Quick Test for Reed Canarygrass Alkaloid Concentration

A plant alkaloid screening procedure by J.R. Frelich & G.C. Marten from

Quick Test for Reed Canarygrass (Phalaris arundinacea L.) Alkaloid Concentration(1)

by J. R. Frelich and G. C. Marten(2)

HTML Produced by Gnostic Garden

ABSTRACT

We modified a generalized, plant-alkaloid screening procedure to provide a quick, semiquantitative test for concentration of total basic alkaloids in reed canarygrass (Phalaris arundinacea L.). The test is based upon ocular assessment of discernible differences in turbidity or precipitates produced by silicotungstic acid added to plant extracts. The extracts are prepared by macerating frozen grass in ammoniacal chloroform and then by partitioning into 2 N H2SO4.

Alkaloid concentrations determined by a summation of readings over five dilutions using the quick test were highly correlated (r = 0.94, P < .01) with concentrations determined by the quantitative laboratory test of Simons and Marten (1971). The quick test is about four times faster than the Simons and Marten test and should enable plant breeders to select low-alkaloid plants or alkaloid-free plants among large numbers of genotypes.

Additional index words: Palatability, Plant toxicity, Forage anti-quality components, Alkaloid assay.

1 Paper No 8190, Scientific Journal Series, Minnesota Agricultural Experiment Station and North Central Region, Agricultural Research Service, USDA. The authors acknowledge with gratitude the assistance of R. E. Stucker in statistical analyses. Part of a thesis submitted by the senior author as partial fulfillment of the requirements for a Ph.D. degree. Received March 18, 1973.

2 Research Assistant and Research Agronomist, ARS, USDA, Department of Agronomy and Plant Genetics, University of Minnesota, St. Paul 55101.

Relative palatability of reed canarygrass (Phalaris arundinacea L.) is inversely related to alkaloid concentration (Simons and Marten, 1971). Large numbers of genotypes must be screened for alkaloid concentration if plant breeders are to develop highly palatable, low-alkaloid cultivars (Marten, 1973). To facilitate this screening, a quick, simple, and reproducible alkaloid assay would be very desirable. Our objective was to develop such a procedure that would increase the efficiency of the quantitative test reported by Simons and Marten (1971), which allows a maximum of about 75 analyses per week per technician.

Culvenor and Fitzgerald (1963) described a field test that estimates alkaloids in plants by assessing the amount of precipitate formed when silicotungstic acid or Mayer's reagent is added to plant extracts prepared by ammoniacal chloroform extraction and partitioning into 2 N H2S04- Macdowall (1966) also used silicotungstic acid to precipitate nicotine in a quick field test for tobacco (Nicotiana tabacum L.) alkaloids.

Simons and Marten (unpublished data) found that a field test for plant alkaloids proposed by Burns (1964), which employs color development of paper impregnated with Dragendorff's reagent, was unsatisfactory for testing alkaloids in reed canarygrass because of ambiguous color development.

This paper outlines a quick laboratory test for relative reed canarygrass alkaloid concentration, patterned after the field test of Culvenor and Fitzgerald (1963). .4 recommended procedure follows the description of developmental work.

DEVELOPMENT OF THE TEST

Extraction and Partitioning. 'Total alkaloids in 60 samples of reed canarygrass (representing a wide range of alkaloid concentrations) were extracted with ammonical chloroform, with methanol, and with ammoniacal ether. Only the chloroform did not cause formation of foam and excess emulsions at the solution interface during partitioning with sulfuric acid. Chloroform therefore was selected as the extractant during other phases of the work. Fresh-frozen grass (5 g) was macerated in 50 ml of ammoniacal chloroform in a Sorval Omnimixer3, and the extract was filtered through an Eaton-Dikeman No. 615 filter paper into a separatory funnel.

Sulfuric acid was chosen over HCL to partition the alkaloids from the acid-insoluble plant products in the organic extract, because HCL caused foam and emulsion at the interface of the two liquid phases for a few samples. Exactly 6 ml of 2 N H2SOy were added to the extract, and the contents were shaken briefly before allowing the two liquid phases to separate. The chloroform phase was drained off and discarded, and the aqueous phase was transferred to a test tube. Aliquots of 0.25, 0.50, 1.00, 1.50, and 2 ml of-the aqueous phase were pipetted into test tubes. Then 2 N H2SO4 was added to bring the final volume to 2 ml. The solutions were thus prepared at dilution rates of 8X, 4X, 2X, 1.33X, and none, respectively; these correspond to use of 48, 24, 12, 8, and 6 ml of 2 N H2SO4, respectively, in the partitioning step.

Alkaloid Determination. Both Mayer's reagent and silicotungstic acid caused precipitates to form in the presence of reed canarygrass alkaloids, but we elected to use the latter because it gave a coarser precipitate that was easier to assess quantitatively. Also, the color reactions for specific alkaloids were more pronounced with silicotungstic acid than with Mayer's reagent.

