Use of Lilium longiflorum, cv. ace pollen germination and tube elongation as a bioassay for the hepatocarcinogens, aflatoxins.

Although various animal tissues are used for bioassay of aflatoxins (B1, B2, G1, G2), a rapid bioassay dependent upon a plant part's response does not exist. Both pollen germination (G) and tube elongation (TE) were enhanced in a 3.0 mM KH2PO4 (K)-containing but AFB1-lacking, modified Dickinson's medium. The B1 did not affect G when K was withheld but K supplementation impaired G above 15 micrograms/ml B1. Without K, 5-20 stimulated but 25 and 30 micrograms/ml B1 inhibited TE which was suppressed by every B1 conc tested in K-containing medium. Addition of NaH2PO4(N) instead of K to medium did not promote G. Slight G stimulation occurred at 16.6 micrograms/ml mixed aflatoxins (MA) in medium lacking either K or N but low G inhibitions were observed with K or N. The MA at 33.3 micrograms/ml reduced G 2.5% in K's of N's absence and 26 or 17% in their presence. While K did not stimulate TE without MA, N did 26%. At 16.6 and 33.3 micrograms/ml MA, TE was reduced 19, 6, 19% and 24, 25, 31%, respectively, in control, K- and N- media. Pollen G and TE were markedly sensitive to G1. Significant inhibitions of Zea mays seed G were observed at 5.8 and 11.6 micrograms/ml B1 but not root elongation (RE) from 0.4-11.6 micrograms/ml. The MA (31.5 micrograms/ml) administered for 72-240 hr did not influence either Arachis hypogeae seed G or RE. However, imbibing 5 cultivars each of Avena sativa (65-117 hr) and Hordeum vulgare (39-89 hr) inhibited RE 4/15-62%. Thus, except for Z. mays, pollen G and TE appear to be more B1-sensitive than seed G and RE. But, the pollen bioassay is less sensitive than both certain animal bioassays (0.025 micrograms/ml) and analytical methodologies (10 pg.).


Introduction
Although a number of bioassays for aflatoxin B1 (AFB1) and certain other aflatoxins exist which employ a variety of animals (1), a bioassay which utilizes a plant or alternatively one or more of its parts has not been developed. However, there are *Department of Biology, West Virginia University, Morgantown, West Virginia 26506. tDepartment of Biology, Virginia Commonwealth University, Richmond, Virginia 23229.
This report is part of a series of papers on Pollen previously published in Environ. Health Perspect. 33:   (1980).  published reports, which have been reviewed by Dashek and Llewellyn (2), that describe the effects of aflatoxins on seed germination, seedling growth or elongation, chlorophyll synthesis, enzymatic activities, amino acid uptake, protein or nucleic acid syntheses and cellular ultrastructure of or by a variety of plant parts. A number of these effects could possibly be of service in the development of a bioassay.
Because our initial investigations (3) suggested that both Lilium longiflorum, cv. Ace pollen germination and tube elongation responded to AFB1 at concentrations as low as 4 ,ug/ml (approximately 30% inhibition of germination) and 8 jig/ml (about 25% inhibition of tube elongation), we have attempted to develop a bioassay utilizing these two parameters of pollen growth and development. This paper reports the results of that attempt, an examination of aflatoxin effects on seed gernination and elongation of attached roots of certain crop plants, a comparison of the sensitivities of various plant and animal tissues to aflatoxins and a comparison of the sensitivities of the pollen bioassay and analytical methods for quantitating aflatoxins.

