ECOTOXICOLQGY
AND
ENVIRONMENTAL
SAFETY
23, 118- 13 1 ( 1992)
Coplanar PCBs in Fish and Mussels from Marine and Estuarine Waters of New York State’ CHIA-SWEE HONG,* BRIAN BUSH, AND Ju XIAO Wadsworth Center for Laboratories and Research, New York State Department of Health, P.O. Box 509, Albany, New York 12201-0509 Received May 16, 1991 Thirty samples of striped bass from marine and estuarine waters of New York State, six samples of mussels from Long Island Sound, and one composite sample of freshwater mussels from Troy were analyzed by a recently developed method which combines sulfuric acid cleanup, carbon chromatography, and high-resolution gas chromatography for the determination of non-orthoand mono-ortho-substituted polychlorinated biphenyls (PCBs), which are biologically active congeners of PCBs and approximate isostereomers of 2,3,7,8-tetrachlorodibenzo-P-dioxin of (2,3,7,8TCDD). Non-ortho coplanar PCBs ranged from 0.2 to 37.1 ppb in fish. The highest concentrations of 37.1 rig/g of 3,3’,4,4’-tetrachlorobiphenyl (77) and 7.5 rig/g of 3,3’,4,4’,5-pentachlorobiphenyl (126) were detected in a fish caught near Troy/Albany, New York. Mono-orth&substituted PCBs ranged from 0.4 to 790 ppb in fish, with the major components identified as 2,3’,4,4’,5penta(118) and 2,3,3’,4,4’-pcntachlorobiphenyls (105). When concentrations were converted to 2,3,7,8TCDD picogram equivalents, 99% of the equivalents were derived from three congeners (77,105, and 126), 3,3’,4,4’,5pentachlorobiphenyl(l26) accounting for approximately 80% of the activity. At Troy pg/g TCDD equivalents derived from PCB were approximately 3000, whereas the actual 2,3,7,8-TCDD concentration was only 20 pg/g; hence, accurate coplanar PCB measurement is important. 0 1992 Academic Press, Inc.
INTRODUCTION Polychlorinated biphenyl (PCB) contamination of the Hudson River and the bioaccumulation of PCBs in species such as striped bass began nearly 30 years ago when PCBs were first discharged into the river and is still evident today. Among the theoretically possible 209 PCB congeners, 20 of them can attain coplanarity due to the lack of chlorine substitution at the ortho positions lateral to the biphenyl rings (Erickson, 1986). Quantitative structure-activity relationship studies indicated that PCBs with non-or-rho, two pm-u, and at least two meta chlorines are isostereomers of the most toxic 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and elicit toxic and biologic responses typical of TCDD (Tanabe et al., 1987; Safe, 1984). Several studies suggest that toxicity resembling that of TCDD of technical PCB mixtures may be associated with the presence of low levels of particular coplanar PCBs having four or more chlorine atoms at both paru and metu positions such as 3,4,4’5-,3,3’,4,4’-tetra-, 3,3’,4,4’,%penta(IUPAC Nos. 8 1, 77, 126), and 3,3’,4,4’,5,5’-hexachlorobiphenyl (IUPAC No. 169) (Safe and Hutzinger, 1987; Luu et al., 1985; Yoshimura et al., 1978; Poland and Glover, 1977). The other eight mono-ortho analogs of the coplanar PCBs, namely 2,3,4,4’,5-, 2,3’,4,4’,5’-, 2,3,3’,4,4’-, and 2,3’,4,4’,5-penta- (IUPAC Nos. 114, 123, 105, ’ The U.S. Government’s right to retain a nonexclusive royalty-free license in and to the copyright covering this paper, for governmental purposes, is acknowledged. 2 To whom correspondence should be addressed. 0147-6513192 $3.00 Copyright 0 1992 by Academic Pres Inc. All rights of reproduction in any form resewed.
