Clinicoimmunologic studies on Phoenix sylvestris Roxb. pollen: an aeroallergen from Calcutta, India


Pampa Chakraborty, Division of Palynology and Environmental Biology, Department of Botany, Bose Institute, 93/1 Acharya P.C. Road, Calcutta – 700 009, India


Background: This study highlights the allergenicity and allergenic components of the pollen of Phoenix sylvestris Roxb. (PS), or date sugar palm, which is predominantly airborne in the air of Greater Calcutta.

Methods: A 2-year aerobiologic survey was performed by Burkard sampler. PS pollen extract was used in skin tests of allergic patients, fractionated by (NH4)2SO4 and the SephacrylS-200 column. The allergenicity of each fraction was checked by skin test and IgE ELISA inhibition. The principal allergenic fraction, Fr.IIa, was separated in 11% SDS–PAGE, and its allergenicity was confirmed by IgE ELISA inhibition and immunoblotting.

Results: PS pollen grains were found to be prevalent in the air of the suburban zone of Calcutta from January to March with a peak in February. The pollen extract showed high (44.07%) positive skin reaction on 540 respiratory allergic patients. Among the (NH4)2SO4 cut fractions, Fr.II was the most active one, and it was resolved into four subfractions in the Sephacryl S-200 column. Fr.IIa was the principal allergenic fraction, showing the presence of two components of 33 and 66 kDa in SDS–PAGE. In IgE immunoblotting, both of the components were found to be allergenic.

Conclusions: The PS pollen grain is an important aeroallergen from Calcutta, India. The 33- and 66-kDa components are the major allergens present in the relevant pollen extract.

Phoenix sylvestris Roxb. (PS), or date sugar palm, commonly occurs in India up to an altitude of 1500 m ( 1), in both wild and cultivated forms. The tree is economically important as it yields edible fruit (khejur), sugar (gur), and a traditional alcoholic drink (toddy). PS pollen was reported to be airborne and allergenic ( 2, 3). Other reports have established that P. dactylifera (date palm) is an allergenic pollen-producing tree in Saudi Arabia ( 4, 5), like P. canariensis in Spain ( 6).

The present study provides more clinicoimmunologic data on another species of the same genus. After having recorded the seasonal and diurnal periodicity of this species, studies were performed to identify and isolate its major allergenic components from the extracts used in the allergy clinic for diagnosis and therapy. Such a study may be helpful in standardization of PS pollen extracts in order to optimize the diagnosis and sensitization and to ensure the consistency of extracts from batch to batch.

Material and methods

Aerobiologic studies

Continuous air sampling was conducted at the Madhyamgram Field Station of the Bose Institute, in a suburban area 19 km north of central Calcutta over a 2-year period (1 July 1994 to 30 June 1996) with a Burkard 7-day sampler placed on a 0.5-m-high concrete base. The exposed tapes were mounted and counted according to the guidelines of the British Aerobiology Federation ( 7) for study of the seasonal and diurnal periodicity.

The vertical profile, i.e., the frequency changes of airborne PS pollen with height, was studied in the same site with six rotorod samplers (2800 rpm) mounted on a mast at equidistant heights ( 8), up to 5 m during peak days.

Source material

Fresh pollen samples were collected from the male date sugar palm trees during the pollination months. The batch contained less than 5% of nonpollen impurities.

Preparation of extract

The pollen was defatted with diethyl ether and then extracted in phosphate-buffered saline (PBS, 0.1 M Na-phosphate containing 0.15 M NaCl, pH 7.2) by continuous stirring at 4°C for 16 h in a 1:50 (w/v) ratio. After centrifugation at 12 500 g for 40 min, the clarified extract was dialyzed in PBS and passed through a 0.22-μm Millipore filter (Millipore Corp., Bedford, MA, USA). The filtrate was then lyophilized and stored at −20°C in aliquots of known volume in sterile vials.

Determination of soluble protein

The protein content of crude extract and different fractions were determined according to Lowry et al. ( 9) with BSA as the standard.

Fractionation of PS pollen allergen

The whole extract was fractionated in the ranges of 0–30%, 30–60%, and 60–90% saturation of (NH4)2SO4. Each precipitated fraction after being dissolved in PBS was separately dialyzed (molecular weight cutoff limit of 10 000) to remove the traces of ammonia, and stored at −20°C. These fractions were referred to as Fr.I, Fr.II, and Fr.III, respectively. The allergenic activities of the fractions were determined by skin prick tests (SPT) and ELISA inhibition analyses.

