Content uploaded by Fanie R Van Heerden
Author content
All content in this area was uploaded by Fanie R Van Heerden on Jul 30, 2014
Content may be subject to copyright.
Physical and chemical characteristics of Aloe ferox leaf gel
C. O'Brien, B.-E. Van Wyk ⁎, F.R. Van Heerden
1
Department of Botany and Plant Biotechnology, University of Johannesburg, P.O. Box 524, Auckland Park 2006, Johannesburg, South Africa
Abstract
Aloe ferox leaf gel differs substantially from that of Aloe vera but almost no commercially relevant data is available this species. Leaf dimen-
sions, gel yields and gel compositions were studied, based on samples from several natural populations. Glucose is the only free sugar in aloe gel
(0.1 to 0.4 mg ml
−1
in A. ferox). Monosaccharides released after hydrolysis show potential for gel fingerprinting and allow for a distinction be-
tween A. ferox and A. vera. The former yields various combinations of glucose and galactose as main monosaccharides, while the latter yields only
mannose. Further variation studies are recommended because A. ferox appears to have three different gel chemotypes. Conductivity shows species-
specific ranges —in A. ferox below 3000 μScm
−1
in fresh gel and above 3100 μScm
−1
in aged gel (corresponding values for A. vera were 1670
and 1990 μScm
−1
). The level of phenolic (bitter) compounds in A. ferox gel can be reduced by treatment with activated charcoal, resulting in a
small loss of total dissolved solids. Alcohol precipitable solids and insolubility are useful variables for quality control of gel powder. The methods
and data presented are the first steps towards developing quality criteria for A. ferox leaf gel.
© 2011 SAAB. Published by Elsevier B.V. All rights reserved.
Keywords: Free sugars; Gel composition; Gel conductivity; Gel yield; Hydrolyzed sugars; Organic acids; Quality control
1. Introduction
Aloe ferox Mill. (=Aloe candelabrum A.Berger), commonly
known as the bitter aloe or Cape aloe (also khala,umhlaba,bit-
teraalwyn) is a polymorphic species indigenous to the Cape
coastal region, from Swellendam in the west to the southern
parts of KwaZulu-Natal in the east (Reynolds, 1950; Van
Wyk and Smith, 1996; Glen and Hardy, 2000). It is a single-
stemmed aloe with erect racemes of red, orange, yellow or rare-
ly white flowers and spreading or gracefully curved thorny
leaves. Northern forms of the species, previously known as A.
candelabrum, are morphologically, genetically and chemically
within the range of variation of A. ferox (Viljoen et al., 1996).
Aloe marlothii A.Berger is a possible alternative source of
gel products. This species formed the basis of a pharmaceutical
product known as “Natal aloes”that was discontinued in the
late 19th century. Aloe arborescens Mill., currently under com-
mercial development in Japan and Italy (known as “Japan aloe”
or “Kidachi aloe”) is another possible alternative source of gel
products (Van Wyk et al., 2009).
A. ferox has been used since ancient times as traditional med-
icine —a San rock painting depicts the plant (Reynolds, 1950)
and it has a well-documented history of use as medicine (see
Grace, 2011). However, the use of the inner, non-bitter gel as a
food supplement is a recent development —no documentation
of a food use is found in the literature except the production of
jam (preserve) by Cape farmers (Palmer and Pitman, 1972;
Fox and Norwood Young, 1982; Palmer, 1985; Rood, 1994).
The gel of Aloe maculata All. and Aloe zebrina Baker, however,
is used as famine food in case of an emergency, and the flowers
of several species, including A. ferox, contain nectar which is
eaten by children (Fox and Norwood Young, 1982).
Aloe vera L. is undoubtedly one of the most important me-
dicinal plants of the world. The plant provides the raw materials
for a well-researched, well-established, multi-billion dollar in-
dustry with an estimated annual turnover exceeding 110 billion
US dollars (International Aloe Science Council, 2004). The gel
from this species is used mainly in cosmetics and as a tonic
drink but there are numerous other uses in the food industry
(Grindley and Reynolds, 1986; Reynolds and Dweck, 1999;
World Health Organization, 1999; Waller et al., 2004); see
⁎Corresponding author.
E-mail address: bevanwyk@uj.ac.za (B.-E. Van Wyk).
1
Current address: School of Chemistry, University of KwaZulu-Natal Pieter-
maritzburg, Private Bag X01, Scottsville 3209, South Africa.
0254-6299/$ - see front matter © 2011 SAAB. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.sajb.2011.08.004
www.elsevier.com/locate/sajb
Available online at www.sciencedirect.com
South African Journal of Botany 77 (2011) 988 –995
also Reynolds (2004), Park and Lee (2006), Du Preez (2008),
Grace et al. (2008, 2009) and Grace (2011).
In South Africa, the A. ferox gel industry has gained momen-
tum since 1994 when the first gel was produced through a patent-
ed process in a factory at Albertinia (Botha, 1994; Newton and
Vaughan, 1996). Despite the increasing commercial importance
of A. ferox gel, only one scientific study of the gel components
has been published (Mabusela et al., 1990). In view of the large
number of Aloe species, it is surprising that no research has
been done to investigate the commercial potential of other spe-
cies. In 2007, Standards South Africa (a division of the South Af-
rican Bureau of Standards) published a South African National
Standard for Aloe raw materials (Standards South Africa, 2007).