118
COPLANAR
PCBs
119
118); 2,3,3’,4,4’,5-, 2,3’,4,4’,5,5’-, and 2,3,3’,4,4’,5’-hexa- (IUPAC Nos. 156, 167, 157); and 2,3,3’,4,4’,5,5’-heptachlorobiphenyl (IUPAC No. 189), also have moderate dioxinlike toxicity. These mono- and non-o&o-substituted PCBs constitute a considerable proportion of commercial PCB preparations. PCB residues have been identified in almost every component ofthe global ecosystem (Norstrom et al., 1988; Nisbet and Sarofim, 1972). Hence we expected a widespread occurrence of mono- and non-orthu-substituted PCB residues also. It must be borne in mind, however, that several other toxic effects of PCB have been reported (Kimbrough, 198 1). Neurochemical effects that appear not to be related to the dioxin-like coplanar PCB, but more to the di-o&o-substituted PCB, have been reported recently (Seegal et al., 1991). Hence complete congener-bycongener elucidation of PCB residues is needed for accurate risk assessment of the PCB congener residues. A carbon column was used to separate mono- and non-ortho-substituted PCBs from other PCBs (Tanabe et al., 1987; Hong and Bush, 1990) to achieve higher sensitivity and accuracy by GC/EC measurements on a DB-5 column. Thirty striped bass samples from the Hudson River Estuary, the Atlantic shore of Long Island, and Long Island Sound, one composite sample of freshwater mussels from Troy, New York, and six mussel samples from Long Island Sound were analyzed for the above 12 PCB congeners. EXPERIMENTAL
Reagents. All solvents used were nanograde from Burdick & Jackson, Muskegon, Michigan. The Aroclors were obtained from the U.S. Environmental Protection Agency, Research Triangle Park, North Carolina, and PCB congeners from AccuStandard Inc., New Haven, Connecticut. LPS-2 low-pressure silica gel (particle size, 37-53 pm) was obtained from Whatman Lab Sales, Hillsboro, Oregon, and Super-A activated carbon (AX-2 1) from Anderson Development Company, Adrian, Michigan. METHODS PCBs were determined according to the method of Hong and Bush (1990) which, in brief, contained the following steps: (a) extraction and H2S04 cleanup, (b) fractionation and enrichment of mono- and non-o&o-substituted PCBs on an activated carbon-silica gel ( 1: 12) column, (c) isomer-specific determinations using HRGC/EC, and (d) confirmation on CC/MS on the M+ and (M + 2)+ cluster ions of coplanar PCBs (Hong and Bush, 1991). PCBs with two to four ortho substituents were eluted in the first 50 ml of 20% dichloromethane in hexane, and mono- and non-ortho-substituted PCBs were eluted in the second fraction of 150 ml of 50% benzene in ethyl acetate from carbon chromatography. High-resolution GC/EC analysis. The sample extracts were analyzed on a HewlettPackard 5840A gas chromatograph equipped with a Ni-63 electron-capture detector using a 25-m 5% phenyl methyl silicone fused capillary column (HP-Ultra II). The oven temperature was held at 100°C for 2 min and then raised lO”C/min for 6 min and then 3”C/min for 40 min. The carrier gas used was helium (27.5 psi) and the make-up gas was 5% argon-methane (40 ml/min). The temperature of the injection port was 250°C and the electron-capture detector was maintained at 250°C. The calibration mixture that was used to quantitate two-to-four ortho-substituted PCBs in
120
HONG, BUSH, AND XIAO
the sample was a 1: 1: 1: 1 mixture of 200 @ml each of Aroclors 1221, 1016, 1254, and 1260. A coplanar PCB standard that contained 10 rig/ml each of four non-o&z+ and eight mono-o&o-substituted PCBs was used to quantitate mono- and non-or& substituted PCBs. Totally 37 specimens including striped bass and mussels were analyzed for monoand non-or&-substituted PCBs. The analyzed fish were full grown at or near sexual maturity. An analysis for 78 PCB congeners has already been reported (Bush et al., 1989). RESULTS
AND
DISCUSSION
Figure 1 shows the chromatogram of a 1: 1: 1: 1 mixture of Aroclors 1221, 1016, 1254, and 1260 (used for GC calibration in our congener-specific method) spiked with the 12 mono- and non-o&&substituted PCBs and the chromatogram of the 12 monoand non-o&u-substituted PCBs. It is shown that 3,4,4’,5-tetra-, 3,3’,4,4’-tetra-, 2,3’,4,4’,5’-penta-, 3,3’,4,4’,5-penta-, 2,3’,4,4’,5,5’-hexa-, 2,3,3’,4,4’,5-hexa-, and 2,3,3’,4,4’,5’-hexachlorobiphenyl were not resolved from PCB components having two to four ortho chlorine substitution on the DB-5 column. Therefore, carbon separation is necessary for the quantitation of individual mono- and non-o&o-substituted PCBs.
A
FIG. 1. Cbromatograms on a DB-5 column of(A) a 1:l:l:l mixture ofAroclors 1221, 1016, 1254, and 1260 spiked with 12 mono- and non-ortho-substituted PCBs. (B) Twelve mono- and non-OrthcFsubstituted PCBs.
COPLANAR
PCBs
121
Figure 2 shows the chromatograms of a l&h from eastern Long Island Sound, fraction 1 contains two to four o&o-substituted PCBs, and fraction 2 contains mono- and non-or&-substituted PCBs. Levels of those mono- and non-orrho-substituted PCBs in fraction 2 were determined by comparing with appropriate gas chromatogram peaks of reference solutions. Table 1 presents the concentrations of individual mono- and non-o&o-substituted PCBs, total mono- and non-ortho-substituted PCBs, and total PCBs in the striped bass collected from the indicated area as shown in Fig. 3. Figure 3 shows a map of the New York Bight, indicating the sampling locations for this study (the striped bass in this study are not the same fish as those in 1990 paper, although they were collected at the same time), and the mean total concentrations of mono- and non-ortho-substituted PCBs for the striped bass sampled in each area. Table 2 shows the concentration of total mono- and non-o&o-substituted PCBs and total PCBs in striped bass from western and eastern Long Island Sound, the western and eastern Atlantic shore, and the Hudson River Estuary. Figure 4 shows each mono- and non-or&-substituted PCB found in the fish plotted in order of increasing distance from the Troy Dam, assuming free passage from Long Island Sound to the Hudson River. The Troy Dam is the upstream limit of the spawning migration. The figure shows a general trend in reduction of the amount of each mono-
FR-1
FA-2
FIG. 2. Chromatograms of the first and second fractions of a fish from Long Island Sound after fractionation on a carbon column.
