COMPARATIVE RICE SEED TOXICITY TESTS USING FILTER PAPER, GROWTH POUCH-TM, AND SEED TRAY METHODS WUNCHENG WANG Illinois State Water Survey, P.O. Box 697, Peoria, IL 61652, U.S.A.* (Received: July 1991)

Abstract. Paper substrate, especially circular filter paper placed inside a Petri dish, has long been used for the plant seed toxicity test (PSTT). Although this method is simple and inexpensive, recent evidence indicates that it gives results that are significantly different from those obtained using a method that does not involve paper, especially when testing metal cations. The study compared PSTT using three methods: filter paper, Growth Pouch-TM, and seed tray. The Growth Pouch-TM is a commercially available device. The seed tray is a newly designed plastic receptacle placed inside a Petri dish. The results of the Growth Pouch-TM method showed no toxic effects on rice for Ag up to 40 mg L -1 and Cd up to 20 mg L -1. Using the seed tray method, ICs0 (50% inhibitory effect concentration) values were 0.55 and 1.4 mg L -1 for Ag and Cd, respectively. Although results of filter paper and seed tray methods were nearly identical for NaF, Cr(VI), and phenol, the toxicities of cations Ag and Cd were reduced by using the filter paper method; IC5o values were 22 and 18 mg L -1, respectively. The results clearly indicate that paper substrate is not advisable for PSTT.

1. Introduction The purpose of this study was to compare plant seed toxicity testing (PSTT) using three methods: filter paper, Growth Pouch-TM, and the seed tray. Rice (Oryza sativa) was used as the test species. PSTT can be performed using soil, an inert substance, paper substrate, or no solid substrate (Adema and Henzen 1989, Gorsuch et al. 1990, Ratsch and Johndro 1986, Walsh and Weber 1991, Walsh et al. 1991, Wang 1987, 1990, Wang and Williams 1988, 1990, Wong and Bradshaw 1982). The varieties of test methods are perhaps a reflection of vastly different plant species residing in various environments, such as terrestrial, aquatic, marsh, and estuarine regions. A detailed discussion of plant test methodology and results is given in a recent literature review (Wang 1991). When different test methods are used, test results can vary significantly. For example, Adema and Henzen (1989) reported that Cd 50% inhibitory effect concentrations (IC50s) for lettuce, oats, and tomato were 0.84 to 6.0 mg L -1 tested in nutrient solution. While the ICs0 values ranged from 114-561 and 53-430 mg L -1 in loam and humic sand, respectively. Similarly, Ratsch and Johndro (1986) conducted lettuce root elongation tests using solution and inclined filter paper methods. They reported that Cd (expressed in CdC12) IC50 values were 0.23 and 6.5 mg L -1 , respectively. These results suggest that P S T r requires more investigation, especially on test methods. Present address: U.S. Geological Survey P.O. Box 1230, Iowa City, IA 52244. Environmental Monitoring and Assessment 24: 257-265, 1993. @ 1993 Kluwer Academic Publishers. Printed in the Netherlands.