The scope of the standard is to specify the “requirements and
test methods for A. ferox raw material intended for use in consum-
er products including health, cosmetic, health food, medicinal,
veterinary and industrial products”.
Gel composition at a species level differs substantially.
Available published information shows that the gel composi-
tion of A. vera (Choi and Chung, 2003; Waller et al., 2004) dif-
fers from that of A. ferox (Mabusela et al., 1990), and A.
arborescens (Yagi et al., 1985). These species differ in their
acetylated polysaccharides. Nuclear magnetic resonance spec-
troscopy (NMR) is used as a quality control method for A.
vera gel but the absence of acetylated compounds in A. ferox
complicates the application of NMR methods in this species.
Thus, quality control methods that are applied to A. vera gel
cannot be applied with the same success to A. ferox gel. The
gel polysaccharides are different within these two species,
with A. vera releasing mannose (Choi and Chung, 2003) after
a hydrolysis treatment and A. ferox releasing mainly glucose
and galactose (Mabusela et al., 1990).
Since almost all published information is only applicable to
A. vera gel, this study was conducted to explore some chemical
and physical characteristics of the leaf parenchyma gel of A.
ferox. Leaf dimensions, gel fillet yields and gel powder yields
were studied in detail. Other parameters investigated included
total dissolved solids, free and hydrolysed sugars, conductivity,
alcohol precipitable solids and solubility. The expectation was
that such data can provide a better understanding of the basic
principles underlying gel variability, with possible applications
in chemotaxonomy and especially in developing commercial
quality control procedures for A. ferox gel.
2. Materials and methods
2.1. Materials
Mature leaves were harvested from various individual plants
at several different localities throughout most of the natural dis-
tribution area of A. ferox (with permission from the land
owners). Locality details and voucher specimens are listed
with the results in Tables 1 to 4.
2.2. Leaf dimensions and gel firmness
Leaf width and leaf thickness measurements were taken at
the base of the leaf. A penetrometer was used to determine
Table 1
Leaf dimensions and gel firmness in eight populations of Aloe ferox (six leaves from three individual plants were sampled). The values given are averages for six
leaves per plant.
Locality and voucher specimens (all in JRAU) Plants Leaf weight
(g)
Leaf length
(mm)
Leaf width
(mm)
Leaf thickness
(mm)
Gel firmness
(kg pressure)
1. Albertinia (S 34° 05.532′; E 21° 38.528′)O'Brien 60 1 726.0 471.7 101.7 19.5 10.5
2 532.7 437.2 106.0 16.2 10.0
3 628.6 462.7 107.8 17.2 7.1
2. Albertinia (S 34° 08.689′; E 21° 39.309′)O'Brien 61 1 523.0 401.7 132.8 10.8 9.6
2 334.4 323.3 96.8 14.3 6.4
3 305.0 336.7 88.3 13.5 8.1
3. Seweweekspoort (S 33° 27.480′; E 21° 25.485′)O'Brien 62 1 917.7 597.7 119.7 18.7 8.0
2 950.7 596.7 127.5 22.2 6.9
3 994.7 534.2 136.3 20.2 6.2
4. Uniondale (S 33° 47.003′; E 23° 29.115′)O'Brien 63 1 612.0 480.5 101.5 11.8 8.8
2 868.2 554.3 100.3 18.3 8.1
3 469.3 472.5 94.2 11.8 9.3
5. Uniondale (S 33° 34.582′; E 23° 10.390′)O'Brien 64 1 1028.2 619.5 109.8 18.7 7.4
2 800.5 493.7 106.3 18.2 12.0
3 773.7 519.0 104.2 17.5 11.1
6. Fort Beaufort (S 32° 55.119′; E 26° 28.647′)O'Brien 65 1 474.0 449.7 104.0 14.5 7.9
2 480.8 425.3 90.0 17.0 9.3
3 569.8 460.4 98.5 14.1 8.4
7. Balfour (S 32° 32.318′; E 26° 41.414′)O'Brien 66 1 224.1 406.7 83.7 08.3 5.7
2 753.9 528.0 132.4 18.4 6.3
3 333.0 384.8 89.8 11.2 9.4
8. Seymour (S 32° 34.010′; E 26° 44.868′)O'Brien 67 1 766.3 550.2 113.2 18.5 9.1
2 653.0 565.8 121.8 15.0 7.9
3 956.5 560.5 112.5 20.3 7.6
Mean value 653.1 484.7 107.5 16.1 8.3
989C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
the firmness of each individual A. ferox gel fillet. A probe of
11 mm in diameter was pushed into the gel from the middle
portion of each leaf to a depth of 8 mm. The values recorded
were an average of three repetitions and the unit of measure-
ment is kg pressure.