Pollen Germination Conditions
Experiments with AFBI and KH2PO4. Two stock solutions of Dickinson's medium (4) without tetracycline were adjusted to pH 5.2; the control and experimental stocks lacked and contained 3.0 mM KH2PO4, respectively. Aflatoxin B1 (A grade, Calbiochem., LaJolla, Calif.) was dissolved in 7 ml of acetone which was then added to 500 ml of experimental medium to yield 30 ,ug/ml toxin. This concentration was verified by combined thin layer chromatography and a visual dilution technique which is sensitive to 2 ppb (5). Acetone (7 ml) was supplied to the control medium prior to autoclaving. The experimental medium was diluted to yield media containing 5, 10, 15, 20 and 25 ,ug/ml toxin.
Lilium longiflorum, cv. Ace pollen (6 months, 4°C) was added in lots of 10 mg fresh weight to 20 ml aliquots of sterile medium in sterile, disposable Petri dishes which were then incubated at 26 + 2°C for 4 hr. To measure tube lengths and assess percent germination, photomicrographs ofthe dishes were taken subsequent to positioning the dishes over a partitioned grid. The germination percentages and tube lengths presented in Table 1  (control) and the other (experimental) contained it; AFB1 (grade B, dried in situ, Calbiochem, LaJolla, Ca.) was dissolved in 7 ml acetone and then added to 500 ml of the experimental medium with a resulting AFB1 concentration of 30 ,g/ml; subsequent to autoclaving, AFB1 concentration was verified by thin layer chromatography coupled with a visual dilution technique which is sensitive to 2 ppb; 7 ml of acetone were also added to the control medium prior to autoclaving; the 30 ,ug/ml AFB1 experimental medium was diluted to yield media which contained 5, 10, 15, 20 or 25 ,ug/ml; 10 mg fresh weight lots of stored (6 months, 4°C) Lilium longifljom, cv. Ace pollen were sown in 10 ml aliquots of sterile medium in sterile, disposable, plastic Petri dishes; pollen was incubated for 4 hr at 27 ± 2°C. To obtain tube lengths and percent germinations, photomicrographs were made of the Petri dishes following their positioning over a 3 mm grid; data tabulated from the photomicrographs are means and standard deviations for three experiments; each experiment had two controls; 350 pollen grains were scored and tubes measured per treatment; % change equals the % change from the control; a variant of this table and figure legend will appear (10). 268 Environmental Health Perspectives treatment varied, that number approximated 3000. The total number of tubes measured was 260.
Experiments with AFB2, AFG1 andAFG2. The specificity of the pollen bioassay was examined by dissolving aflatoxin B2 (AFB2) aflatoxin G, (AFG1) and aflatoxin G2 (AFG2) in acetone with subsequent transfer to Petri dishes of aliquots which yielded 15 and 30 ,ug/ml upon addition of medium which either contained or lacked 3.0mM KH2PO4.4

Seed Germination Conditions
Fifty Arachis hypogaea, cv. Florigiant seeds were treated with Botec fungicide, rinsed with sterile H20 and then sown in lots of five per Petri dish on three layers of sterile standard laboratory paper toweling. The seeds were arranged such that their embryo ends pointed toward the center of the dish. To each dish 15 ml of sterile H20 containing 31.5 ,ug/ml mixed aflatoxins (5.6 ,g/ml AFB1, 0.2 ,ug/ml AFB2, 25.0 ,g/ml AFG1 and 0.8 jig/ml AFG2) were added. The dishes were sealed with parafilm and tilted to an angle which permitted the roots to elongate in a straight fashion. The seeds were incubated in the dark at 24 + 20C for 72, 144, 168 and 240 hr when percent germination and root lengths were measured, the latter with a mm rule.
The same experimental design was also used for both Avena sativa and Hordeum vulgare seeds except that five cultivars of each genera were tested. These cultivars were: Coker, Norline, Moregrain, Windsor and Roanoke for A. sativa and Surry, Henry, Volbar, McNair and Barsoy for H. vulgare. Other differences in the design included 10 seeds and 10 ml of test solution per dish. The germination times were 65, 89 and 177 hr for A. sativa and 29, 63 and 89 hr for H. vulgare.
Zea mays seeds were surfaced sterilized with 10% Chlorox for 10 min, at which time they were rinsed with 300 ml distilled H20. Seeds in lots of 15 were sown on two layers of sterile ifiter paper and imbibed as above except that the imbibition medium included 50 pg/ml chloramphenicol and the dishes were not tilted. The experiment was replicated three times with duplicate dishes for each treatment within an experiment.

Statistical Analyses
The data were analyzed by a two-tailed t-test for evaluating the difference between population means.