258

WUNCHENGWANG

Recently, the author designed a device, the seed tray, for use specifically for PSTT. The device, a plastic receptacle, is placed inside a Petri dish (see the Methods section). In contrast to the filter paper method (Swanson 1946, Wang 1987, 1990), the test substance in the seed tray method remains in aqueous solution. Unlike the solution method reported by Ratsch and Johndro (1986), where plant seeds tumble continuously inside a test vessel, in this method the plant root extends vertically. This method also differs from the solution method reported by Walsh and Weber (1991), who used a Petri dish containing aqueous solution, so that root systems could develop only horizontally. The Growth Pouch-TM (Northrup King, Downers Grove, Illinois) is a commercially available product used for testing plant seed responses (Gorsuch et al. 1990, Wang and Williams 1990). The product offers a convenient, safe approach for testing hazardous samples in either aqueous or slurry form. 2. Methods The six chemical compounds used for this study were silver nitrate (AgNO3), cadmium chloride (CdCI2), potassium chromate (K2CrO4), sodium fluoride (NaF), phenol (C6HsOH), and 2,4-D (2,4-dichlorophenoxyacetic acid). The first five compounds were reagent-grade and 2,4-D was assayed as 96%. Liquefied phenol (88%) was used and the aqueous solution was prepared by factoring this value. These compounds were selected so that the results could be compared with those reported by Ratsch and Johndro (1986). Nominal concentrations of all compounds were used. The stock solution of 2,4-D was prepared using 100 mg per 100 mL aqueous solution containing 15 mL triethylene glycol. The highest working solution was 2 mg L -1 2,4-D, so that the concentration of the carrier solvent was 0.03% (v/v), a value below the recommended limit of 0.05% (American Public Health Association et al. 1989). All other toxicant solutions were prepared directly in an aqueous solution, without the stock solution step. Hard, reconstituted water (American Public Health Association et al. 1989) contains a sufficiently high concentration of chloride ions to cause Ag ion precipitation. For this reason deionized water (treated with water softener and reversed osmosis) was used throughout this study as the water control and dilution water. The highest concentration of each compound was selected on the basis of results by Ratsch and Johndro (1986) and Wang (1987). There were six to seven diluted solutions, using 0.5 dilution factor. About 10 kg of rice seeds were purchased in January 1990 and stored at - 10°C until use less than 12 months later. The seeds were pretreated with 3.3 g of 10% Clorox-TM (CIO-L -1) for 20 min and washed ten times with deionized water. Different vessels were used for the three methods: a 100 x 15-millimeter glass Petri dish (plus a 90-mm Whatman #1 filter), Growth Pouch-TM, and a 100 × 15-mm plastic Petri dish (plus a seed tray). Different volumes of test samples were used: 9, 18, and 31 mL, respectively.

COMPARATIVE RICE SEED TOXICITY TESTS

259

TABLE I Summary of rice seed toxicity test. Test type Seed pretreatment Temperature Light quality Test vessels

Test solution/vessel

Seeds/vessel Diluted solutions Dilution factor Water control and dilution water Replicates Test period Test indicator

Static 10% Clorox solution for 20 min (3.3 g OC1-L - l ) 25 to 25.2°C Dark 100 x 15-ram glass Petri dish (plus a 90-mm Whatman #1 filter) Growth Pouch-TM 100 x 15-mm plastic Petri dish (plus a seed tray) 9 mL filter paper 18 mL Growth Pouch-TM 31 mL seed tray 12 6to7 0.5 Deionized water 4 120 hours Dry-root biomass

Twelve seeds were placed in each vessel and incubated at 25-25.2°C for six days. At the end of incubation, all primary roots in each vessel were severed, dried at room temperature for 24 h, combined, and weighed to the nearest 0.1 rag. The root biomass method offers a simple and less time consuming alternative to the root elongation measurement (Wang and Williams 1990). Each test contained a water control and all tests were performed in quadruplicate. Table I summarizes the test conditions. The seed tray, a custom-made plastic material, was designed to fit into a 100 × 15-ram Petri dish (Figure 1). With 31 mL test solution, the seed tray was covered with liquid, and test seeds were placed on the tray and were slightly submerged. New trays were hydrophobic and contained impurities (e.g., plasticizers), but soaking the new trays in a detergent bath ovemight eliminated this problem. During incubation, plant roots extended through the holes and were immersed in the bulk liquid below. To obtain root biomass as the test endpoint, the extended roots at the bottom of the trays were pulled and stretched gently using a forceps. A sharpened 5.1centimeter putty knife was used to cut off the roots. One cut was usually sufficient to sever all roots. If the roots entangled and developed horizontally at the top of the trays, individual handling was required. Using the filter paper method, all primary roots required individual cutting (Wang 1987). Using the Growth Pouch-

260

WUNCHENGWANG

!

