1124 SREEVIDYA & MEHROTRA: JOURNAL OF AOAC INTERNATIONAL VOL. 86, NO.

6, 2003

DIETARY SUPPLEMENTS

Spectrophotometric Method for Estimation of Alkaloids

Precipitable with Dragendorff’s Reagent in Plant Materials
NARASIMHAN SREEVIDYA and SHANTA MEHROTRA1
National Botanical Research Institute, 2, Rana Pratap Marg, Lucknow, India 226 001

A rapid, easy, and simple spectrophotometric thiocyanates (2) and reinecke salts (3) are available. Chro-
method was developed for the estimation of total matographic methods such as paper (4), high-performance
alkaloids precipitated by Dragendorff’s reagent thin-layer chromatography (5), and liquid chromatography (6)
(DR) in plant materials. It is based on the formation are also available for individual alkaloids, depending on the
of yellow bismuth complex in nitric acid medium type. However, there are certain demerits in the aforesaid
with thiourea. The yellow-colored complex formed methods. With a titrimetric method, there are chances of man-
obeys Lambert-Beer’s law in the concentration ual errors depending on the skill of the person performing the
range of 0.06–50 mg/mL with lmax at 435 nm. Using procedure; hence the values for the samples may vary signifi-
this method, the alkaloidal percentage of certain al- cantly, and an instrument method would be more reliable for
kaloids (ajamalicine, papaverine, cinchonine, validation purposes and in accordance with GLP norms. Mod-
piperine, berberine) and some plant materials con- ern chromatographic methods are not suitable for herbal mate-
taining alkaloids (Berberis aristata, Solanum rials where the therapeutic activity derives from a complex
nigrum, and Piper longum) were determined. The mixture of many closely related compounds or class of com-
method was compared with other methods. It can pounds, which could still be quantified by spectrophotometric
be used for routine analysis of commercial sam- methods or other traditional methods. Generally, reinecke
ples by industries dealing with herbal drugs for salts and cobalt thiocyante salts are not used for routine alka-
standardization of plant materials containing alka- loid detection. When compared with the above 2 reagents,
loids and for alkaloid-containing pharmaceutical Dragendorff’s reagent (DR) is very common and is used for
products. routine detection of alkaloids. In his review on the biological
and phytochemical screening of plants, Fransworth (7) re-
ported that in at least 19 surveys for alkaloids, DR was used.

A
lkaloids are responsible for the therapeutic effect of Many other workers have used DR for screening of alka-
many plant materials such as Rauwolfia serpentina, loids (8–10). Alkaloids were quantified with DR
Strychnos nux-vomica, and cinchona bark. Some alka- titrimetrically by estimation of liberated EDTA with 0.01M
loids, such as aristolochic acids, are toxic. Many references ZnSO4 with Na3BO3 buffer (11) and iodometrically (12). A
are available for the toxicity of aristolochic acids, and plants spectrophotometric method of quantification method based on
bearing them are banned for use in medications used by hu- DR is more convenient for a routine analysis of the total
man beings (1). No single chemical component is responsible alkaloidal content.
for the medicinal properties of plant-based drugs, and their Here we propose a simple, rapid, efficient spectrophotomet-
synergic action or bioenhancement is due to the presence of ric method, which can be used for routine estimation of alka-
other chemical substances in the plant material. Therefore, the loids. The amount of bismuth present is estimated after precipi-
determination of the total amount of different classes of com- tating the alkaloids with DR. DR cannot precipitate purine
ponents is essential for the standardization of the plants. As al- alkaloids (13); however, very few plants contain such alkaloids.
kaloids have therapeutic efficacy and bioenhancing proper-
Experimental
ties, the estimation of total alkaloids in plants bearing alka-
loids and formulations that contain them as therapeutic agents Apparatus
becomes essential.
Even though there are many methods for determining indi- (a) Spectrophotometer.—UV 1601 UV-Vis
vidual alkaloidal content in formulations and extracts, there spectrophotometer (Shimadzu, Tokyo, Japan) with 1 cm UV
are relatively few methods for determination of total quartz cell.
alkaloidal content. Titrimetric methods using different types (b) Analytical balance.—AB54–S (Metler Toledo,
of dyes as indicators and colorimetric methods using cobalt Greifensee, Switzerland).
Reagents
Received February 8, 2003. Accepted by JS May 20, 2003.
1
Author to whom correspondence should be addressed; e-mail: (a) Bismuth nitrate pentahydrate.—Qualigens (Mumbai,
[email protected]. India); standard solution; purity 99.8%.
 SREEVIDYA & MEHROTRA: JOURNAL OF AOAC INTERNATIONAL VOL. 86, NO. 6, 2003 1125