2.3. Hand filleting and gel yields
Harvested leaves were manually filleted in the laboratory on
a stainless steel surface. Whole leaves were thoroughly washed
and scrubbed to remove mud and bitter exudates before weigh-
ing. The sides, base and tip of each leaf were removed, and the
inner leaf portion cut longitudinally into strips. The gel paren-
chyma was then cut away from the rind with a sharp knife.
Gel fillet weights were recorded for each leaf. All fillets were
then liquidised and filtered through Whatman No. 4 filter
paper, applying vacuum until all liquid was removed. The gel
liquid recovered was weighed. Gel fillet yield and gel liquid
yield were calculated using the following equations:
Gel f il let yield %ðÞ¼
W eight of gel f illet
W eight of leaf 100
Gel liquid yiel d %ðÞ¼
W eight of gel l iquid
W eight of gel f illet 100:
2.4. Activated charcoal treatment
For some applications, gel samples were de-bittered using
10 g of granular activated charcoal per 250 ml liquid gel. The
suspension was left at 4 °C for 2 h (with periodic stirring)
after which it was passed through Whatman No. 1 filter paper
to remove the coarse granules of charcoal and then through
Celite Filteraide to ensure complete removal of the charcoal.
2.5. Freeze-drying of aloe gel
A bench-top Virtis freeze-dryer (8 l) was used for the drying
of gel samples. Gel extracted from the leaves was loaded in
250 ml volumes onto the dryer and dried at a low temperature
(−55 °C) under a high vacuum (50 mTorr) for a cycle of
36 h. Samples were ‘shelled’using liquid nitrogen to increase
the drying surface area and thus decrease the total drying
time. The weight of the liquid gel (250 ml) and resulting gel
powder was recorded for further yield determination. The gel
powder yield was calculated using the equation below.
Gel powder yield %ðÞ¼
W eight of gel powd er obtained f rom 250 ml gel
W eight of 250 ml gel 100
2.6. Hydrolysis of aloe gel
Freeze-dried, de-bittered aloe gel (20 mg) was suspended in
2 ml of a 2 M trifluoroacetic acid (TFA) aqueous solution and
heated at 120 °C for 2.5 h. Vials were shaken at 15 min inter-
vals to stop residues from sticking to the bottom of the vials.
Once hydrolysed, samples were taken to dryness under vacuum
with the addition of 1 ml methanol to accelerate the drying
process.
Table 2
Individual leaf weights, gel fillet weights, gel liquid weights and gel yields (as
fillet or liquid) for eight Aloe ferox plants (six leaves per plant) from the same
eight localities listed in Table 1.
Locality
and plant
number
Leaf
number
Leaf
weight
(g)
Gel
fillet
weight
(g)
Gel liquid weight
(g) (pooled value of
six leaves)
Gel
fillet
yield
(%)
Gel
liquid
yield
(%)
1/1 1 496.3 200.6 1067.6 40.4 88.9
2 523.9 232.0 44.3
3 440.2 172.0 39.1
4 454.3 232.7 51.2
5 464.8 205.8 44.3
6 464.8 158.4 34.1
2/1 1 1057.8 515.9 1570.3 48.8 73.9
2 698.4 328.7 47.1
3 703.9 350.3 49.8
4 616.9 229.3 37.2
5 795.9 361.9 45.5
6 754.9 337.1 44.7
3/1 1 176.7 71.9 426.0 40.7 76.6
2 266.9 110.9 41.6
3 179.5 76.7 42.7
4 220.8 108.6 49.2
5 237.4 79.2 33.4
6 263.2 108.9 41.4
4/1 1 1315.5 714.8 1359.3 54.3 43.9
2 1305.9 698.1 53.5
3 854.0 379.4 44.4
4 1099.3 580.9 52.8
5 894.8 456.8 51.1
6 699.5 263.0 37.6
5/1 1 557.2 264.3 903.8 47.4 56.8
2 655.4 288.5 44.0
3 694.1 313.6 45.2
4 658.4 269.2 40.9
5 656.3 286.1 43.6
6 450.4 160.0 35.5
6/1 1 783.1 400.9 1310.6 51.2 59.7
2 737.4 397.4 53.9
3 764.7 361.6 47.3
4 752.9 387.5 51.5
5 565.2 283.9 50.2
6 752.9 364.9 48.5
7/1 1 554.7 205.8 1028.8 37.1 75.5
2 555.5 247.1 44.5
3 495.9 210.7 42.5
4 553.5 253.1 45.7
5 522.1 245.0 46.9
6 466.4 201.3 43.2
8/1 1 846.2 423.8 2252.6 50.1 81.1
2 968.9 492.4 50.8
3 845.3 414.3 49.0
4 925.4 483.7 52.3
5 878.6 458.3 52.2
6 1037.4 503.9 48.6
990 C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
2.7. Chromatographic analysis of free and hydrolyzed sugars
Monosaccharides present in aloe gel or in hydrolyzed gel
were analysed directly by thin-layer chromatography (TLC)
and high-performance liquid chromatography (HPLC) or as
alditol acetate derivatives by gas chromatography (GC).