i

O00000~O00000OO 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O00000OO0000000 OOOO0000000OO00 OOOO00000000000 O000UO0000OO000 O000000OOOOO000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ODOOOOOUO000000 00000000~000000 OOO000OO0000000 O0000UOOOOOUO00 O0000000000OO00 O 0 0 0 0 0 U 0 0 0 0 0 0 0 0 0000t_lf_lO

01_1EIO

1

TOP VIEW

(3",

3

rJo013 I

'

Illll:_-cr3:r_Irti:ct]-ri12tt3:rll:tt33:]~:tt.-CC:lllo 111 I /1 SIDE VIEW il . / . . . . . ~ tile ..1l-I

Fig. 1. Seedtray for plant toxicity test. TM methods, roots required one cut along the seed trough and severed roots were removed using forceps (Wang and Williams 1990). The ICs0 values and 95% confidence limits were calculated using the moving average method (Peltier and Weber 1985). The t test was used to calculate significant differences between mean values. :,

3. R e s u l t s

3.1. COMPARISONOF TEST QUALITY The quality of test results can be compared using control samples. Specifically for PSTT, results of the seed germination rate of the control samples are given in Table II. The mean rates for the filter paper, Growth Pouch-TM, and seed tray methods were 83, 82, and 87% respectively. There was no significant difference among these rates. On one occasion, the filter paper and Growth Pouch-TM methods each produced a relatively low germination rate: 73 and 63%, respectively. Only the seed tray method showed a consistently high germination rate of 79% or better. Amond the three methods, Growth Pouch-TM yielded the highest mean root biomass, 9.2 mg per dish in the control samples (Table II). The respective mean root biomass values for the filter paper and seed tray methods were 6.3 and 3.9 mg per dish, respectively. The relatively high biomass values obtained from the Growth Pouch-TM and filter paper methods, however, had no significant advantage

COMPARATIVERICE SEED TOXICITYTESTS

261

TABLE II Control samples of repeated rice toxicity tests. Tests Germination, % Filterpaper Growth Pouch-TM Seed tray Biomass, mg per dish Filter paper GrowthPouch-TM Seed tray