50 mL 2% methanolic acetic acid for 1.5 h, and the extract was

diluted to 100 mL with 2% methanolic acetic acid.
(h) Extraction of Piper longum.—10 g coarsely powdered
plant material (fruits) was Soxhlet-extracted with 100 mL eth-
anol for 2 h, and the extract was concentrated to 10 mL. To the
concentrated extract, 10 mL 10% alcoholic KOH was added
and further diluted to 100 mL with ethanol.
Procedure for Calibration Curve
The calibration curve was obtained with Bismuth nitrate
pentahydrate stock solution. Series dilutions of the stock solu-
tion were made by pipetting out 1, 2, 3, 4, 5, 6, 7, 8, and 9 mL
stock solution into separate 10 mL standard flasks and diluting
Figure 1. Absorption spectrum of bismuth to volume with distilled water. A 1 mL amount of this solution
nitrate–thiourea complex. was taken, and 5 mL thiourea solution was added to it. The
absorbance value of the yellow solution was measured at
435 nm against colorless reagent blanks. Figure 1 shows the
(b) Thiourea.—SD Fine (Mumbai, India). spectrum of the yellow bismuth thiourea solution.
(c) Disodium sulfide.—Qualigens.
Procedure for Assay of Alkaloids and Plant Extracts
(d) Ajamalicine.—Sisco (Sisco Research Laboratories,
Bombay, India); purity 98%. A 5 mL amount of the extract/solution was taken and the
(e) Papaverine.—Lancaster (Morecambe, UK); purity pH was maintained at 2–2.5 with dilute HCl. A 2 mL amount
98%. of DR was added to it, and the precipitate formed was centri-
(f) Cinchonine.—Lancaster; purity 98%. fuged. The centrifugate was checked for complete precipita-
(g) Piperine.—Lancaster; purity 98%. tion by adding DR. After centrifugation, the centrifugate was
(h) Berberine.—Lancaster; purity 98%. decanted completely and meticulously. The precipitate was
All other chemicals used were of analytical grade. We collected further washed with alcohol. The filtrate was discarded and
genuine plant materials of Solanum nigrum and Piper longum the residue was then treated with 2 mL disodium sulfide solu-
from the botanic gardens, National Botanical Research Institute, tion. The brownish black precipitate formed was then centri-
Lucknow, India, and Berberis aristata from Dehradun, India. fuged. Completion of precipitation was checked by adding 2
drops of disodium sulfide. The residue was dissolved in 2 mL
Solutions concentrated nitric acid, with warming if necessary. This solu-
(a) DR.—Prepared by mixing (1) solution of 0.8 g bis- tion was diluted to 10 mL in a standard flask with distilled wa-
muth nitrate pentahydrate in 40 mL distilled water and 10 mL ter; 1 mL was then pipetted out, and 5 mL thiourea solution
glacial acetic acid, and (2) solution of 8.0 g potassium iodide was added to it. The absorbance was measured at 435 nm
in 20 mL distilled water.
(b) Standard bismuth nitrate solution.—Bismuth nitrate
stock solution was made by dissolving 10 mg Bi(NO3)3·5H2O Table 1. Optical characteristics and precision for
in 5 mL concentrated nitric acid and diluting to 100 mL with bismuth–thiourea complex
distilled water. Parameters Characteristics
(c) Thiourea, 3%.—Prepared by dissolving 3 g in 100 mL
Color Yellow
distilled water.
(d) Disodium sulfide, 1%.—Prepared by dissolving 1 g in lmax, nm 343 and 435
100 mL distilled water. Stability Stable
(e) Stock solutions of alkaloid.—10 mg of each pure alka- Beer’s law range, mg/mL 0.05–50
loid was dissolved in 10 mL solvent, e.g., 10 mg berberine was
Molar absorptivity, L/mol 3.039 ´ 103
dissolved in warm distilled water and diluted to 10 mL with
warm distilled water. Similarly, stock solutions of piperine in Limit of detection, mg/mL 0.02
95% alcohol and papaverine, cinchonine, and ajamalicine Regression equation y = mx + c
were prepared in 95% alcohol containing 2% acetic acid. Regression coefficient (R2) 0.9994
(f) Extraction of Berberis aristata.—10 g coarsely pow- Slope (m) 0.0662
dered plant material (root) was extracted with 25 mL 2% Intercept (c) 0.0144
aqueous acetic acid at room temperature for 10 min. The pro- a
Standard error 0.0051
cedure was repeated 3 times. The extracts were mixed and di-
luted to 100 mL with 2% aqueous acetic acid. Coefficient of varianceb 0.1565
(g) Extraction of Solanum nigrum.—10 g coarsely pow- a,b
n = 3 replicates.
dered plant material (whole plant) was Soxhlet-extracted with
1126 SREEVIDYA & MEHROTRA: JOURNAL OF AOAC INTERNATIONAL VOL. 86, NO. 6, 2003