Preparation of Mixed Aflatoxins for Seed Germination Studies
Mixed aflatoxins were prepared according to The inoculated coconuts were extracted with cloroform and the extract analyzed for aflatoxins by the combined thin layer chromatography and visual dilution technique. The choloroform extract contained 1270 ,ug mixed aflatoxins (225 jig/ml AFB1, 6 jig/ml AFB2, 1000 ,ug/ml AFG1 and 39 ,g/ml AFG2). A 50 ml portion of this crude extract was transferred to 1 liter of warm sterile distilled water which was gently heated to drive off the chloroform. This solution contained 31.5 ,ug/ml mixed aflatoxins in the ratio of individual aflatoxins reported above and was stored in the dark at 8°C.

Results and Discussion
Does AFB1 Inhibit Pollen Germination and Subsequent Tube Elongation When KH2PO4 Is Withheld or Included in the Growth Medium?
Germination was not inhibited when pollen was sown in medium lacking phosphate but containing 5-30 pg/ml AFB1 (Table 1). In contrast, pollen germination was inhibited 10.6, 6.3, 27.3 and 45.1% upon addition of 15, 20, 25 and 30 pg/ml AFB1 to an incubation medium containing 3.0 mM KH2PO4. In the absence of KH2PO4, only 25 and 30 ,ug/ml AFB1 inhibited tube elongation. However, this elongation was inhibited at every AFB1 concentration when KH2PO4 was added to the germination medium. Maximum inhibition occurred at 25 (24%) and 30 (36%) pg/ml.

Do Mixed Aflatoxins Inhibit Pollen
Germination and Tube Elongation When Different Phosphate Salts are Withheld or Added to the Growth Medium?
Both percent germination and tube lengths for pollen sown in medium containing or lacking mixed aflatoxins and either KH2PO4 or NaH2PO4 are shown in Table 2. Without phosphate, 16.6 pg/ml and 33.3 pug/ml mixed aflatoxins stimulated and reduced percent germination by 7.8 and 34.7%, respectively. When 3.0 mM NaH2PO4 was included in the medium, the percent germination inhibitions were 17% (16.6 ,ug/ml) and 16.5% (33.3 pug/ml). In contrast, addition of3.0 mM KH2PO4 to the incubation medium yielded germination inhibition percentages of8.3% and 25.6% at 16.6 and 33.3 pg/ml, respectively. Tube elongations for pollen germinated in the absence of phosphate but presence of 16.6 or 33.3 ,g/ml mixed aflatoxins were reduced 18.9% and 23.7%, respectively. When the medium was provided A comparison of the effects of AFB2, AFG1 and AFG2 on both germination and tube elongation is shown in Table 3. Whereas sowing pollen in medium lacking 3.0mM KH2PO4 but containing 15 g±g/ml AFB2 did not significantly inhibit percent germination, 30 ,ug/ml suppressed germination by 29% and tube elongation by 15% (Table 3). Both ofthese reductions were statistically significant at 95% confidence level. In contrast, both 15 and 30 ,ug/ml AFG1 in medium lacking KH2PO4 suppressed germination by 97.5 and 99.7%, respectively, and tube elongation by 100%. When KH2PO4 was added to the medium, 15 ,ug/ml AFG1 decreased germination and tube elongation by 88 and 55%, respectively. However, 30 ,ug/ml AFG1 impaired germination by 99.7% and tube elongation by 100%.
The addition of 15 or 30 gig/ml AFG2 to media which either contained or lacked 3.0mM KH2PO4 did not inhibit germination except at 30 ,ug/ml together with KH2PO4, but, tube elongation was suppressed 34 and 26% at 15 and 30 ,ug/ml, respectively, for medium which was not supplemented with KH2PO4. Both suppressions were significant at the 95% confidence level. When the medium was provided with 3.0mM KH2PO4, a 12% (significant) reduction in tube elongation was observed. Therefore, both lily pollen germination and tube elongation appear to be more sensitive to AFG1 than AFB1, 270 AFB2 or AFG2. However, germination experiments with AFG1 which would employ pollen of high viability should be performed to substantiate the results reported here for pollen of low viability. In addition, it is desirable to assess the ability of lily pollen to germinate and elongate tubes at < 15 ,ug/ml AFG1.