2.7.1. Thin-layer chromatography (TLC)
Various solvents systems were used for TLC on silica gel
plates, of which butanol:acetic acid:diethyl ether:water in a
ratio of 9:6:3:1 (Harborne, 1988) gave the best results. Plates
were developed twice to allow for better separation of poorly
separating sugars (they were taken from the solvent, dried and
then redeveloped to the original solvent front level). The sugars
were detected by spraying with aniline phthalate dip or with
chromic acid. Dried plates were heated at 110 °C for approxi-
mately 20 min or until a colour change occurred. Spot colours
and R
f
values were compared to those of reference standards.
Freeze-dried aloe gel samples were reconstituted in distilled
water to form a 1% solution (w/v). This dissolved sample was
passed through a Cameo 0.22 mm nylon filter after which
5μl was spotted onto the TLC plates and treated and developed
in the same manner as the standards.
2.7.2. High-performance liquid chromatography (HPLC)
A Shimadzu LC-6AD HPLC system equipped with a Waters
differential refractometer R401 was used. Fresh aloe gel was
passed through a Cameo 0.22 mm nylon filter and 20 μl
injected directly into the system. The separation was performed
on a Hamilton HC-75 H
+
form (305 mm × 7.8 mm) column
specifically designed for the simultaneous detection of sugars
and organic acids. Conditions of analysis were: 0.01 N H
2
SO
4
as solvent with a flow rate of 0.6 ml min
−1
at an ambient tem-
perature. Pure monosaccharides (8 mg ml
−1
) were used as ex-
ternal standards. Peaks generated from the aloe gel were
identified by comparison of their retention times. Concentra-
tions of monosaccharides were calculated using peak area.
2.7.3. Gas chromatography (GC)
Derivatives of the free sugars were prepared according to the
method of Hoebler et al. (1989). Freeze-dried gel was reconsti-
tuted in a 1% solution (w/v) with distilled water, after which
Table 3
Gel powder yield, total dissolved solids (TDS, before and after treatment with activated charcoal to remove phenolic compounds, as % Brix), free glucose level (as
mg ml
−1
), monosaccharides released by hydrolysis (as % of total sugars) and conductivity (in fresh gel and gel left at room temperature for three days, as μScm
−1
)in
nine Aloe ferox gel samples. Some comparative data for Aloe vera gel is presented in the last row. nd = not determined.
Locality and voucher specimen
(all in JRAU)
Plants Gel
powder
yield
(%)
TDS (%
Brix) before
and after
charcoal
treatment
Free
glucose
(mg ml
−1
)
Monosaccharides released by
hydrolysis (% of total sugars)
Conductivity
(μScm
−1
)
Gal Glu Man Xyl Fresh Old
Aliwal North O'Brien 69 1 0.55 2.0 1.9 0.1 42 58 ––2890 3410
Albertinia (Brakkloof)
No voucher
1 0.46 1.7 1.6 0.1 –100 ––2760 3190
Albertinia (Middeldrift)
No voucher
1 0.21 1.5 1.4 0.2 –100 –tr 2702 3325
Albertinia (Vinklaagte)
No voucher
1 0.16 1.5 1.4 0.4 –100 –tr 2780 3395
Fort Beaufort O'Brien 60 1A nd nd nd nd 33.5 66.5 ––nd nd
1C nd nd nd nd 31.5 68.5 –tr nd nd
1E nd nd nd nd –100 ––nd nd
Uniondale O'Brien 63 1A nd nd nd nd –100 ––nd nd
1C nd nd nd nd 49.7 50.3 ––nd nd
1E nd nd nd nd 48.2 51.8 ––nd nd
Aloe vera (ex hort.) O'Brien 51 1 0.60 1.8 1.6 3.3 ––100 –1670 1990
Table 4
Alcohol precipitable solids (APS) and insolubility percentages obtained for gel
samples from eight populations of Aloe ferox as listed in Table 1 (six leaves
from three individual plants were sampled) and a single sample of A. vera for
comparison.
Population and plants Average APS (%) Average insolubility (%)
1A 52.3 25.5
1C 37.5 3.5
1E 33.2 3.0
2A 25.2 7.7
2C 26.5 8.0
2E 27.8 4.7
3A 48.8 7.0
3C 19.5 7.0
3E 27.7 6.0
4A 38.8 4.0
4C 18.8 4.7
4E 59.5 23.0
5A 39.6 8.0
5C 22.2 3.3
5E 37.0 4.5
6A 21.0 5.0
6C 31.6 5.0
6E 23.0 8.0
7A 36.6 4.0
7C 24.8 3.7
7E 20.0 3.5
8A 23.1 2.0
8C 33.5 12.0
8E 27.6 7.1
Aloe vera (ex hort.) O'Brien 51 34.2 20.2
991C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
2 ml of ethanol was added. The solution was centrifuged, the
supernatant decanted, taken to dryness and then reconstituted
with 100 μl distilled water. This sample was further processed
for GC analysis: 2 ml of a sodium borohydride solution in di-
methyl sulfoxide (2:100, w/v) was added to 0.1 ml of sample.