1 83 90 85

2 94 85 83

3 83 88 90

4 85 81 79

5 73 85 92

6.2 7.7 6.2 6.3 6.8 12.4 10.5 10.5 8.5 9.3 4.0 3.3 3.9 2.9 4.8

Coeffi~entofvafiation,% Filter paper 7 Growth Pouch-TM 13 Seedtray 23

16 18 20

12 8 8

17 28 17

10 14 16

6 79 83 88

Mean S.D. 7 79 90 94

11 90 77 88

83 82 87

6 6 5

5.9 5.5 8.9 9.3 3.8 5.4

5.6 6.2 6.5 6.6 6.5 8.2 7.5 9.7 3.2 4.3 3.5 3.9

6.3 9.2 3.9

0.6 1.6 0.7

18 12 32

34 38 18

15 16 15

7 10 8

12 9 6

8 83 63 81

9 85 79 92

10 10 7

10 79 79 81

21 22 9

11 7 10

o v e r the seed tray method. For example, precision o f tests as indicated by the m e a n coefficients o f variation (standard deviation/mean) were practically identical, 15, 16, and 15% for the three tests, respectively. Furthermore, the filter p a p e r and G r o w t h P o u c h - T M methods yielded erroneous results, as s h o w n in the n e x t section. 3.2. COMPARISON OF TOXICITY TEST RESULTS Toxicity test results (Figure 2) clearly showed that the toxic effects were dependent on test methods and test c o m p o u n d s . T h e G r o w t h P o u c h - T M m e t h o d produced a n o n r e s p o n s e effect to cations Ag and Cd, a highly subdued effect to anion Cr(VI) and salt NaF, and a slightly decreased effect to organic c o m p o u n d s phenol and 2,4-D. T h e filter p a p e r method also showed decreased toxic effect to cations Ag and Cd, w h e n c o m p a r e d with the seed tray method (Figure 2). T h e IC5o values were 22 and 18 mg L -1, respectively. W h e n herbicide 2,4-D at concentrations o f 0.125 m g L -1 or greater was present, the roots attached to the p a p e r firmly, m a k i n g the root biomass m e a s u r e m e n t impossible. F o r this reason, the results o f 2,4-D were not presented in Figure 2. Table III gives the IC50 values and the 95% confidence limits determined b y using the three methods. A m o n g six toxicants, 2,4-D was the m o s t toxic; IC5o values were 0.10 and 0.11 m g L -1 using the G r o w t h P o u c h - T M and the seed tray methods, respectively. T h e seed tray method was m o s t sensitive for other toxicants. T h e IC50 values for Ag, Cd, Cr(VI), phenol, and N a F were 0.55, 1.4, 4.8, 75, and 320 m g L -1, respectively. T h e IC50 values for NaF, Cr(VI), and phenol obtained

WUNCH W ENA GNG

262 100

Cd

b,-.Cr --

I

0

I

I

B ~

I

20

9

,

.n

10

40

20

100

t-

=

fJ

NaF

"

iX.

l

I

I

I 100

50

I 0

I 2000

1000

100

_,X

A~A/ I

0

I

200

2,4-D

Phenol I

I

400

I

0

CONCENTRATION.

!

I

I

I

2

mg/L

Fig. 2. Rice seed toxicity test using three test methods, filter paper (O), Growth Pouch-TM (C]), and seed tray (A), using mean values of two experiments.

by the filter paper method approximated those obtained by the seed tray method, 331, 7.3, and 79 mg L - I , respectively.

4. Discussion Paper substrate has been used in PSTT because it is simple, convenient and inexpensive. Attempts are being made to standardize the method by the American Society for Testing and Materials, Philadelphia, Pennsylvania, and the Standard Methods Committee, which comprises representatives from the American Public

263

COMPARATIVE RICE SEED TOXICITY TESTS

TABLE III IC5o (50% inhibitory effect concentration) values and 95% confidence limits of six toxicants, all mg L -1 . Filter paper ICs0

Growth Pouch-TM

95% CL

ICs0

95% CL

Seed tray ICs0

95% CL

Ag

17

16-20

NA a

0.55

0.50~0.61

Cd

18

16-21

NA"

1.4

1.27-1.53

331

298-368

827

747-920

320

284-359

Cr(VI)

7.3

6.6-7.9

63

53-77

4.8

3.92-5.65

Phenol 2,4-D

79 NA b

72-88

118-166 0.08-0.12

75 0.11

68-84 0.096-0.12

NaF

138 0.10

NA a = Not available because 50% inhibitory effect was not reached NA b = Not available because roots attached firmly on filter paper

Health Association, American WaterWorks Association, and Water Pollution Control Federation. The results of this and related studies, however, suggest that the use of paper substrate is not advisable. There are three major problems using any paper substrate. First, the paper substrate interfered with toxicant bioavailability, especially cations such as Ag and Cd. A companion study confirmed that chemisorption was the cause of cation toxicity reduction (Wang, forthcoming). Second, the paper substrate used in the test tended to stimulate the root development. For example, the mean root biomasses in the control samples were 6.3, 9.2, and 3.9 mg per dish for the filter paper, Growth Pouch-TM, and seed tray methods, respectively. The paper substrate, in proportion with the mean root biomass from these tests, weighed more in the Growth Pouch-TM device than the filter paper, 3.06 grams (g) versus 0.54 g, respectively. The companion study (Wang, forthcoming) also reported a significant difference between two test methods with and without filter paper, using Japanese millet seed as the test species. Four repeated experiments gave the mean (and standard deviation) of the tests with and without filter paper as 3.7 (0.24) and 2.1 (0.18) mg per dish root biomass, respectively. Two hypotheses of root stimulation have been proposed. One, unknown stimulant(s) is (are) present in the paper. Two, gas diffusion is facilitated through the paper substrate. Either way, the stimulation introduced by the paper substrate can only complicate test results. Third, the use of filter paper presented difficulty because the roots sometimes attached strongly to the paper, making the root biomass measurement impossible. All these problems are sufficiently serious and using paper substrate for PSTT is not advisable, whether testing a single compound or a complex, unknown mixture. The seed tray method is convenient for PSTI'. The device is small, simple, inexpensive, strong, reusable, and does not require large amounts of laboratory

