determined from the linear portion of the activity curve, is
expressed in nanomoles of L-citrulline produced per mg of protein per minute. The lowest
detection limit under the above conditions was 0.1 picomole/mg/min.
Investigation of the possible presence of microcystins in the livers was accomplished using high
pressure liquid chromatography and UV detection (Hewlett-Packard HPLC Series 1050 with UV-
DAD). Remaining liver fractions of each sample were combined and extracted with 15 ml of
methanol in a Brinckmann Polytron. The supernate was removed by centrifugation and the

-------
precipitate was re-extracted with 15 ml methanol. The combined methanol/water extracts were
extracted three-times with 5 ml portions of hexane before final evaporation to a volume of about
2 ml. No clean-up was performed before injection. A water/acetonitrile gradient was developed
for the H-P ODS-Hypersil column (10 cm), starting with 100% water, going to 55% water in 18
min, and then to 0% water at 21 min and holding this mixture for 5 min to elute all nonpolar
materials from the column. The MC-LR primary standard peak was confirmed by spectral
comparison to the UV spectrum of MC-LR obtained with a Cary Model 118-C.
Ninety-six hour bioassay tests were performed on the liver extracts using freshly-hatched first
stage zoeae of fiddler crabs (Uca panacea) collected from habitats maintained at EPA GED.
Twenty-five ml of filtered seawater and 25 zoeae were placed into 6 cm Carolina bowls to which
were added 50 fj.1 of the liver extract. The controls received only seawater. Dead zoeae were
counted and removed twice daily.

-------
Results
The physical appearance of three of the liver samples during the first homogenization step was
quite different from all others. Samples 10/23/95: #004, #005 and #006 had a bright red
precipitate, in contrast to brown or reddish-brown precipitates from the remaining samples (Table
1). These samples were later identified as controls. The purpose of the Dowex-50W treatment
was to remove L-arg, which was originally added to stabilize iNOS during purification, as well as
to remove constitutive L-citr. The treatment had no effect on the color of the supernates.
Induced nitric oxide synthase activity was measured in all liver samples. Initial activity
measurements indicated a rapid depletion of L-arginine (L-arg) without concommitant increase in
L-citrulline (L-citr), the substrate for iNOS, and its metabolite, respectively. While L-citr was
present initially, no consistent increases were found during the assays. Since the disappearance of
L-arg was most likely due to arginase activity, L-valine (L-val) was added to inhibit that activity
(Table 2). L-val did not inhibit the iNOS activity of the positive iNOS control (murine
macrophage homogenate obtained from Dennis J. Stuehr, Cleveland Clinic Foundation). Addition
of L-citr to reaction mixtures containing liver homogenates showed no degradation, indicating the
absence of enzymes degrading L-citr, thus allowing for the use of the appearance of L-citr as an
indicator of iNOS activity. This was also found to be true for iNOS activity of head kidney
homogenates from channel catfish (Schoor and Plumb 1994). The time courses for the specific
iNOS activity of the three active liver preparations are shown in Tables 3, 4 and 5.
Table 6 shows the iNOS activities for all catfish liver samples and, for comparative purposes,
activities from EdwardsieHa-exposed channel catfish (Ictaluruspunctatus), as well as the

-------
activities from murine macrophage homogenates which were used as a positive control. No
differences were found in the activities from E-l-1 and E-2 homogenates.
All liver extracts were analyzed by HPLC for the presence of microcystins and nodularins, but
none were found at the detection limit, 5 ng microcystin-LR/g liver tissue. The toxicity of the
extracts was assayed by exposing first-stage zoeae of fiddler crabs (Uca panacea) to 50 |il of
aqueous extract in two separate experiments. The results are shown in Tables 7 and 8.

-------
Discussion
When the initial liver sample homogenates were prepared for iNOS determinations, it became
clear that samples 10/23/95: #004, #005 and #006 behaved quite differently than did any other
liver sample. While there was little difference in the color of the supernates, the color of the
precipitates was a bright red in the above samples, whereas it was mostly a muddy dark brown in
the others. The brown precipitates were packed much more tightly than the red ones after
centrifugation. It was learned later that the above samples came from a control site.
The time courses of the appearance of L-citrulline, the byproduct of the formation of nitric oxide
from L-arginine, from samples 10/23/95: #001, #002 and #003 are shown in Tables 3, 4 and 5.
The iNOS activities from those samples were calculated from regression analyses and are shown
in Table 6. The above samples show an induced NOS activity about ten-fold higher than found in
the controls or the other samples. There appears to be no correlation to any initial concentration
of L-citrulline.
HPLC analyses of the liver extracts revealed no microcystins or nodularins, reducing the chance
of involvement of algae which produce these groups of toxins. However, bioassays of the liver
extracts show toxicity which parallels the iNOS activity (Tables 7 and 8). It should be noted that
this assay system is under development for sediment toxicity testing and is not the most suitable
biassay system for water soluble toxins. It is remarkable that in both tests the relative mortality in
samples 10/23/95: #001, #002 and #003 is close, in spite of the drastic differences in the control
mortalities.

-------
Conclusion
In the absence of more detailed information on liver histopathology, it is very difficult to associate
any cause with the findings above. It is tempting to speculate that an organism is responsible
which is capable of turning-on iNOS in certain hepatocytes either via a toxin, cell wall fragment
(LPS) or some cytokine-related mechanism, not so much because of the induction of NOS but
because of the paralleling toxicity data.
Acknowledgements
The authors thanks Dr. Jan H. Landsberg, Florida Marine Research Institute, Florida Department
of Environmental Protection, for providing the catfish liver samples. Mention of trade names does
not imply endorsement by the U.S. Environmental Protection Agency.