The mixture was stirred for 1.5 h at 40 °C. Glacial acetic acid
(0.2 ml) was added and the test tubes placed on ice. Once
cooled, acetylation was achieved by the addition of 4 ml acetic
anhydride and 0.4 ml 1-methylimidazole. The solution was stir-
red and allowed to stand for 10 min at room temperature. To
decompose any excess acetic anhydride, 20 ml of distilled
water was added and the tubes again cooled on ice. When
cold, 8 ml of dichloromethane was added and the mixture shak-
en and separated in a separation funnel. The lower phase was
removed and washed a further three times using 20 ml of dis-
tilled water. The dichloromethane extract was concentrated
under vacuum and the resultant alditol acetates were taken up
in 1 ml of dichloromethane (CH
2
Cl
2
) and placed in a freezer
30 min prior to GC analysis. Alditol acetates were analysed
on a 30 m × 0.25 mm capillary column DB 100 (J & W Scientif-
ic, Inc.) on a Shimadzu Class GC-17A gas chromatograph with
a flame ionisation detector (FID). Helium was used as a carrier
gas at a flow rate of 4 ml min
−1
. The column temperature was
set at a constant 220 °C, the injector port at 270 °C and the de-
tector at 250 °C. A 1 μl sample in dichloromethane was ana-
lysed for a total run time of 25 min. Peak retention times of
gel samples were compared to those of monosaccharide stan-
dards derivatised as their alditol acetates, following the same
procedure.
2.8. Total dissolved solids (TDS)
To determine the effect of the activated charcoal treatment,
gel was placed on the stage of a hand-held refractometer,
which measures the total dissolved solids (TDS) expressed as
percentage Brix (% Brix). Readings were taken before and
after treatment of aloe gel with activated charcoal.
2.9. Conductivity
Measurements were taken with a Crison Conductimeter 525.
The instrument was calibrated according to the manufacturer's
specifications and readings were expressed in μScm
−1
.Aloe
gels were analysed fresh and then again after a three-day stor-
age at room temperature.
2.10. Alcohol precipitable solids and solubility
These two variables were studied by dissolving freeze-
dried aloe gel (20 mg) into 2 ml distilled water. Samples
were stirred at room temperature for 15 min to obtain a max-
imum solution of freeze-dried gel powder. Thereafter 6 ml of
ethanol (75%) was added and the solution centrifuged for
10 min. The bulk of the upper soluble water phase was dis-
carded. The insoluble precipitate was then taken to complete
dryness by freeze-drying. This procedure was repeated three
times to obtain an average. The precipitate was weighed and
the APS was expressed as a percentage using the following
equation:
APS %ðÞ¼ W eig ht of precipit ate
Original g el powder weight 20 mgðÞ
100:
The solubility (or actually the insolubility) of the gel was de-
termined by reconstituting the APS fraction (obtained above) in
0.5 ml distilled water. Reconstituted samples were stirred for
15 min. The water layer was decanted and the precipitate
remaining was dried (freeze-dried) and weighed. The procedure
was repeated three times to get an average. The insoluble frac-
tions were expressed as a percentage of the original gel powder
weight in the following equation:
Insolubility %ðÞ¼ W eig ht of insolubl e precipitate
Original g el powder weight 20 mgðÞ
100:
3. Results and discussion
3.1. Leaf dimensions and gel firmness
The leaf dimensions of 24 A. ferox plants from eight popula-
tions representing most of the natural distribution area of the
species are presented in Table 1. The mean leaf weight is
about 0.65 kg, the mean leaf length nearly 485 mm, the mean
leaf width almost 108 mm and the mean leaf thickness just
over 16 mm. The exceptional variability is noteworthy, proba-
bly reflecting a combination of genetic and phenotypic effects.
This is in sharp contrast to the remarkable uniformity in com-
mercial A. vera, which represents a single clone (an ancient cul-
tigen) and hence almost no variability except those that can be
ascribed to environmental effects during cultivation. The results
presented in Tables 1 and 2 allowed us to recommend that A.
ferox leaves should not be harvested if they are less than
30 cm long (Standards South Africa, 2007). The gel firmness
and gel yield of small leaves tend to be rather low. Furthermore,
smaller leaves are likely to be immature, indicating that they
have been harvested in the upper part of the rosette, which
should always be left for the plant to recover.
Gels are relatively firm in A. ferox leaves, with firmness
readings ranging between 6 kg and 12 kg. Firmness is impor-
tant, especially for hand filleting, because gel cannot be effec-
tively removed if the leaves are too flaccid.
3.2. Gel and gel powder yields
Table 2 shows the individual leaf weights and gel fillet
weights for eight plants of A. ferox (each with six leaves sam-
pled), as well as the gel liquid weight for all six leaves removed
from each A. ferox plant. Leaf weights collected from different
A. ferox populations are very variable (± 177 to 1316 g per leaf).
Using the hand filleting technique, the average yield of gel fillet
per leaf (w/w) is approximately 50% but the actual values vary
greatly. For a relatively small leaf weighing only 177 g, the
weight of the gel fillet that is removed is 72 g (41% gel fillet
yield). For a very large leaf weighing 1316 g, a gel fillet weight
992 C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
of 715 g is obtained (54% gel fillet yield). In general, small
leaves have proportionally lower gel yields than large leaves.