3 Mention of a trade name or proprietary product is for identification purposes only and does not constitute a guarantee or warranty of the product by the U.S. Department of agriculture or the University of Minnesota and does not imply its approval to the exclusion of other products that may also be suitable.

The color of the silicotungstate precipitates formed from reagent gramine, DMT (N,N-dimethyltryptamine), and 5-Me0-DMT (5-methoxy-N, N-dimethyltryptamine), were peach, gray, and lavender, respectively. We hypothesized that this meant we could identify the particular type of these three alkaloids (predominant in reed canarygrass) that was present in a grass sample from the color of the precipitate formed; however, the color in most instances was not distinct when two or more alkaloid types were present in one sample. If gramine was the only alkaloid in the grass, it could be readily identified. DMT and 5-Me0-DMT frequently occurred in mxiture, and therefore relative amounts of the two alkaloids could not be determined by this method.

Precipitate Rating System. Silicotungstic acid (12°0) was added to each tube of the aqueous phase extract, and the tubes were agitated. The resulting turbidity or precipitation was rated immediately using the following rating system: 0, an absolutely clear solution, indicative of a very small amount of alkaloids or none; 1, a transparent solution with faint traces of turbidity; 2, a transparent solution with medium turbidity; 3, a translucent and turbid solution; 4, formation of a small amount of precipitate almost immediately; and 5, formation of a large amount of precipitate immediately, indicative of large amounts of alkaloids.

Diluting the aqueous phase of each sample extract provided several opportunities to observe differences. Because we are primarily interested in retaining only those plants with low alkaloid concentration, we must be able to easily distinguish relatively low-alkaloid from relatively high-alkaloid samples. To accomplish this, some samples of each set to be screened must be rated 0 or 1. Thus, the specific dilution or dilutions required must be decided for each major set of samples.

The ratings obtained from each test tube (all dilutions) where summed for the final rating score. An example of the rating system from seven samples, compared with the actual alkaloid concentration, clearly shows that the rating given a particular sample decreased as the dilution increased (Table 1). A 0 rating was common in the 8X solution, whereas a 0 rating was nonexistent in the undiluted solutions. Thus, if we were looking for very low alkaloid-containing plants, only sample one (rated 1 in the undiluted solution) should be retained (undiluted solution is equivalent to using 6 ml of acid during partitioning). On the other hand, the limits might be extended to retaining only those samples rated 1 or 0 in the 1.33X or 2X solutions; these correspond to use of 8 and 12 ml of dilute acid during partitioning.

Table 1. Silicotungstate precipitate ratings for diluted extracts compared with the percentage total alkaloid concentration of seven reed canarygrass samples.

Sample no.	Dilution*					Sum of precipitate ratings	Total alkaloid concentration
	None	1.3x	2x	4x	8x
	----- Ratings -----						% dry wt
1	1	0	0	0	0	1	0.11
2	2	2	0	0	0	4	0.15
3	4	3	1	0	0	8	0.11
4	5	3	1	0	0	9	0.25
5	5	5	3	1	0	14	0.48
6	5	5	5	3	1	19	0.72
7	5	5	5	3	1	21	1.43

* None 1.3x, 2x, 4x, and 8x ape equivalent to use of 6, 8, 12, 24, and 48ml of partitioning solution, respectively.

Relationship of Slammed Precipitate Ratings and Total Alkaloid Concentration, A plot of the sum of the precipitate ratings vs the alkaloid concentration for all 60 plant samples is presented in Fig. 1, and a plot for 24 standard-reagent, alkaloid solutions of gramme and 5-We0-D MT is presented in Fig. 2. Both the plant and the reagent ratings were significantly correlated (P < .01) to the actual alkaloid concentration. The regression coefficients of the two prediction equations were not significantly different (P <\ .05)

Fig. 1. Relationship between total alkaloid concentration, determined by the procedure of Simons and Marten (1971), and the sum of silicotungstate precipitate ratings of reed canarygrass extracts.

Fig. 2. Relationship between total alkaloid concentration, determined by the procedure of Simons and Marten (1971), and the sum of the silicotungstate precipitate ratings of standard alkaloid solutions containing gramine and' 5-Me0-DMT.

The alkaloid ratings by this precipitate test were not corrected for moisture in the sample, because we were primarily interested in a quick test that would provide relative rather than absolute alkaloid concentration. If the sum of precipitate ratings scale (Fig. I) is subdivided into classes and the mean percent dry matter of samples in each class is computed, the following is obtained:

Sum of precipitate ratings	Number of Samples	Mean % DM
<5	16	20
5.8	8	22
9.12	16	22
13.6	9	21
>16	11	21

Because of the uniformity of the means, dry matter concentration evidently did not greatly influence the relationship between the sum of the precipitate ratings of wet samples and actual alkaloid concentration of dry matter.