The principal allergenic fraction, i.e., Fr.II, was then subjected to gel filtration on a 2.8×48 cm Sephacryl S-200 column equilibrated with 10 mM PBS, pH 7.2. An extensively dialyzed and concentrated (28 mg) protein sample was run at the flow rate of 12 ml/h at 4°C after loading. Fractions were collected in 3-ml aliquots (void volume=70 ml), and the elution was monitored at 280 nm with a spectrophotometer.

Skin prick tests (SPT)

SPT were carried out with the P. sylvestris pollen extract (1:50 w/v) and different fractions (100 μg protein content/ml in PBS) on adult respiratory allergic patients with relevant case reports attending the Institute of Child Health, Calcutta. Histamine diphosphate (1 mg/ml) and PBS were used as positive and negative controls, respectively. Tests were performed with 20-μl aliquots of allergen solution placed on the ventral side of the forearm, and each site was pricked with a disposable hypodermic (no. 26) needle. The reactions to whole extracts or fractions of the relevant pollen were measured after 20 min and were graded from 1+ to 3+ (1+= erythema of <20 mm in diameter, 2+= erythema of >20 mm diameter, 3+= wheal and erythema) ( 10). None of the patients were receiving immunotherapy during SPT or sera collection. Control sera were collected from five nonallergic subjects with negative skin reaction. The consent of all patients was obtained before SPT and sera collection.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was performed to measure specific IgE levels, the unique immunoglobulin that causes allergy, in individual patient serum sensitive to P. sylvestris pollen by the method of Engvall & Perlman ( 11). Here the conjugate used was antihuman IgE tagged with horseradish peroxidase enzyme (Sigma Chemical Co., USA). The absorbance was measured at 492 nm with the ELISA reader. The individual patient sera having a P/N (the ratio of OD of patient serum to control sera) value greater than 3.5 were pooled for ELISA inhibition.

ELISA inhibition

This analysis was performed by preincubation of pooled patient sera with equal volumes of whole extracts and different fractions of P. sylvestris pollen with different protein content for 18 h at 4°C. Such inhibited sera were added to allergen-coated wells. Then the assays were performed by the method described before (under ELISA). The percent inhibition was calculated by the following formula:


The point of 0% inhibition was obtained by incubating the serum pool with PBS. A dose-response curve was defined by plotting the percent inhibition against the allergen concentration. From this curve, the quantitative value of each fraction needed to obtain 50% inhibition (C50) was determined.

Eleven percent polyacrylamide gel electrophoresis (PAGE) with sodium dodecyl sulfate (SDS) and in native condition

SDS–PAGE analyses of whole-pollen extract and different fractions were performed according to Laemmli ( 12). The isolated allergenic fraction from P. sylvestris pollen was studied in native PAGE according to Hames ( 13).

IgE-specific immunoblotting

Immunoblotting was performed on PVDF membrane according to Sambrook et al. ( 14). After transfer, the membrane was incubated in blocking solution containing 3% nonfat dry milk in Tris-buffered saline (TBS, 50 mM Tris, 150 mM NaCl, pH 7.5) with 0.02% NaN3 for 2 h at 25°C. Then the membrane was incubated with pooled patient sera diluted (1:5 v/v) in blocking solution for 16 h at 4°C, after which the membrane was washed in TBS with 0.02% Tween 20 (TBST) for 3×15 min, under gentle shaking. After washing, the membrane was incubated with antihuman IgE peroxidase conjugate (Sigma) diluted in blocking solution (1:500) for 1 h at 25°C, and this was followed by 3×15 min washing in TBST. The detection was performed with substrate solution (6 mg of diaminobenzidine and 10 μl of 30% H2O2 dissolved in 10 ml Tris-Cl, pH 7.6) containing 0.3% CoCl2 (0.1 ml/cm2) for a few minutes.


In a 2-year aerobiologic survey in the northern suburb of Greater Calcutta, the pollen grains of PS were found to be airborne from January to March. Although the occurrence of the pollen in air was seasonal, during the peak month, i.e., in February, it contributed 15.97% and 14.84% of the total aeropollen load in 1995 and 1996, respectively. Regarding the diurnal periodicities, in the days of peak concentration (up to 26 pollen grains per day/m3), the pollen grains were generally first detected at 10 a.m., they peaked at 11 p.m., and then they progressively declined until 7 p.m. The vertical profile of pollen concentration was studied up to the height of 5 m, and the highest frequency was found at that height.

Skin reaction was studied with whole-pollen extract on 540 respiratory allergic patients with relevant case history, and 44.07% of them showed a positive response. Among the patients with positive response, 16.38% elicited +2 to +3 reactions. Sixty patients were then selected (with positive skin reaction and ELISA P/N value above 3.5) for further SPT with (NH4)2SO4 cut fractions at one time. Among Fr.I, II, and III, the maximum response (68.33%) was elicited by Fr.II. On the other hand, in vitro study by ELISA inhibition with pooled sera of 10 Phoenix-sensitive patients (P/N value above 3.5) also showed Fr.II to be the principal allergic fraction. Fr.II required 0.9 μg protein to reach the C50 value.