PCB
coplanar PCBs hki9 Total PCBs bm % total coplanar PCBs
Total
3,4,4:5 3,3’,4,4’ 2,3’,4,4:5’ 2,3’,4,4’,5 2,3,4,4:5 2,3,3’,4,4’ 3,3’,4,4’,5 2,3:4,4’,5,5’ 2,3,3’,4,4’,5 2 >.3 3’ ,.4 4’ 35’ 3,3*,4,4:5,5 2 ,>3 3’ ,>4 4’ ,T 5 5’
congener
HUD(I0)
24.1
8.4
7.1
2.02
2.4 37.1 0.6 779 112 562 7.5 232 202 67.1 18.8
6.7
20.5
1.37
0.5 24.9 581 77.6 324 3.3 164 131 49.8 16.4
Albany/Troy
20.0
1.41
27.0 0.8 588 85.0 367 5.9 152 124 48.8 14.4
-@
IUPAC HUD(O) NO. Troy
81 77 123 118 114 105 126 167 156 157 169 189
1
-
19.9
5.8
7.6
1.16
8.9
21.9 0.6 499 57.7 339 4.1 108 90.4 31.0 -
21.1
1.61
4.1
790 81.8 359 4.7 188 146 5.0 -
4.8 29.1 -
Pt./Bay
HUD(168)
Croton
5.8
4.14
0.24
4.6
95.7 21.5 44.3 3.9 31.6 19.5 12.0 -
7.0 -
-
9.1
4.64
0.42
2.3
191 60.9 109 6.1 9.4 16.8 1.9 -
19.2 -
-
Bridge
11.3
2.01
0.23
1.2
6.4 1.2 85.2 24.9 45.3 0.9 34.4 20.4 7.0 -
-
Bridge
HUD(214) HUD(190) George Tappanzee Washington
8.1
6.52
0.53
1.8
300 25.3 113 49.6 27.8 12.5 -
4.3 -
-
NW(241) Eatons Neck
CONCENTRATIONS (ng/gwet W-QOFMONO-ANDNON-~&~-SUBSTITUTED FROMMARINEANDESTUARINEWATERSOFNEWYORKSTATE
TABLE
7.6
3.57
0.27
123 27.6 49.3 1.6 37.9 17.5 7.5 0.9
7.0 -
-
8.0
2.01
0.16
1.6
10.7 60.8 28.1 28.7 4.3 9.4 13.6 7.4 -
NW(248) Lilco
11.2
3.76
0.42
3.0
13.4 1.8 154 52.8 101 3.9 50.1 26.3 12.1 -
-
-
8.9
3.72
0.33
0.5
7.7 153 39.6 64.7 0.7 38.4 20.4 5.6 -
PCBs IN STRIPED BASS
10.0
4.02
0.40
7.2
10.7 1.0 137 46.3 113 1.9 43.1 28.1 11.5 -
-
SW(2368) Breezy Pt.
9.0
6.52
0.59
3.3
266 49.4 133 2.8 60.6 45.1 16.6 -
0.6 11.5 -
Note. HUD, Hudson in parentheses indicates ’ CO.1 rig/g.
PCBS
coplanar
4.9
River Estuary; distance (km)
4.8
4.7 1
3.54
( MJg)
o/total
0.23
0.17
(NM Total PCBs
4.3 I.1 73.4 30.1 56.6 1.1 26.3 23.2 9.9 -
8.8
4.22
6.4
3.12
0.20
93.7 21.1 41.6 0.8 24.0 10.5 5.2 0.4
0.2 4.2 9.6 -
-
7.7
5.33
0.41
1.8
200 41.2 58.6 1.6 45.8 33.6 14.4 -
Inlet
Fire Island
SE(328)
NW, Western Long Island Sound; from Troy Dam.
7.9
3.69
0.37
118 82.9 88.5 2.6 12.4 20.5 21.7 10.0
116 42.0 56.2 0.9 34.9 21.8 13.5 1.7
0.29
15.6 -
-
Beach
Jones
0.3 6.7 -
Inlet
3.3
-
Rockaway
NO.
81 77 123 118 114 105 126 167 156 157 169 189
SW(275)
IUPAC
6.8 0.5 66.0 19.9 36.5 0.8 19.4 13.2 4.6 0.7
3,4.4’,5 3,3:4,4 2,3’,4,4’,5’ 2$,4,4’S 2.3,4,4’,5 2.3,3’,4,4’ 3,3’,4,4’,5 2,3:4,4’,5,5’ 2,3,3’,4,4’,5 2,3,3’,4,4’,5’ 3,3:4,4:5,5 2,3,3’,4,4’,5,5’ Total coplanar PCBS
PCB congener
SE(302)
SW, Western
9.6
5.94
Atlantic
6.6
6.03
0.40
9.6
12.7 0.5 116 40.6 90.5 2.5 63.2 50.7 17.4 -
-
l-Continued
SE(342) Fire Island
0.57
258 70.6 118 1.5 58.0 29.4 15.0 7.7
1.5 13.8 -
TABLE
Atlantic
10.3
1.36
0.14
1.4
2.1 0.4 52.2 17.7 20.9 1.1 18.4 15.8 8.1 -
SE(362) Moriches Bay
Shore: SE, Eastern
10.2
2.35
0.24
1.5
1.1 97.3 26.1 49.8 1.6 28.9 19.6 7.8 -
-6.8
SE(348) Westhampton Beach
8.9
1.12
0.10
12.6 0.5 13.8 33. I 2.9 2.5
34.1 -
5.1 -
-
GUI1 Island
Little
NE(382) SE(389)
SE(395)
6.6
2.42
0.16
9.4 54.7 17.8 38.9 4.9 18.1 10.9 4.4 -
-
Amagansett
Long Island Sound.