-------
Table 1. Physical appearance of liver samples after first homogenization and ultracentrifugation at
9,000xg for 60 minutes.
Physical appearance
Sample identification
E-l
Precipitate
10/23/95 001
clear, amber
brown/red
002"
clear, red/brown
brown/red
003
clear, brown
brown
004
clear, amber
bright red
005
clear, amber
bright red
006
clear, amber
bright red
10/25/95 001
clear, amber
brown
002
clear, amber
brown/red
10/30/95 001
clear, brown
brown
001
clear, brown
brown
11/02/95 001
clear, amber
brown/red

-------
Table 2. Effects of addition of L-valine on the disappearance of L-arginine.
Time
L-arginine
L-arginine + L-valine
(min)
(nanomoles)
(nanomoles)
0
305
3051
60
250
310
120
215
300
1 Value normalized to 305

-------
Table 3. Time course of induced nitric oxide synthase activity from catfish liver 10/23/95 001, E-
1-1 preparation.
Time
(min)
0
45
105
165
220
275
L-citrulline
(picomoles)
38
68
100
190
240
390
L-arginine
(nanomoles)
350
370
370
375
360
370

-------
Table 4. Time course of induced nitric oxide synthase activity from catfish liver 10/23/95 002, E-
1-1 preparation.
Time
(min)
0
60
90
120
180
330
L-citrulline
(picomoles)
31
84
110
150
220
330
L-arginine
(nanomoles)
370
375
370
355
360
375

-------
Table 5. Time course of induced nitric oxide synthase activity from catfish liver 10/23/95 003, E-
1-1 preparation.
Time	L-citrulline	L-arginine
(min)	(picomoles)	(nanomoles)
0	29	375
45	50	370
80	" 86	355
120	100	360
250	360	360
340	630	350

-------
Table 6. Induced NOS activity in hardhead catfish, Arins fells, liver homogenates from various
sites and comparative iNOS activities from other sources.
Initial
L-Citrulline (T=0)	iNOS Activity (E-l-1)
	Liver Sample	(picomoles)	(picomoles/mg/min)
10/23/95 001 38	4.9
002	31	6.2
003	29	4.2
004	20	0.4
005	40	0.4
006	26	0.31
10/25/95 001 19	0.3
002 12	0.4
10/30/95 001 11	0.5
11/02/95 001 26	0.71
Murine Macrophage Homog.	5,3002
1253
Channel Catfish4
Exposed	160
Control	2.0
1	Values from E-2 fraction
2	Murine macrophage homogenates, when originally received (Stuehr)
3	Murine macrophage homogenates, stored at -80 °C for two years (Stuehr)
4	From head kidney (Schoor and Plumb 1994)

-------
Table 7. Toxicity of liver extracts to first-stage zoeae of fiddler crabs, Ucapanacea. Test date
6/24/96.
Liver sample
Liver extract1
(mg)
22 hrs
Relative mortality2 (Percent mortality)
29 hrs 46 hrs 53 hrs
10/23/95 001 -
59
0.9(52)
1.0(59)
1.4(81)
1.7(100)
002
33
1.2(39)
1.3(43)
2.1(70)
2.8(91)
003
83
0.4(30)
0.5(39)
0.6(48)
1.2(96)
004
140
0.1(17)
0.1(17)
0.7(96)
0.7(100)
005
430
0.1(52)
0.2(60)
0.2(60)
0.2(60)
006
130
0.3(32)
0.3(44)
0.4(52)
0.7(88)
10/25/95 001
130
0.3(43)
0.4(47)
0.5(70)
0.7(90)
002
140
0.1(19)
0.2(31)
0.4(58)
0.7(96)
10/30/95 001
160
0.2(24)
0.2(36)
0.4(56)
0.4(69)
11/02/95 001
170
0.2(35)
0.2(35)
0.3(52)
0.5(87)
Control

(18)
(31)
(63)
(80)
'As used per assay; liver extract equivalent to frozen liver weight in 50 ul aqueous solution
2Percent mortality divided by relative mortality

-------
Table 8. Toxicity of liver extracts to first-stage zoeae of fiddler crabs, Uca panacea. Test date
Ull 196.


Relative mortality2 (Percent mortality)
Liver sample
Liver extract1
44 hrs
53 hrs

(mg)


10/23/95 001 .
59
1.6(96
1.7(100)
002
33
2.5(83)
3.0(100)
003
83
1.2(96)
1.2(100)
004
140
0.7(96)
0.7(100)
005
430
0.2(93)
0.2(93)
006
130
0.6(74)
0.7(93)
10/25/95 001
130
0.4(52)
0.7(88)
002
140
0.3(48)
0.5(74)
10/30/95 001
160
0.5(78)
0.6(100)
11/02/95 001
170
0.4(74)
0.5(81)
Control

(20)
(25)
'As used per assay; liver extract equivalent to frozen liver weight in 50 jj.1 of aqueous solution
^Percent mortality divided by relative mortality

-------
References
Hawke, J.P., McWhorter, A.C., Steigerwalt, A.G., Brenner, D.J. (1981). Edwardsiella ictalnri
sp. nov., the causative agent of enteric septicemia of catfish. Intern. J. Syst. Bact. 31: 396-400
Schoor, W.P., Plumb, J.A. (1994). Induction of nitric oxide synthase in channel catfish Ictahmis
punctatus by Edwardsiella ictaluri. Dis. Aquat. Org. 19:153-155
Stuehr, D.J., Cho, H.J., Kwon, N.S., WeiseM.F., Nathan, C.F. (1991). Purification and
characterization of the cytokine-induced macrophage nitric oxide synthase: An FAD- and FMN-
containing flavoprotein. Proc. Natl. Acad. Sci. USA. 88: 7773-7777

-------