When considering gel liquid yields (Table 2), the overall recov-
ery of gel liquid from gel fillet (w/w) was high (up to 89% gel
liquid recovery). This is strongly influenced by the water con-
tent of the gel. Cellular particles removed by filtering account
for a relatively small fraction (sometimes b22%) of the liqui-
dised gel fraction but the extreme variability is again notewor-
thy. It is important to note that polysaccharides have a
tendency to bond fairly strongly to the cellulose in filter
paper. A glass fibre filter (Diallo et al., 2003) or centrifugation
is therefore recommended for future work.
The percentage gel powder yields obtained in this study are
found in Table 3. Gel powder yields range from 0.16 to 1.05%.
Gel powder yields obtained after freeze-drying vary greatly be-
tween gel batches from different populations (0.16 to 0.55%).
These populations were wild harvested and undoubtedly dif-
fered in the available soil water and thus the water content of
the leaves. The yield of gel powder per leaf may well be fairly
uniform but when expressed as a percentage of wet gel weight
the yield figures are very variable.
3.3. Total dissolved solids (TDS)
Total dissolved solids (TDS) decreased in gel treated with acti-
vated charcoal (Table 3). This is an indication that some solids are
lost during the treatment. After industrial filtration of A. vera gel,
the mixture of charcoal and celite that remain behind may have a
hexose content of up to 37% (Waller et al., 2004). The decline in
solids using the activated charcoal method developed in this
studyshowedminimalremovalofsolids with an average decrease
of only 0.1 to 0.6% Brix. This is important, as sugars and other
solids are considered to be important components of the gel.
3.4. Free and hydrolysed sugars
All three of the chromatographic techniques (TLC, HPLC
and GC) confirmed that glucose is the only free sugar found
in the gel of A. ferox (and several other species —see O'Brien,
2005). The presence of glucose as practically the only free
monosaccharide in A. vera gel (Christopher and Holtum,
1996; Femenia et al., 2003) is confirmed here. Glucose levels
in A. ferox gel varied from 0.1 to 0.4 mg ml
−1
.
All samples were successfully hydrolysed after treatment at
120 °C for 2.5 h. Glucose was found in the hydrolysed gel of
all species investigated (O'Brien, 2005) except A. vera, which
yielded only mannose. Other monosaccharides released after
hydrolysis included arabinose, galactose and xylose (O'Brien,
2005), with galactose so far found only in A. ferox and A. coop-
eri Baker. In the A. ferox populations investigated, galactose is
either present or absent (Table 3). It seems that there are three
gel chemotypes in this species, irrespective of geographical lo-
cation: glucose only, galactose–glucose in ratio of 1:1 and ga-
lactose–glucose in a ratio of 1:2. Galactose appears to have
potential as a diagnostic fingerprint for A. ferox gels in the
same way as mannose is used in A. vera. However, this idea
needs further investigation. Xylose appears to be a minor
sugar released by hydrolysis, with only trace amounts sporadi-
cally present in A. ferox. It is important to note that pentoses
and 6-deoxyhexoses degrade at a higher rate than hexoses
under acidic conditions and high temperature (Aspinall,
1983), so that our results (aimed at chemical markers for quality
control) do not reflect the actual composition of hydrolysable
sugars. Mabusela et al. (1990) have shown that water extracts
of A. ferox leaf gel contain arabinose and rhamnose.
3.5. Gel conductivity
An increase in conductivity is associated with the prolonged
storage of A. ferox gel at room temperature (Table 3). A possi-
ble explanation for this relationship is that glucose can be con-
verted into lactic acid which can result in an increase of free
ions or conductivity within decaying aloe gels. The conversion
of malic acid to lactic acid deserves further study, taking into
account the daily fluctuations of malic acid (Christopher and
Holtum, 1996; O'Brien, 2005). Levels of conductivity in Aloe
gels appear to be species specific. The conductivity of A.
ferox gel is around 3000 μScm
−1
(Table 3), A. speciosa
Baker above 3000 μScm
−1
(O'Brien, 2005)andA. vera
around 2000 μScm
−1
(O'Brien, 2005). Conductivity appears
to be a good indicator of gel freshness and may have practical
value in the quality control of A. ferox gel.
3.6. Alcohol precipitable solids and solubility
The percentages of alcohol precipitable solids (APS) and in-
solubility of 24 gel samples from eight populations of A. ferox
(and one sample of A. vera) are shown in Table 4. The values
for APS range from 18.8 to 59.5%, with an average of about
32%, while the insolubility varies from 2.0 to 25.5% (average
about 7.1%). Despite this extreme variability, APS (together
with insolubility) is a useful measure for quality control of gel
powders and have practical implications in product formula-
tion. Suppliers link the biological activity of aloe gel with the
APS weight of the gel because the APS value is assumed to
be a direct indication of the amount of polysaccharides within
the gel. As a result, the method described here was taken up
in the South African National Standard for Aloe raw materials
(Standards South Africa, 2007). Alcohol precipitable solids
are nevertheless a crude analytical parameter, as organic
acids, metal ions and glycoproteins are also included in the pre-
cipitate, together with polysaccharides. Thus, APS determina-
tion is directly linked to mineral content and gel solubility. In
the South African aloe industry, a unique second process is
used in combination with alcohol precipitation (Botha, 1994).