10.8
1.30
0.14
5.0 14.0 2.0
4.1 1.0 55.8 20.0 35.8 -
-
Shinnecock
Shore: NE, Eastern
12.4
1.61
0.20
0.8
5.3 31.8 65.5 20.5 37.2 1.4 22.9 14.6 4.8 -
-
NE(375) Orient Pt.
SE(422)
Number
6.5
3.86
0.25
0.5 4.5 0.7 137 19.8 38.8 0.9 27.3 10.9 6.2 -
Montauk
E
0
124
HONG, BUSH, AND XL40
FIG. 3. Sampling points in theNewYork BightandHudson River showing concentrations of totalmonoand non-o&o-substituted PCBs at each point &@g wet weight).
and non-o&u-substituted PCBs, 3,3’,4,4’-tetrachlorobiphenyl was found to be higher than 3,3’,4,4’,5-pentachlorobiphenyl in all the cases. The highest concentrations of about 37.1 rig/g 3,3’,4,4’-tetrachlorobiphenyl and 7.5 rig/g of 3,3’,4,4’,5pentachlorobiphenyl were detected in a fish caught near Troy/Albany, New York. 3,3’,4,4’,5,5’Hexachlorobiphenyl was not detected in any samples. Mono-o&o-substituted PCBs ranged from 0.4 to 790 ppb. The main components in all the species were 2,3’,4,4’,5and 2,3,3’,4,4’-pentachlorobiphenyl, which reached 30-56% and 12-27s oftotal monoand non-o&o-substituted PCBs, respectively. TABLE 2 CONCENTRATION (fig/g wet wt) OF TOTAL MONO- AND NON-O~~~&~JB~TITI.JTED PCBs AND TOTAL PCBs IN STRIPED BASS FROM FIVE AREAS OF NEW YORK BIGHT
Concentration (mean f standarderror of the mean) Area’
Number of fish
Total coplanars
Total PCBs
NW NE SW
4 2 6
SE
10
0.35 -+ 0.08 0.15 & 0.05 0.34 +-0.06 0.29 t- 0.05 1.06 +- 0.24
3.97 +- 0.94 1.37f 0.24 4.37 + 0.46 3.59 +- 0.56 14.55+- 3.25
HUD
8
BArea: NW, Western Long Island Sound; NE, Eastern Long Island Sound; SW, Western Atlantic Shore; SE, Eastern Atlantic Shore; HUD, Hudson River Estuary.
COPLANAR
PCBs
125
800 600 400 200 0
FIG. 4. Concentration of mono- and non-orfho-substituted PCBs in fish plotted in order of increasing distancefromtheTroyDam.(l)3,4,4’,5;(2)3,3’,4,4’;(3)2,3’,4,4’,5’;(4)2,3’,4,4’,5;(5)2.3,4,4’,5;(6)2,3,3’,4.4’; (7) 3,3’,4,4’,5: (8) 2,3’.4,4’,5,5’; (9) 2,3,3’,4,4’,5; (10) 2,3,3’,4,4’,5’; (11) 3,3’,4,4’,5,5’: (12) 2.3,3’,4.4’.5,5’.
Table 3 presents the concentrations of individual mono- and non-ortho-substituted PCBs, total mono- and non-ortho-substituted PCBs, and total PCBs in six mussels from Long Island Sound and one composite of freshwater mussels from Troy. Figure 5 shows the pattern of mono- and non-o&o-substituted PCBs present in mussels from Troy and Long Island Sound. The sample from Troy, nearest the pollution source at Fort Edward, New York, has the higher level of total mono- and nono&o-substituted PCBs, 112.9 ppb. Mussels from Long Island Sound contain only 8.5-18 ppb of total mono- and non-o&o-substituted PCBs. Figure 6 shows the total PCBs and total mono- and non-ortho-substituted PCBs in fish plotted in order of increasing distance from the Troy Dam. Clearly the total PCB burden of the samples is strongly correlated with the total mono- and non-orthosubstituted PCB burden. Figure 7 shows the correlation of total coplanar PCBs with total PCBs in striped bass. The correlation coefficient (R) is 0.98 and the regression coefficient is 13 + 0.5
COPLANAR
PCB
FIG. 5. Concentration of mono- and non-otiho-substituted PCBs in one composite of freshwater mussels from Troy and six mussels from Long Island Sound. Mono- and non-ortho-substituted PCBs are the same as those in Fig. 4.