Table 2. Determination of concentrations of different Results and Discussion

alkaloids for recovery and accuracya

Actual concn, Observed concn, The alkaloids are precipitated as (BiI3)(Alk·HI) (15) by DR
Alkaloids mg/mL mg/mL Recovery, % (KBiI4). Bismuth forms a yellow bismuth complex
Ajamalicine 0.100 0.109 109.00 ± 0.82 {Bi[CS(NH2)3]}(NO3)3 in nitric acid medium with thiourea.
Maintaining the pH is very important for precipitation by DR.
Papaverine 0.104 0.105 100.96 ± 0.86
If the medium is more acidic, the precipitate formed dissolves
Cinchonine 0.104 0.102 98.07 ± 0.57 in excessive acid. The bismuth from the alkaloidal complex is
Piperine 0.133 0.128 96.24 ± 0.62 completely released by disodium sulfide (12). If the concen-
Berberine 0.205 0.201 98.04 ± 0.45 tration of bismuth is high, precipitation occurs after addition
of thiourea solution. Also, if the concentration of thiourea is
a
Average of 3 determinations ± RSD, %. high, precipitation occurs. Therefore, dilute solutions and 3%
solution of thiourea was used. The reagents used are stable, and
the complex formed is also stable for more than 24 h. The cali-
against the blank containing nitric acid and thiourea. The bration graph between concentration and absorbance is linear
amount of bismuth present in the solution was calculated by with R2 value 0.9994 and obeys Lambert-Beer’s law in the con-
multiplying the absorbance values with the factor, taking suit- centration range of 0.06–50 mg/mL (Table 1). Because the com-
able dilution factor into consideration. The factor is obtained plex formed is 1:1, the amount of bismuth corresponds to the
from the standard curve, which is a constant for different con- amount of alkaloids present; hence the method of preparation of
centrations. DR does not affect the result. The amount of alkaloid present is
calculated by multiplying absorbance with factor.
Factor = concentration/absorbance The recovery was appreciable, indicating the accuracy of
this experimental procedure (Table 2). Differences of esti-
A comparison of total alkaloids in all the plant materials mated concentrations to actual concentrations varied from
taken was made by using the Method II given in the CCRUM –1.5 to +10%. The differences may be due to the presence of
publication, Physicochemical Standards of Unani Formula- some traces of unreacted reagent in the positive side, and dif-
tions (14). According to this method, the weighed plant mate- ferences in the negative side may be due to the impurity of the
rial (15 g) is extracted with 200 mL chloroform–ether–alcohol alkaloids, the purity of which was only 98%.
(90%; 23 + 8 + 2.5) for 10 min. To this, 6 mL dilute ammonia
When this method was applied to plant materials, the same type
is added, and the flask is shaken for 1 h and allowed to stand
of accuracy was observed, which was determined by addition of
for 8 h with occasional shaking. To this, 10 mL water is added,
and the mixture is shaken vigorously. When the drug has set- different amounts of berberine solution to Berberis aristata
tled, 100 mL of the solution is drawn out and filtered into a alkaloidal extract (Table 3). A comparison with the
separator and washed with a few milliliters of a mixture of sol- Physicochemical Standards of Unani Formulations method (14)
vent ether and chloroform, and shaken with 0.5N sulfuric acid was made to check whether the results obtained by this method
for complete extraction of alkaloids. The combined acid ex- were comparable with the method developed using DR (Table 4).
tract is filtered and made alkaline with dilute ammonia solu- In conclusion, this method will determine the amount of al-
tion. The alkaloids are liberated with chloroform and washed kaloids precipitable by DR and can be applied to single alka-
with chloroform. The chloroform washes and the extract are loid or to alkaloid-bearing plants. This method can be used for
poured in a tared conical flask. The chloroform is distilled off routine analysis of commercial samples by industries dealing
and the solvent is removed completely in a vacuum desiccator. with herbal drugs for standardization of plant materials con-
A 5 mL amount of alcohol is added to the residue, and the sol- taining alkaloids and alkaloid-containing pharmaceutical
vent is again removed. The evaporation with alcohol is re- products. This method is not applicable to plant materials con-
peated, and the residue is dried to constant weight in vacuum taining purine alkaloids because DR cannot precipitate purine
desiccator and weighed as total alkaloids. alkaloids. However, few plants contain such alkaloids.