Do Mixed Aflatoxins Affect Seed Germination and/or Root Elongation of Peanuts, Corn, Barley and Oats?
Inclusion of mixed aflatoxins at 31.5 ,ug/ml in the imbibition medium was without a significant effect on either germination ofArachis hypogaea seeds or elongation of their roots at 72, 144, 168 and 240 hr of imbibition (Table 4).
The lack of an aflatoxin effect on either seed germination or elongation of attached Arachis hypogaea roots is somewhat surprising, since an aflatoxin incidence rate of 19% in consumer peanut products has been reported for the United States and Canada during the years 1972-1975 (7). This unexpected result is coupled with another. Aflatoxin B1 inhibits the germination and elongation of both attached and excised Glycine max, cv. Essex roots (8-10), but field-grown soybeans are relatively resistant to invasion by Aspergillus flavus (11).
Aflatoxin was detected in only 2 of 866 soybean samples analyzed by the Department of Agriculture and at total aflatoxin amounts of 10-11 ugi/kg. When Zea mays seeds were imbibed in mixed aflatoxins over the concentration range of 0.36-11.60 ,ig/ml, the percent inhibitions of seed germination were 9.5, 6.0, 13.1, 22  aLots of stored (6 months, 400) Lilium longiflorum, cv. Ace pollen (20 mg fresh weight) were added to 10 ml Dickinson's medium with or without 3.0mM KH2PO4 and 15 or 30 kg/ml AFB2, AFG1 or AFG2; pollen was germinated in sterile Petri dishes at 2600 for 4 hr when 1 ml of formaldehyde was added to the dishes; drops were removed at random and % germination and tube lengths measured with an ocular micrometer; data are means and standard deviations of four replicates for the AFB2 and AFG1 treatments and six replicates for AFG2 without phosphate; data for AFG2 with phosphate are averages for two experiments; the number of grains scored was 2300-2900 and tubes measured were 300 per treatment; statistical analyses indicated that the differences in mean germination percentages and tube lengths between toxin-treated and untreated pollen were significant in every case. aOne hundred peanut seeds were treated with Botec fungicide; 50 seeds were treated with aflatoxin and 50 seeds with H20; seeds were arrayed in a straight line on laboratory paper toweling in sterile Petri dishes so that the embryo ends pointed toward the dish's center; 15 ml of sterile H20 containing 31.5 ,g/ml mixed aflatoxins or 15 ml H20 were added to each dish; the dishes were sealed with parafilm and tilted to obtain an angle which allowed the roots to elongate linearly in order to facilitate measurement; seeds were germinated for the above times in the dark at 260C when root lengths were measured with a mm rule; root data are means and standard deviations for seeds in ten dishes (five seeds/dish). and 11.6 ,ug/ml were statistically significant. aSeeds were surfaced sterilized in 10% Chlorox for 10 min and then rinsed five times in 300 ml distilled H20 prior to sowing; seeds in lots of 15 were placed in sterile Petri dishes containing a single layer of ifiter paper; 10 ml of aflatoxin solution was added to each dish, and the dishes were wrapped and allowed to incubate 40 hr at 2400; at 40 hr, percent germination and root elongation were determined by using a dissecting microscope; to measure uptake 1 ml germination medium was removed from each dish prior to and following germination; the data are means and standard deviations of three experiments with duplicate dishes for each treatment within an experiment.
117 hr of imbibition, respectively ( Table 6). The effects of 31.5 ,ug/ml mixed aflatoxins on seed germination and root elongation of various Hordeum vulgare cultivars are summarized in  39,63 and 89 hr, respectively.
As for the incidence of aflatoxins in small grains in "commercial channels" in the United States  Table 4 except that the incubation times were 65, 89 and 117 hr; the differences in root lengths between treated and nontreated seeds of every cultivar were statistically significant at all times tested except for cultivar Windsor at 65 hr.  Table 4 except that 10 seeds and 10 ml of test solution were added per dish and the imbibition times were 39, 63 and 89 hr; the differences in root lengths between treated and non-treated seeds of every cultivar were statistically significant at all times tested with the exception of cultivar Henry at 39 hr. 272 Environmental Health Perspectives during the years 1968-1975, 3 of 416 oat and 0 of 254 barley samples examined contained AFB1 (7).