264

WUNCHENG WANG

space. The device does not interfere with test compounds, and offers a quick way to obtain test results. The device is especially attractive for toxicity testing using the renewal method, which is a prerequisite for municipal effluent monitoring; the device containing test specimens can easily be transferred to a vessel containing fresh sample. The method can be used in a light-dark cycle. The seed tray method, however, should be conducted with care. Almost all roots in the control sample extended into the reservoir containing bulk liquid. Roots could be cut easily. With the presence of a toxicant, however, roots occasionally stayed on the top side of the device and required individual handling. An alternative test method for PSTT which does not use a paper substrate was reported by Walsh and Weber (1991), who measured the dry biomass of whole seedlings as the test endpoint. This approach may not be ideal because both shoots and seeds are included, which results in decreased sensitivity of the toxic effect. Improved sensitivity can be achieved by severance and measurement of root portions only. Another approach uses fine nylon mesh suspended in a beaker (Wong and Bradshaw 1982). The approach is similar to the seed tray method and also provides a convenient way for measuring root length. The results of this study (seed tray method) and those of Adema and Henzen (1989) offer interesting comparisons. They used wet shoot biomass of lettuce, tomato, and oats, and tested in nutrient solution for 14 days. They reported IC50 values for Cd of 0.84, 3.0, and 6.0 mg L -1 , respectively, while those for Cr were 0.16, 0.29, and 1.4 mg L -1, respectively. For comparison, the IC50 values for Cd and Cr using dry-root biomass of rice for six days were 1.4 and 4.7 mg L -1, respectively, using the seed tray method. Although the seed tray method gave considerably better results than the filter paper and the Growth Pouch-TM methods, the results were different than those obtained by Ratsch and Johndro (1986). For example, the IC50 values of Ag, Cd, NaF, and 2,4-D in this study were 0.55, 1.4, 320, and 0.11 mg L -1, respectively. The respective values of AgNO3, CdC12, NaF, and 2,4-D reported by Ratsch and Johndro were 0.011, 0.23, 660, and 0.005 mg L -1. The differences are possibly due to a combination of factors: plant species (rice vs. lettuce), light (dark vs. 16-8 hours light-dark), air agitation (no vs. yes), nutrient availability (deionized water vs. 0.1 strength NCSU phytotron nutrient solution), and other unknown factors. The results of this study and that by Ratsch and Johndro (1986) appeared that rice was seemingly less sensitive than lettuce. Rice, however, had the following advantages over lettuce. First, because rice is an aquatic and/or wetland species (Correll and Correll 1972), its response is considered relevant to aquatic environments. Second, rice is the most important crop in the world (Correll and Correll 1972). Third, rice developed only short shoots (1-2 cm) in the first 120 h, making seedling tangling less likely and handling easier. Fourth, rice roots were stout when exposed to toxicants, unlike lettuce roots, which rot when exposed to toxicity and thus preclude handling.