If one had reason to believe that a wide range of dry matter concentrations existed among members of a specific set of genotypes to be screened for alkaloid concentration by this method, he should actually measure dry matter of each genotype and correct precipitate ratings accordingly. However, determining dry matter in each sample would partly defeat the objective of a quick screening test; and from our experience, this determination usually would not be necessary if field plots were uniformly treated.


RECOMMENDED PROCEDURE
The following is the recommended procedure for sample preparation, extraction, and partitioning:

1. Collect a 50-g sample of fresh grass and freeze as soon as possible at -20 C.

2. Sometime before analysis, chop the grass into small pieces with a heavy-weight paper cutter while the grass is still frozen -` (chopping is easier, faster, and more efficient in the frozen state). Store the chopped sample at -20 C until alkaloid analysis.

3. Mix each sample thoroughly and weigh a 5-g subsample. 

4. Macerate the frozen subsample in 50 ml ammoniacal reagent grade chloroform (0.10 A' ammonia; 6.71 ml conc. NH4OH in 993.29 ml chloroform) in a Sorval Omnimixer, or equivalent blender, at medium speed for 1 minute. (Remove the polystyrene gasket from the blender cup because chloroform dissolves it.)

5. Filter the plant extract through an Eaton-Dikeman No. 615, or equivalent, tilter paper into a 125-ml separatory funnel. Racks of about 15 funnels lead to efficient filtration. Rinse the fibrous residue three times, using 10 ml of ammoniacal chloroform per rinse. Press the filter paper with a spatula to insure sufficient extract recovery, and discard the residue.

6. Add 6 ml of 2 N H2SO4 to the extract and shake vigorously for 15 seconds'. Allow the two liquid phases to separate; usually 5 minutes are sufficient (release pressure buildup in separatory funnels by opening the top at pauses during shaking).

7. Drain the lower chloroform layer and emulsions and discard. Drain the aqueous layer into a test tube (1.5 x 15 cm), stopper, and place in a test tube rack.

For alkaloid determination, we recommend the following procedure:

8. Pipet a 2-ml aliquot of the aqueous extract into a 1.25 )( 10 cm test tube. Add six drops of silicotungstic acid reagent (12 g/100 ml water) per tube to groups of 5 tubes at one time.

9. Agitate the mixture by tapping the bottom of the tubes, and observe the precipitate formation. Alkaloid concentration of the sample dry matter is greater than 0.20% if a dense silicotungstate precipitate (rating of 5) forms immediately (assuming about 20 to 25% DM in 5 g of fresh grass and use of 6 ml of HMSO, for partitioning). Alkaloid concentration is not greater than about 0.20%a if a minute amount of silicotungstate precipitate forms immediately (rating of 4), or if the solution is densely turbid (rating of 3). Alkaloid concentration is less than 0.20%a if the solution is cloudy but transparent (ratings of 1 or 2), and 0.10%a or less if the solution is clear (rating of 0).

4 Other dilutions of the extract may be desirable depending on the operator's selected definition of "low alkaloid" plants. This extraction procedure applies to immature reed canarygrass (regrowth from 3- to 6-weeks-old) containing about 20 to 25% dry matter. Adjustments in extract dilution may be necessary if the grass does not have these properties. Some groups of samples may have such generally low alkaloid concentrations that more than 5 g of fresh grass should be extracted initially to obtain a meaningful range of precipitate ratings. Other groups may have such high concentrations that greater dilutions are required, as shown in Table 1.

CONCLUSIONS

We are proposing a simple test for quick screening of total alkaloid concentration in reed canarygrass' It requires much less time than the quantitative-qualitative test previously used in our laboratory, allowing up to 300 analyses per operator per week (the previous procedure allowed analysis of about 75 samples with equally concentrated effort). The quick test can be undertaken rather successfully by inexperienced people using simple equipment. This precipitate test is designed to allow the operator to set the limits of a negative response. The test provides a means of screening large numbers of plants for relative alkaloid concentration when absolute quantitative tests are not practical (such as in plant breeding programs).

REFERENCES

Burns, R. E. 1964. Field screening of lupines and other plants for alkaloid content. Agron. J. 56 246.
Culvenor, C. C. J., and S. S. Fitzgerald. 1963 A field method for alkaloid screening of plants. J. Pharm. Sci. 52:303-304.
Macdowall, F. D. H. 1966. Rapid field test for tobacco alkaloid content. Can. J. Plant Sci. 46:332-335.
Marten, G. C. 1973. Alkaloids in reed canarygrass. In A. G. Matches (ed.) Anti-quality components in forages. Special Publ. Amer. Soc. Agron., Madison, Wis. (In press).
Simons, A. B., and G. C. Marten. 1971. Relationship of indole alkaloids to palatability of Phalaris arundinacea L. Agron J. 63:915-919.