In this patented process, the fibrous pulp that remains after
liquidised aloe gel has been filtered off, is repeatedly washed
to remove all bitter substances, after which it is treated with so-
dium citrate. Water is added and the mixture is heated. The so-
dium citrate causes the bonds between polysaccharides and
calcium to be broken. The mixture is then filtered, and the liq-
uid fraction, which contains the calcium-free polysaccharides,
is known as aloe ‘jelly’. This ‘jelly’forms the basis for A.
ferox health drinks in South Africa. The quantitative and
993C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
qualitative relationships between APS and ‘jelly’polysaccha-
rides in A. ferox gel deserve further study. When pectic poly-
saccharides are present, an HPLC method which can
simultaneously measure the presence of both neutral and acidic
monosaccharides in hydrolyzates (Dai et al., 2010) would be
ideal for quality control purposes.
4. Conclusions
Leaf dimensions are variable between plants from the same
population as well as from population to population. The leaves
of A. ferox are firm, allowing for easy hand filleting. Larger, heavi-
er leaves generally result in larger gel fillets using manual filleting
techniques. Gel liquid yields vary considerably within populations
as well between populations, much of which can probably be as-
cribed to varying climatic conditions (especially rainfall and hu-
miditywhichinturndeterminethe water content of the leaves).
Gel yields are very variable and often different to what would be
expected from a superficial inspection of leaf dimensions (without
consideration of the relative thickness of the rind and parenchyma).
Gel powder yields are exceptionally variable and probably strongly
influenced by the water content of the leaves at the time of harvest.
Treatment with activated charcoal (to remove phenolic com-
pounds) may result in some loss of solids from the gel.
It can be concluded that considerable quantitative variation is
found in monosaccharides released after hydrolysis but that the
qualitative pattern is surprisingly invariable. Individual plants of
A. ferox (regardless of provenance) appear to belong to one of
three gel chemotypes (glucose only or galactose–glucose in a
ratio of 1:1 or 1:2). This interesting result should be explored in
more detail, based on a wider sample. In contrast, A. vera gel ap-
pears to be invariable and yield only mannose after hydrolysis.
This study has confirmed that the age and storage conditions
of aloe gels may influence their conductivity. An increase in gel
conductivity may prove to be a useful quality control method
for the local gel industry. Fresh gel of A. ferox has a conductiv-
ity between 2702 and 2890 μScm
−1
, while the older gel sam-
ples are well above 3190 μScm
−1
. Fresh A. vera gel has a
conductivity of well below 2000 μScm
−1
. It will be useful to
measure larger numbers of gel samples to get a better idea of
the total range of variation in this character so that definite cri-
teria can be established for conductivity as a quality control pa-
rameter for A. ferox. Alcohol precipitable solids and
insolubility show much variation in A. ferox, perhaps reflecting
genetic differences between plants and populations, in addition
to environmental effects.
The data presented here highlights important physical and
chemical differences between the leaves and leaf gel of A. ferox
and those of other species (notably A. vera). Despite considerable
variation, several parameters deserve further study and more thor-
ough evaluation as potential quality control variables for A. ferox
leaf gel.
Acknowledgements
Financial support from the National Research Foundation,
The University of Johannesburg and Organic Aloe (Pty) Ltd
is gratefully acknowledged. We also wish to thank various per-
sons from the South African Aloe ferox industry for logistic
support during field work and for providing samples, especially
Chris Pattinson (Organic Aloe) and Ken Dodds (African Aloe).
An anonymous referee provided very useful suggestions for im-
proving the manuscript.
References
Aspinall, G.O. (Ed.), 1983. The Polysaccharides, Vol. II. Academic Press, New
York.
Botha, M.C., 1994. Process for extracting polysaccharides from plant material.
South African Patent ZA 941581.
Choi, S., Chung, M.H., 2003. A review on the relationship between Aloe vera
components and their biological effects. Seminars in integrative medicine
1, 53–62.
Christopher, J.T., Holtum, J.A.M., 1996. Patterns of carbon partitioning in
leaves of crassulacean acid metabolism species during deacidification.
Plant Physiology 112, 393–399.
Dai, J., Wu, Y., Chen, S.-W., Zhu, S., Yin, H.-P., Wang, M., Tang, J., 2010.
Sugar compositional determination of polysaccharides from Dunaliella sa-
lina by modified RP-HPLC method of precolumn derivatization with 1-
phenyl-3-methyl-5-pyrazolone. Carbohydrate Polymers 82, 629–635.
Diallo, D., Paulsen, B.S., Liljebäck, T.H.A., Michaelsen, T.E., 2003. The mali-
an medicinal plant Trichilia emetica; studies on polysaccharides with com-
plement fixing ability. Journal of Ethnopharmacology 84, 279–287.
Du Preez, F.M., 2008. Aloe ferox. Die bittersoet geneser. The Office Publishers,
Port Elizabeth.