3
11 Duplicateofsamesample. b
AND
ENVIRONMENTAL
SAFETY
23, 118- 13 1 ( 1992)
Coplanar PCBs in Fish and Mussels from Marine and Estuarine Waters of New York State’ CHIA-SWEE HONG,* BRIAN BUSH, AND Ju XIAO Wadsworth Center for Laboratories and Research, New York State Department of Health, P.O. Box 509, Albany, New York 12201-0509 Received May 16, 1991 Thirty samples of striped bass from marine and estuarine waters of New York State, six samples of mussels from Long Island Sound, and one composite sample of freshwater mussels from Troy were analyzed by a recently developed method which combines sulfuric acid cleanup, carbon chromatography, and high-resolution gas chromatography for the determination of non-orthoand mono-ortho-substituted polychlorinated biphenyls (PCBs), which are biologically active congeners of PCBs and approximate isostereomers of 2,3,7,8-tetrachlorodibenzo-P-dioxin of (2,3,7,8TCDD). Non-ortho coplanar PCBs ranged from 0.2 to 37.1 ppb in fish. The highest concentrations of 37.1 rig/g of 3,3’,4,4’-tetrachlorobiphenyl (77) and 7.5 rig/g of 3,3’,4,4’,5-pentachlorobiphenyl (126) were detected in a fish caught near Troy/Albany, New York. Mono-orth&substituted PCBs ranged from 0.4 to 790 ppb in fish, with the major components identified as 2,3’,4,4’,5penta(118) and 2,3,3’,4,4’-pcntachlorobiphenyls (105). When concentrations were converted to 2,3,7,8TCDD picogram equivalents, 99% of the equivalents were derived from three congeners (77,105, and 126), 3,3’,4,4’,5pentachlorobiphenyl(l26) accounting for approximately 80% of the activity. At Troy pg/g TCDD equivalents derived from PCB were approximately 3000, whereas the actual 2,3,7,8-TCDD concentration was only 20 pg/g; hence, accurate coplanar PCB measurement is important. 0 1992 Academic Press, Inc.
INTRODUCTION Polychlorinated biphenyl (PCB) contamination of the Hudson River and the bioaccumulation of PCBs in species such as striped bass began nearly 30 years ago when PCBs were first discharged into the river and is still evident today. Among the theoretically possible 209 PCB congeners, 20 of them can attain coplanarity due to the lack of chlorine substitution at the ortho positions lateral to the biphenyl rings (Erickson, 1986). Quantitative structure-activity relationship studies indicated that PCBs with non-or-rho, two pm-u, and at least two meta chlorines are isostereomers of the most toxic 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and elicit toxic and biologic responses typical of TCDD (Tanabe et al., 1987; Safe, 1984). Several studies suggest that toxicity resembling that of TCDD of technical PCB mixtures may be associated with the presence of low levels of particular coplanar PCBs having four or more chlorine atoms at both paru and metu positions such as 3,4,4’5-,3,3’,4,4’-tetra-, 3,3’,4,4’,%penta(IUPAC Nos. 8 1, 77, 126), and 3,3’,4,4’,5,5’-hexachlorobiphenyl (IUPAC No. 169) (Safe and Hutzinger, 1987; Luu et al., 1985; Yoshimura et al., 1978; Poland and Glover, 1977). The other eight mono-ortho analogs of the coplanar PCBs, namely 2,3,4,4’,5-, 2,3’,4,4’,5’-, 2,3,3’,4,4’-, and 2,3’,4,4’,5-penta- (IUPAC Nos. 114, 123, 105, ’ The U.S. Government’s right to retain a nonexclusive royalty-free license in and to the copyright covering this paper, for governmental purposes, is acknowledged. 2 To whom correspondence should be addressed. 0147-6513192 $3.00 Copyright 0 1992 by Academic Pres Inc. All rights of reproduction in any form resewed.