The whole-pollen extract of P. sylvestris showed the presence of more than 20 protein bands in 11% SDS–PAGE, whereas Fr.II elicited the presence of nine bands. In the immunoblotting (with P. sylvestris-specific sera pooled from 10 patients with P/N value above 3.5), the whole extract showed the presence of nine allergenic components of different molecular weights ( Fig. 1). In contrast, the allergenic components in Fr.II, as documented in immunoblotting, were four in number.

Figure 1.

IgE-specific immunoblotting of whole extract (A) and Fr.II (B) of PS pollen.

Fr.II was then subjected to gel filtration on the Sephacryl S-200 column. The eluted protein aliquots of four individual peaks ( Fig. 2) were pooled to obtain four fractions, referred to as Fr.IIa, IIb, IIc, and IId.

Figure 2.

Gel filtration of Fr.II of PS pollen in Sephacryl S-200 column.

The allergenic potential of Fr.II and its subfractions was then studied by ELISA inhibition with Fr.II-specific pooled sera from eight patients. Fr.IIa showed 50% inhibition (C50) with 5 μg, Fr.IIc with 10 μg, and Fr.IId with 20 μg of protein, whereas Fr.IIb failed to attain the C50 value within the concentration range studied. Hence, Fr.IIa showed itself to be the principal allergenic fraction derived from Fr.II.

In the native PAGE study, Fr.IIa showed the presence of a single protein component, which was resolved into two components of 33 and 66 kDa ( Fig. 3) in SDS–PAGE. Immunoblotting was performed with pooled sera from five patients (P/N value above 3.5) sensitive to Fr.IIa, and it confirmed both the components to be allergenic ( Fig. 3).

Figure 3.

Fr.IIa of P. sylvestris pollen in 11% native (A), 11% SDS–PAGE (B), and IgE-specific immunoblotting (C).


In the 2-year continuous aerobiologic survey in a suburban zone of metropolitan Calcutta, the pollen of date sugar palm was found to be an important contributor to the aeropollen load during the pollination months. In the days of peak concentration, the diurnal periodicity of the relevant pollen showed a typical mid-day pattern with its peak at 1 p.m. The pollen of this species showed its highest count at a height of 5m, and lower counts at lower heights. Such observations could be related to the height of the source.

The pollen extract showed a remarkable level of positivity in SPT. Both in vivo and in vitro tests, i.e., SPT and ELISA, established that Fr.II derived from (NH4)2SO4 precipitation of the whole extract had the highest allergenicity, a fraction which was further resolved into four subfractions by gel filtration. In ELISA inhibition, Fr.IIa was found to be the principal allergenic fraction at 66–150 kDa. In native PAGE, Fr.IIa exhibited the presence of a single protein component, resolved into two components of 33 and 66 kDa in SDS–PAGE.

In IgE-specific immunoblotting, the whole extract showed the presence of nine allergenic components, four of which were present in Fr.II. In the case of Fr.IIa, both the protein components (33 and 66 kDa) occurring in SDS–PAGE were found to be allergenically significant in immunoblotting. The results thus demonstrate that Fr.IIa contains the major allergens of P. sylvestris pollen having the potential to cause the IgE-specific respiratory allergy in susceptible individuals. Similar observations were recorded by Harfi et al. ( 4), who recorded approximately 12 IgE-reactive bands in the date palm (P. dactylifera). The report showed that the group of bands between 25–35 and 65–72 kDa were always the first to appear consistently in all IgE blots, significantly corroborating the present result for 33- and 66-kDa major allergens. For date palm, Kwaasi et al. ( 5) reported six major allergens, with the bands of 12, 14.4, 57, and 65–67 kDa binding to 85–93%, and those of 28–30 and 37–40 kDa binding to 60–80% of atopic sera. P. sylvestris pollen has already been reported to have some allergenic relation with the pollen of other tropical palm tree species ( 15). Here the 33- and 66-kDa components could be the members of major allergens shared between the two species of this genus having two distinctly different geographic origins.


We thank Dr (Mrs) I. Roy, Dr (Mrs) S. Chatterjee, and Ms N. Bhattacharya of the Institute of Child Health, Calcutta, for help in clinical study and sera collection. We also thank the Council of Scientific and Industrial Research, India, for providing a scholarship to P. Chakraborty, enabling the team to complete this work.