Femenia, A., García-Pascual, P., Simal, S., Rosselló, C., 2003. Effects of heat
treatment and dehydration on bioactive polysaccharide acemannan and
cell wall polymers from Aloe barbadensis Miller. Carbohydrate Polymers
51, 397–405.
Fox, F.W., Norwood Young, M.E., 1982. Food from the Veld. Delta Books,
Johannesburg.
Glen, H.F., Hardy, D.S., 2000. Aloe, Aloaceae (First part). In: Germishuizen, G.
(Ed.), Flora of Southern Africa 5,1,1. National Botanical Institute, Pretoria.
Grace, O.M., 2011. Current perspectives on the economic botany of the genus
Aloe L. (Xanthorrhoeaceae). South African Journal of Botany 16, 51.
Grace, O.M., Simmonds, M.S.J., Smith, G.F., Van Wyk, A.E., 2008. Therapeu-
tic uses of Aloe L. (Asphodelaceae) in southern Africa. Journal of Ethno-
pharmacology 199, 604–614.
Grace, O.M., Simmonds, M.S.J., Smith, G.F., Van Wyk, A.E., 2009. Documen-
ted utility and biocultural value of Aloe L. (Asphodelaceae): a review. Eco-
nomic Botany 63, 167–178.
Grindley, D., Reynolds, T., 1986. The Aloe vera phenomenon: a review of the
properties and modern uses of the leaf parenchyma gel. Journal of Ethno-
pharmacology 16, 117–151.
Harborne, J.B., 1988. Phytochemical Methods: A Guide to Modern Techniques
of Plant Analysis, 2nd ed. Chapman and Hall, London. pp. 143–175, 213–
269.
Hoebler, C., Barry, J.L., David, A., Delort-Laval, J., 1989. Rapid acid hydroly-
sis of plant cell wall polysaccharides and simplified quantitative determina-
tion of their neutral monosaccharides by gas–liquid chromatography.
Journal of Agricultural and Food Chemistry 37, 360–367.
International Aloe Science Council, 2004. How large is the aloe market? (News
item, October 2004). www.iasc.org/aloemarket.html 2004.
Mabusela, W.T., Stephen, A.M., Botha, M.C., 1990. Carbohydrate polymers
from Aloe ferox leaves. Phytochemistry 29, 3555–3558.
Newton, D.J., Vaughan, H., 1996. South Africa's Aloe ferox plant, parts and de-
rivatives industry. A trade review. Traffic East/Southern Africa, South Afri-
can National Office, c/o Endangered Wildlife Trust, Johannesburg.
O'Brien, C., 2005. Physical and Chemical Characteristics of Aloe Gels. Univer-
sity of Johannesburg, MSc Thesis.
Palmer, E., 1985. The South African Herbal. Tafelberg Publishers, Cape Town.
Palmer, E., Pitman, N., 1972. Trees of Southern Africa. (3 vols). Balkema, Cape
Town.
994 C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
Park, Y.I., Lee, S.K. (Eds.), 2006. New Perspectives on Aloe. Springer Science
+ Business Media, New York.
Reynolds, G.W., 1950. The aloes of South Africa. The Trustees of The aloes of
South Africa Book Fund, Johannesburg (and various later updated eds).
Reynolds, T., 2004. Aloes: The Genus Aloe. CRC Press, Boca Raton.
Reynolds, T., Dweck, A.C., 1999. Aloe vera leaf gel: a review update. Journal
of Ethnopharmacology 68, 3–37.
Rood, B., 1994. Kos uit die veldkombuis. Tafelberg Publishers, Cape Town.
Standards South Africa, 2007. South African National Standard. Aloe raw ma-
terial. SANS 368: 2007. Standards South Africa, Pretoria. 1. 26 pp. [ISBN
978-0-626-18895-5].
Van Wyk, B.-E., Smith, G.F., 1996. Guide to the Aloes of South Africa. Briza
Publications, Pretoria. (and revised ed., 2004).
Van Wyk, B.-E., Van Oudtshoorn, B., Gericke, N., 2009. Medicinal Plants of
South Africa, 2nd expanded ed. Briza Publications, Pretoria.
Viljoen, A.M., Van Wyk, B.-E., Van der Bank, H., Smith, G.F., Van der Bank,
M., 1996. A chemotaxonomic and biochemical evaluation of the identity of
Aloe candelabrum (Aloaceae). Taxon 45, 461–471.
Waller, T.A., Pelley, R.P., Strickland, F.M., 2004. Industrial processing and
quality control of Aloe barbadensis (Aloe vera) gel. In: Reynolds (Ed.),
Aloes: The genus Aloe. CRC Press, London, pp. 139–207.
World Health Organization, 1999. WHO monographs on selected medicinal
plants, Vol. 1. World Health Organization, Geneva. (Aloe bitters mono-
graph, pp. 33–42; Aloe vera gel monograph, pp. 43–49).
Yagi, A., Nishimura, H., Shida, T., Nishioka, I., 1985. Structure determination of
polysaccharides in Aloe arborescens var. natalensis. Planta Medica 3, 213–218.
Edited by AM Viljoen
995C. O'Brien et al. / South African Journal Of Botany 77 (2011) 988–995