118
COPLANAR
PCBs
119
118); 2,3,3’,4,4’,5-, 2,3’,4,4’,5,5’-, and 2,3,3’,4,4’,5’-hexa- (IUPAC Nos. 156, 167, 157); and 2,3,3’,4,4’,5,5’-heptachlorobiphenyl (IUPAC No. 189), also have moderate dioxinlike toxicity. These mono- and non-o&o-substituted PCBs constitute a considerable proportion of commercial PCB preparations. PCB residues have been identified in almost every component ofthe global ecosystem (Norstrom et al., 1988; Nisbet and Sarofim, 1972). Hence we expected a widespread occurrence of mono- and non-orthu-substituted PCB residues also. It must be borne in mind, however, that several other toxic effects of PCB have been reported (Kimbrough, 198 1). Neurochemical effects that appear not to be related to the dioxin-like coplanar PCB, but more to the di-o&o-substituted PCB, have been reported recently (Seegal et al., 1991). Hence complete congener-bycongener elucidation of PCB residues is needed for accurate risk assessment of the PCB congener residues. A carbon column was used to separate mono- and non-ortho-substituted PCBs from other PCBs (Tanabe et al., 1987; Hong and Bush, 1990) to achieve higher sensitivity and accuracy by GC/EC measurements on a DB-5 column. Thirty striped bass samples from the Hudson River Estuary, the Atlantic shore of Long Island, and Long Island Sound, one composite sample of freshwater mussels from Troy, New York, and six mussel samples from Long Island Sound were analyzed for the above 12 PCB congeners. EXPERIMENTAL
Reagents. All solvents used were nanograde from Burdick & Jackson, Muskegon, Michigan. The Aroclors were obtained from the U.S. Environmental Protection Agency, Research Triangle Park, North Carolina, and PCB congeners from AccuStandard Inc., New Haven, Connecticut. LPS-2 low-pressure silica gel (particle size, 37-53 pm) was obtained from Whatman Lab Sales, Hillsboro, Oregon, and Super-A activated carbon (AX-2 1) from Anderson Development Company, Adrian, Michigan. METHODS PCBs were determined according to the method of Hong and Bush (1990) which, in brief, contained the following steps: (a) extraction and H2S04 cleanup, (b) fractionation and enrichment of mono- and non-o&o-substituted PCBs on an activated carbon-silica gel ( 1: 12) column, (c) isomer-specific determinations using HRGC/EC, and (d) confirmation on CC/MS on the M+ and (M + 2)+ cluster ions of coplanar PCBs (Hong and Bush, 1991). PCBs with two to four ortho substituents were eluted in the first 50 ml of 20% dichloromethane in hexane, and mono- and non-ortho-substituted PCBs were eluted in the second fraction of 150 ml of 50% benzene in ethyl acetate from carbon chromatography. High-resolution GC/EC analysis. The sample extracts were analyzed on a HewlettPackard 5840A gas chromatograph equipped with a Ni-63 electron-capture detector using a 25-m 5% phenyl methyl silicone fused capillary column (HP-Ultra II). The oven temperature was held at 100°C for 2 min and then raised lO”C/min for 6 min and then 3”C/min for 40 min. The carrier gas used was helium (27.5 psi) and the make-up gas was 5% argon-methane (40 ml/min). The temperature of the injection port was 250°C and the electron-capture detector was maintained at 250°C. The calibration mixture that was used to quantitate two-to-four ortho-substituted PCBs in
120
HONG, BUSH, AND XIAO
the sample was a 1: 1: 1: 1 mixture of 200 @ml each of Aroclors 1221, 1016, 1254, and 1260. A coplanar PCB standard that contained 10 rig/ml each of four non-o&z+ and eight mono-o&o-substituted PCBs was used to quantitate mono- and non-or& substituted PCBs. Totally 37 specimens including striped bass and mussels were analyzed for monoand non-or&-substituted PCBs. The analyzed fish were full grown at or near sexual maturity. An analysis for 78 PCB congeners has already been reported (Bush et al., 1989). RESULTS
AND
DISCUSSION
Figure 1 shows the chromatogram of a 1: 1: 1: 1 mixture of Aroclors 1221, 1016, 1254, and 1260 (used for GC calibration in our congener-specific method) spiked with the 12 mono- and non-o&&substituted PCBs and the chromatogram of the 12 monoand non-o&u-substituted PCBs. It is shown that 3,4,4’,5-tetra-, 3,3’,4,4’-tetra-, 2,3’,4,4’,5’-penta-, 3,3’,4,4’,5-penta-, 2,3’,4,4’,5,5’-hexa-, 2,3,3’,4,4’,5-hexa-, and 2,3,3’,4,4’,5’-hexachlorobiphenyl were not resolved from PCB components having two to four ortho chlorine substitution on the DB-5 column. Therefore, carbon separation is necessary for the quantitation of individual mono- and non-o&o-substituted PCBs.
A
FIG. 1. Cbromatograms on a DB-5 column of(A) a 1:l:l:l mixture ofAroclors 1221, 1016, 1254, and 1260 spiked with 12 mono- and non-ortho-substituted PCBs. (B) Twelve mono- and non-OrthcFsubstituted PCBs.
COPLANAR
PCBs
121
Figure 2 shows the chromatograms of a l&h from eastern Long Island Sound, fraction 1 contains two to four o&o-substituted PCBs, and fraction 2 contains mono- and non-or&-substituted PCBs. Levels of those mono- and non-orrho-substituted PCBs in fraction 2 were determined by comparing with appropriate gas chromatogram peaks of reference solutions. Table 1 presents the concentrations of individual mono- and non-o&o-substituted PCBs, total mono- and non-ortho-substituted PCBs, and total PCBs in the striped bass collected from the indicated area as shown in Fig. 3. Figure 3 shows a map of the New York Bight, indicating the sampling locations for this study (the striped bass in this study are not the same fish as those in 1990 paper, although they were collected at the same time), and the mean total concentrations of mono- and non-ortho-substituted PCBs for the striped bass sampled in each area. Table 2 shows the concentration of total mono- and non-o&o-substituted PCBs and total PCBs in striped bass from western and eastern Long Island Sound, the western and eastern Atlantic shore, and the Hudson River Estuary. Figure 4 shows each mono- and non-or&-substituted PCB found in the fish plotted in order of increasing distance from the Troy Dam, assuming free passage from Long Island Sound to the Hudson River. The Troy Dam is the upstream limit of the spawning migration. The figure shows a general trend in reduction of the amount of each mono-
FR-1
FA-2
FIG. 2. Chromatograms of the first and second fractions of a fish from Long Island Sound after fractionation on a carbon column.