COMPARATIVERICE SEED TOXICITY TESTS

265

PSTT using rice is especially useful for complex effluent biomonitoring. Wang (1990) compared rice and lettuce for 23 pretreated industrial effluents and reported that rice was invariably more sensitive than lettuce. Furthermore, Wang and Keturi (1990) found rice's high germination rate, long shelf-life, and sensitivity to toxicity made it an ideal candidate among ten species for toxicity testing of a metalengraving effluent sample. Acknowledgements The die mold for the seed tray was made with financial support from Richard G. Semonin, Chief of the Illinois State Water Survey. References Adema, D.M.M. and Henzen, L.: 1989, 'A Comparison of Plant Toxicities of Some Industrial Chemicals in Soil Culture and Soilless Cultures', Ecotoxicol. Environ. Saf. 18, 219-229. American Public Health Association, American Water Works Association, and Water Pollution Control Federation: 1989, Standard Methods for Examination of Water and Wastewater, 17th ed., Washington, De. Correll, D.S. and Correll, H.B.: 1972, Aquatic and Wetland Plants of Southwestern United States, U.S. Environmental Protection Agency, 16030 DNL, Washington, DC. Gorsuch, J.W., K_ringle,R.O. and Robillard, K.A.: 1990, 'Chemical Effects on Germination and Early Growth of Terrestrial Plants', ASTM STP 1091, Amer. Soc. Testing & Materials, Philadelphia, PA, pp. 49-58. Peltier, W.H. and Weber, C.I. (Eds.): 1985, Methods for Measuring the Acute Toxicity of Effluents to Freshwater and Marine Organisms, 3rd ed., EPA/600/4-85/013, U.S. Environmental Protection Agency, Cincinnati, OH. Ratsch, H.C. and 3ohndro, D.: 1986, 'Comparative Toxicity of Six Test Chemicals to Lettuce Using Two Root Elongation Test Methods', Environ. Monit. Assess. 6, 267-276. Swanson, C.P.: 1946, 'A Simple Bioassay Method for the Determination of Low Concentrations of 2,4-Dichlorophenoxyacetic Acid in Aqueous Solution', Bot. Gaz. 108, 39-44. Walsh, G.E., Weber, D.E., Simon, T.L., Brashers, L.K. and Moore, J.C.: 1991, 'Use of Marsh Plants for Toxicity Testing of Water and Sediment', ASTM STP 1115, Amer. Soc. Testing & Materials, Philadelphia, PA, pp. 341-354. Walsh, G.E., Weber, D.E., Simon, T.L. and Brashers, L.K.: 1991, 'Toxicity Tests of Effluents with Marsh Plants in Water and Sediment', Environ. Toxicol. Chem. 10, 517-525. Wang, W.: 1987, 'Root Elongation Method for Toxicity Testing of Organic and Inorganic Pollutatns', Environ. Toxicol. Chem. 6, 409-414. Wang, W.: 1990, 'Toxicity Assessment of Pretreated Industrial Effluents Using Higher Plants', Res. J. Water Poll. Control Fed. 62, 853-860. Wang, W.: 1991, 'Literature Review on Higher Plants for Toxicity Testing', Water, Air, Soil Poll. 59, 381-400. Wang, W.: 1992, 'Use of Plants for the Assessment of Environmental Contaminants', Reviews Environ. Contain. Toxicol., in press. Wang, W.: forthcoming, 'Plant Seed Toxicity Test: Effects of Filter Paper on Cadmium Toxicity'. Wang, W. and Keturi, P.H.: 1990, 'Comparative Seed Germination Tests Using Ten Plant Species for Toxicity Assessment of Metal Engraving Effluent Sample', Water, Air, Soil Poll. 52, 369-376. Wang, W. and Williams, J.M.: 1988, 'Screening and Biomonitoring of Industrial Effluents Using Phytotoxicity Tests', Environ. Toxieol. Chem. 7, 645~552. Wang, W. and Williams, J.M.: 1990, 'The Use of Phytotoxicity Tests (Common Duckweed, Cabbage, and Millet) for Determining Effluent Toxicity', Environ. Monit. Assess. 14, 45-58. Wong, M.H. and Bradshaw, A.D.: 1982, 'A Comparison of the Toxicity of Heavy Metals, Using Root Elongation of Rye Grass, Lolium Perenne', New Phytol. 91,255-261.