Table 3. Recovery studies in Berberis aristataa

Amount of alkaloid in Berberis aristata, Added amount of Observed amount of alkaloid,

mg/10 g material berberine, mg/mL mg/10 g material Recovery, %

56.693 1.000 57.543 99.73 ± 0.62

56.693 5.000 61.050 98.95 ± 0.71
56.693 10.000 65.945 98.87 ± 0.56

a
Average of 3 determinations ± RSD, %.
 SREEVIDYA & MEHROTRA: JOURNAL OF AOAC INTERNATIONAL VOL. 86, NO. 6, 2003 1127

Table 4. Determination of total alkaloids in different (2) Deltombe, J., Leboutte, G., & Rosier, N. (1962) J. Pharm.
plant materials by the proposed method using DR and Belg. 17, 236–238
Method II given in PMUa
(3) Weyers, J., & Skora, M. (1962) Diss. Pharm. 14, 201–205
Amount of total
alkaloid by the Amount of alkaloid by (4) Narain, S.T. (1967) Pharm. Zentralle Dtschl. 106, 443–447
proposed method Method II given in (5) Wang, M., & Zhu, M. (1984) Yaowu Fenzi Zazhi 4, 12–15
using DR, mg/10 g PMU, mg/10 g
Plant material material (6) Cui, J., Zhang, G., & Wang, M. (1985) Yaoxue Xuebao 20,
59–66
Berberis aristata 56.693 ± 0.67 58.059 ± 0.55
(7) Fransworth, N.R. (1966) J. Pharm. Sci. 55, 247
Piper longum 19.540 ± 0.78 18.245 ± 0.48
Solanum nigrum 43.612 ± 0.56 43.578 ± 0.75 (8) Kiang, A.K., & Douglas, B. (1957) in Proc. Third Congress,
Pan Indian Ocean Science Association, Section G,
a
Average of 3 determinations ± RSD, %; PMU = Physicochemical Tananarive, India, pp 19–24
Standards of Unani Formulations (14).
(9) Webb, L.J. (1949) in Bull. 241, Commonwealth Scientific
and Industrial Research Organisation, Melbourne, Australia,
99 pp
(10) Abish, E., & Reichstein, T. (1960) Hel. Chem. Acta 43,
Acknowledgments 1844–1861

We thank P. Pushpangadan (National Botanical Research (11) Surovi, Z., Dugandzic, M., & Blagojevic, Z. (1963) Arch.
Farm 13, 299–304
Institute) for providing encouragement and facilities. N.
Sreevidya is grateful to the Council of Scientific and Indus- (12) Petkov, P.A. (1965) Sb. Tr. Visshiya Med. Inst. IP. Pavlov,
trial Research (New Delhi, India) for financial support and to Plovdiv 17, 23–31
Subha Rastogi and R.P. Rastogi for helpful discussions (and to (13) Evans, W.C. (2002) Trease and Evans Pharmacognosy, 15th
M. Vijayakumar for his assistance throughout the work). Ed., W.B. Saunders, Edinburgh, UK, p. 336
(14) Physicochemical Standards of Unani Formulations (1986)
References
Part 1, Central Council for Research in Unani Medicine, New
Delhi, India, pp 229–230
(1) IARC Monographs on the Evaluation of Carcinogenic Risks
to Humans (2002) Vol. 82, IARC Press, Lyon, France, pp 79, (15) Francois, M., & Blanc, L.G. (1922) Comt. Rend. 175,
83–84 273–274