PCB
coplanar PCBs hki9 Total PCBs bm % total coplanar PCBs
Total
3,4,4:5 3,3’,4,4’ 2,3’,4,4:5’ 2,3’,4,4’,5 2,3,4,4:5 2,3,3’,4,4’ 3,3’,4,4’,5 2,3:4,4’,5,5’ 2,3,3’,4,4’,5 2 >.3 3’ ,.4 4’ 35’ 3,3*,4,4:5,5 2 ,>3 3’ ,>4 4’ ,T 5 5’
congener
HUD(I0)
24.1
8.4
7.1
2.02
2.4 37.1 0.6 779 112 562 7.5 232 202 67.1 18.8
6.7
20.5
1.37
0.5 24.9 581 77.6 324 3.3 164 131 49.8 16.4
Albany/Troy
20.0
1.41
27.0 0.8 588 85.0 367 5.9 152 124 48.8 14.4
-@
IUPAC HUD(O) NO. Troy
81 77 123 118 114 105 126 167 156 157 169 189
1
-
19.9
5.8
7.6
1.16
8.9
21.9 0.6 499 57.7 339 4.1 108 90.4 31.0 -
21.1
1.61
4.1
790 81.8 359 4.7 188 146 5.0 -
4.8 29.1 -
Pt./Bay
HUD(168)
Croton
5.8
4.14
0.24
4.6
95.7 21.5 44.3 3.9 31.6 19.5 12.0 -
7.0 -
-
9.1
4.64
0.42
2.3
191 60.9 109 6.1 9.4 16.8 1.9 -
19.2 -
-
Bridge
11.3
2.01
0.23
1.2
6.4 1.2 85.2 24.9 45.3 0.9 34.4 20.4 7.0 -
-
Bridge
HUD(214) HUD(190) George Tappanzee Washington
8.1
6.52
0.53
1.8
300 25.3 113 49.6 27.8 12.5 -
4.3 -
-
NW(241) Eatons Neck
CONCENTRATIONS (ng/gwet W-QOFMONO-ANDNON-~&~-SUBSTITUTED FROMMARINEANDESTUARINEWATERSOFNEWYORKSTATE
TABLE
7.6
3.57
0.27
123 27.6 49.3 1.6 37.9 17.5 7.5 0.9
7.0 -
-
8.0
2.01
0.16
1.6
10.7 60.8 28.1 28.7 4.3 9.4 13.6 7.4 -
NW(248) Lilco
11.2
3.76
0.42
3.0
13.4 1.8 154 52.8 101 3.9 50.1 26.3 12.1 -
-
-
8.9
3.72
0.33
0.5
7.7 153 39.6 64.7 0.7 38.4 20.4 5.6 -
PCBs IN STRIPED BASS
10.0
4.02
0.40
7.2
10.7 1.0 137 46.3 113 1.9 43.1 28.1 11.5 -
-
SW(2368) Breezy Pt.
9.0
6.52
0.59
3.3
266 49.4 133 2.8 60.6 45.1 16.6 -
0.6 11.5 -
Note. HUD, Hudson in parentheses indicates ’ CO.1 rig/g.
PCBS
coplanar
4.9
River Estuary; distance (km)
4.8
4.7 1
3.54
( MJg)
o/total
0.23
0.17
(NM Total PCBs
4.3 I.1 73.4 30.1 56.6 1.1 26.3 23.2 9.9 -
8.8
4.22
6.4
3.12
0.20
93.7 21.1 41.6 0.8 24.0 10.5 5.2 0.4
0.2 4.2 9.6 -
-
7.7
5.33
0.41
1.8
200 41.2 58.6 1.6 45.8 33.6 14.4 -
Inlet
Fire Island
SE(328)
NW, Western Long Island Sound; from Troy Dam.
7.9
3.69
0.37
118 82.9 88.5 2.6 12.4 20.5 21.7 10.0
116 42.0 56.2 0.9 34.9 21.8 13.5 1.7
0.29
15.6 -
-
Beach
Jones
0.3 6.7 -
Inlet
3.3
-
Rockaway
NO.
81 77 123 118 114 105 126 167 156 157 169 189
SW(275)
IUPAC
6.8 0.5 66.0 19.9 36.5 0.8 19.4 13.2 4.6 0.7
3,4.4’,5 3,3:4,4 2,3’,4,4’,5’ 2$,4,4’S 2.3,4,4’,5 2.3,3’,4,4’ 3,3’,4,4’,5 2,3:4,4’,5,5’ 2,3,3’,4,4’,5 2,3,3’,4,4’,5’ 3,3:4,4:5,5 2,3,3’,4,4’,5,5’ Total coplanar PCBS
PCB congener
SE(302)
SW, Western
9.6
5.94
Atlantic
6.6
6.03
0.40
9.6
12.7 0.5 116 40.6 90.5 2.5 63.2 50.7 17.4 -
-
l-Continued
SE(342) Fire Island
0.57
258 70.6 118 1.5 58.0 29.4 15.0 7.7
1.5 13.8 -
TABLE
Atlantic
10.3
1.36
0.14
1.4
2.1 0.4 52.2 17.7 20.9 1.1 18.4 15.8 8.1 -
SE(362) Moriches Bay
Shore: SE, Eastern
10.2
2.35
0.24
1.5
1.1 97.3 26.1 49.8 1.6 28.9 19.6 7.8 -
-6.8
SE(348) Westhampton Beach
8.9
1.12
0.10
12.6 0.5 13.8 33. I 2.9 2.5
34.1 -
5.1 -
-
GUI1 Island
Little
NE(382) SE(389)
SE(395)
6.6
2.42
0.16
9.4 54.7 17.8 38.9 4.9 18.1 10.9 4.4 -
-
Amagansett
Long Island Sound.