Are Pollen and Seed Germinations
Equally Sensitive to Aflatoxins?
Because both the types of the exogenously supplied aflatoxins and probably their uptakes are variable for the plant systems summarized in Table 8, it may be somewhat misleading to compare the effects of aflatoxins on percent germination of various seeds with those of pollen. However, when such a comparison is made, it is apparent that concentrations above  ,ug/ml are usually required to impair imbibition in a solution containing 5.8 and 11.6 ,ug/ml AFB1 for 18 hr resulted in 40 and 80% germination except for Glycine max. In the latter, inhibitions of seed germination. In contrast, when the medium was supplemented with 3.0mM KH2PO4, administration of 25 and 30 ,ug/ml AFB1 for 4 hr reduced pollen germination by 27.3 and 45.1%. However, mixed aflatoxins consisting primarily of AFG1 do not inhibit germination either with or without supplementing the medium with KH2PO4. This suggests that one reason for the failure of investigators to observe an effect of aflatoxin on seed germination at rather high aflatoxin concentrations could have resulted from the use of mixed aflatoxins rather than pure AFB1.  Table 9 demonstrates that aflatoxin concentrations > 30 ,ug/ml are required to inhibit the elongation of tissues of the majority of those plants thus far examined. This comparison may be more meaningful than that for germination since AFB1 was the aflatoxin of choice in most of the investigations summarized in Table 9. However, data on toxin uptake for the various systems are not available. Given this limitation, it appears that the elongation of Lilium longiflorum, cv. Ace pollen tubes is no more sensitive to aflatoxins than that of a variety of plant tissues. Furthermore, tube elongation is considerably less responsive to AFB1 than Onoclea sensibilis protonemal development and Glycine max root elongation.
Is the Pollen Bioassay as Sensitive to Aflatoxins as the Commonly Employed Animal Bioassays?
Comparison of the summaries for the effects of aflatoxins on germination (Table 8) and elongation ( Table 9) of lily pollen with the effects on various animal systems (Table 10) reveals that pollen bioassay is less sensitive than the commonly employed animal bioassays.  14.2 and 58.0% inhibition at 10 and 100 ,g/ml; no inhibition of radicle elongation at 1 ,ug/ml AFB1; 23.9% inhibitions at 10 and 100 ,g/ml Inhibition of growth of four strains by AFB1 at 1 ,g/ml Nontreated excised roots followed a sigmoidal growth curve with a dry wt. increase from 100% (0 hr) to 108.5% (24 hr); 4.5% dry wt. decline at 4 hr increase to 101.5% (8 hr); and decrease to 99% (12 hr) at 20 Fjg/ml AFB1 Inhibition of root elongation by approximately 50% at 100 ,ug/ml AFB1 % inhibition of attached root elongation was 14 (48 hr) and 26 (140 hr) at 2.9 ,ug/ml AFB1; 21 (10) Dashek et al.  Comparison of the data in Tables 8 and 9 with those in Table 11 shows that the pollen aflatoxin bioassay is not as sensitive as the current analytical methodology for quantifying aflatoxins.

Does the Pollen Bioassay System Have Any Utilitarian Value?
The value of this system is in its rapidity and inexpensiveness as well as the fact that one does not need sophisticated instrumentation and a highly trained technician to perform the bioassay. Furthermore, pollen germination and subsequent tube August 1981 elongation are especially sensitive to AFG1. However, these advantages are offset by the lack of sensitivity of the bioassay. We suggest that Onoclea sensibilis spores and/or Glycine max seeds be adopted as organisms of choice for the development of those aflatoxin bioassays which would employ plants.
Is the Pollen Bioassay as Sensitive to Aflatoxins as the Commonly Employed Animal Bioassays?
Comparison of the summaries for the effects of aflatoxins on germination (Table 8) and elongation (Table 9) of lily pollen with the effects on various animal systems (Table 10) reveals that the pollen bioassay is less sensitive than the commonly employed animal bioassays. Comparison of the data in Tables 8 and 9 with those in Table 11 shows that the pollen aflatoxin bioassay is not as sensitive as the current analytical methodology for quantifying aflatoxins.

Does the Pollen Bioassay System have any Utilitarian Value?
Experimental data indicate that pollen systems are not a particularly useful bioassay for aflatoxin hepatocarcinogenicity.