10.8
1.30
0.14
5.0 14.0 2.0
4.1 1.0 55.8 20.0 35.8 -
-
Shinnecock
Shore: NE, Eastern
12.4
1.61
0.20
0.8
5.3 31.8 65.5 20.5 37.2 1.4 22.9 14.6 4.8 -
-
NE(375) Orient Pt.
SE(422)
Number
6.5
3.86
0.25
0.5 4.5 0.7 137 19.8 38.8 0.9 27.3 10.9 6.2 -
Montauk
E
0
124
HONG, BUSH, AND XL40
FIG. 3. Sampling points in theNewYork BightandHudson River showing concentrations of totalmonoand non-o&o-substituted PCBs at each point &@g wet weight).
and non-o&u-substituted PCBs, 3,3’,4,4’-tetrachlorobiphenyl was found to be higher than 3,3’,4,4’,5-pentachlorobiphenyl in all the cases. The highest concentrations of about 37.1 rig/g 3,3’,4,4’-tetrachlorobiphenyl and 7.5 rig/g of 3,3’,4,4’,5pentachlorobiphenyl were detected in a fish caught near Troy/Albany, New York. 3,3’,4,4’,5,5’Hexachlorobiphenyl was not detected in any samples. Mono-o&o-substituted PCBs ranged from 0.4 to 790 ppb. The main components in all the species were 2,3’,4,4’,5and 2,3,3’,4,4’-pentachlorobiphenyl, which reached 30-56% and 12-27s oftotal monoand non-o&o-substituted PCBs, respectively. TABLE 2 CONCENTRATION (fig/g wet wt) OF TOTAL MONO- AND NON-O~~~&~JB~TITI.JTED PCBs AND TOTAL PCBs IN STRIPED BASS FROM FIVE AREAS OF NEW YORK BIGHT
Concentration (mean f standarderror of the mean) Area’
Number of fish
Total coplanars
Total PCBs
NW NE SW
4 2 6
SE
10
0.35 -+ 0.08 0.15 & 0.05 0.34 +-0.06 0.29 t- 0.05 1.06 +- 0.24
3.97 +- 0.94 1.37f 0.24 4.37 + 0.46 3.59 +- 0.56 14.55+- 3.25
HUD
8
BArea: NW, Western Long Island Sound; NE, Eastern Long Island Sound; SW, Western Atlantic Shore; SE, Eastern Atlantic Shore; HUD, Hudson River Estuary.
COPLANAR
PCBs
125
800 600 400 200 0
FIG. 4. Concentration of mono- and non-orfho-substituted PCBs in fish plotted in order of increasing distancefromtheTroyDam.(l)3,4,4’,5;(2)3,3’,4,4’;(3)2,3’,4,4’,5’;(4)2,3’,4,4’,5;(5)2.3,4,4’,5;(6)2,3,3’,4.4’; (7) 3,3’,4,4’,5: (8) 2,3’.4,4’,5,5’; (9) 2,3,3’,4,4’,5; (10) 2,3,3’,4,4’,5’; (11) 3,3’,4,4’,5,5’: (12) 2.3,3’,4.4’.5,5’.
Table 3 presents the concentrations of individual mono- and non-ortho-substituted PCBs, total mono- and non-ortho-substituted PCBs, and total PCBs in six mussels from Long Island Sound and one composite of freshwater mussels from Troy. Figure 5 shows the pattern of mono- and non-o&o-substituted PCBs present in mussels from Troy and Long Island Sound. The sample from Troy, nearest the pollution source at Fort Edward, New York, has the higher level of total mono- and nono&o-substituted PCBs, 112.9 ppb. Mussels from Long Island Sound contain only 8.5-18 ppb of total mono- and non-o&o-substituted PCBs. Figure 6 shows the total PCBs and total mono- and non-ortho-substituted PCBs in fish plotted in order of increasing distance from the Troy Dam. Clearly the total PCB burden of the samples is strongly correlated with the total mono- and non-orthosubstituted PCB burden. Figure 7 shows the correlation of total coplanar PCBs with total PCBs in striped bass. The correlation coefficient (R) is 0.98 and the regression coefficient is 13 + 0.5
COPLANAR
PCB
FIG. 5. Concentration of mono- and non-otiho-substituted PCBs in one composite of freshwater mussels from Troy and six mussels from Long Island Sound. Mono- and non-ortho-substituted PCBs are the same as those in Fig. 4.
3
11 Duplicateofsamesample. b
