Abstract: The functional properties of sesame (Sesamum indicum L.) protein isolate were studied. The Sesame Protein Isolate (SPI) was produced by varying four parameters (pH, temperature, time and ratio of flour to water). The aim of the study was to compare the SPI with a commercial soya protein isolate to see if the production had some negative effects to the SPI. According to the results, the functional performances of SPI compared to the commercial soya protein isolate were not significantly different (p<0.05). Both of them showed very similar solubility curves when subjected to the same pH treatment. But the commercial soya protein showed very low protein solubility between pH 4-5.5 (5-10%) for that product but the solubility of SPI was higher (15-19%). For fat absorption and water holding capacity SPI was observed to show low (129 g) and high (302 g), respectively, for the bulk density the SPI was also low (0.169 g mL-1). We observed similar performance for the viscosity at both low and high temperatures (40 and 60°C) and it exhibited good foaming capacity below and above its pI. Similar dispersibility characteristics for both of them and good whipping properties but the SPI exhibited when whipped, a higher foam expansion of 79%, as against 76% for the commercial soya protein isolate. SPI displayed good functional properties which could provide the food industry with new high protein food ingredients to be used as protein supplementation in different food formulations.
INTRODUCTION
Today plant proteins play significant roles in human nutrition, particularly in developing countries where average protein intake is less than that required. Because of inadequate supplies of animal proteins, there has been a constant search for new protein sources, for use as both functional food ingredients and nutritional supplements (Onweluzo et al., 1994). Although plant protein products are gaining increased interest as ingredients in food systems throughout many parts of the world, the final success of utilizing plant proteins as additives depends greatly upon the favorable characteristics that they impart on foods. Therefore, the interrelationships of protein quality and processing parameters, like pH, temperature, time and ratio, that affect the functional performance of plant protein products, is worthy of an extensive investigation.
The development of SPI from defatted sesame flour would provide the food industry with new high protein food ingredients for product formulation and protein fortification. The latter is critically needed in many developing countries particularly in Africa where protein deficiencies remain a major health problem, especially among children (Lawal, 2004).
Functional properties of food protein are important in food processing and food product formulation. Some of these properties are solubility, water holding capacity, oil binding, emulsification, foaming properties, bulk density and viscosity (Jung et al., 2005; Bandyopadhyay and Ghosh, 2002). Nonetheless, some of these properties are affected by the intrinsic factors of proteins such as molecular structure and size and some environmental factors, including the method of protein isolation from the seed (Fuhrmeister and Meuser, 2003). The importance of these properties varies with the type of food products in which the protein isolate is to be used. For instance protein isolates with high oil and water binding capacities are desirable for use in meat, sausages and bread, while proteins with high emulsifying and foaming properties are good for salad dressing, confectionaries, frozen foods and soups (Ahmedna et al., 1999).
Attention on plant protein isolates has been focused mainly on cotton seed, peanut, rapeseed, soya protein and sunflower seed and in some areas commercial preparations are available (Anonymous, 1979; Schenz and Morr, 1996). In contrast the functionality of sesame protein has received little attention particularly when different production parameters say more than two are combined to isolate the protein from sesame that could be used in food formulations. The few studies that have been made are mainly on the properties of the defatted flour or meal, sesame oil, the antioxidant property of sesame and functional properties of sesame protein as influenced by pH only (Khalida et al., 2003; Shahidi et al., 2006; Aly et al., 2000). Very limited information is available for the functional properties of sesame isolate as influenced by other factors during its protein extraction from dehulled sesame seeds.
Thus, the objectives of this study were to study the functional properties of sesame protein isolate which was produced in lab where the pH, temperature, time and ratio of flour to water were varied during it production. Compared some of its functional properties with those of a commercial soya protein isolate.
MATERIALS AND METHODS
Materials
The SPI was produced in the Key Laboratory, School of Food Science and Technology,
Southern Yangtze University Wuxi, P.R. China and the commercial soya protein
isolate was purchased from a supermarket in Wuxi, P.R. China, produced by Fuxin
Flour mill Company, Shanghai, P.R. China. This research was conducted in the
key laboratory of Food Science and Technology, Southern Yangtze University,
Wuxi, PR China between September, 2006 to January 2007.
The chemicals and reagents used were of analytical and food grade quality obtained from the chemical store of Southern Yangtze University Wuxi, P.R. China, manufactured by (SCRC) Sinopharm Chemical Reagent Co., Ltd. Shanghai P.R. China.
Methods
The Sesame Protein Isolate (SPI) was prepared from defatted sesame flour
as described in the flow chart (Fig. 1). The isolation was
done by varying the pH, temperature, time and ratio of flour to water as 12,
45°C, 45 min and 6/100 g mL-1, respectively. The response surface
methodology technique was used to optimize the recovery process of the SPI and
the maximum protein recovered was 94% at the above working parameters, thus
making them the working parameters employed for the production of the SPI used
in this research. The particle size was determine after freeze drying the SPI
by passing it through a 100 mm sieve which was taken that all the particles
of the SPI were less than 100 mm in size.
| Fig. 1: | The isolation of the sesame protein and the production of the Sesame protein isolate |
Solubility
Nitrogen solubility of the proteins isolates was determined by the method
of Beuchat et al. (1975) over a pH range of 1-14. The dispersions were
stirred at different pHs at 24°C for 30 min and then centrifuged in a Beckman
Coulter centrifuge (Avanti J-26XPI; USA) at 4500 rpm for 20 min for both the
SPI and the commercial soya protein isolate. Protein content of the supernatants
was determined by the method of AOAC (1995) using 6.25 as conversion factor:
Where
| PS | = | Amount of protein in supernatant |
| PIS | = | Protein in initial sample |
| PR% | = | Protein Recovered in percentage |
Protein solubility was expressed as percent of the protein content of the sample.
Water-and Oil-Holding Capacity
Water absorption was calculated using the method described by Makri et
al. (2005). The estimations were performed in triplicates. For the fat absorption
the method of Carcea-Bencini (1986) was used with slight modifications. One
gram of protein sample was stirred in 10 mL of distilled water or corn oil and
then centrifuged at 3000 rpm for 30 min. The volume of the supernatant was measured.
The oil-holding capacity was expressed as the number of g of oil held by 1.0
g of protein sample for both the SPI and the commercial soya protein isolate.
Bulk Density
The bulk density was determined according to Wang and Kinsella (1976) using
samples of 20 g and a 150 mL graduated cylinder. Bulk density was calculated
as g mL-1. Estimations were performed in triplicates for both SPI
and commercial soya protein isolate.
Viscosity
Apparent viscosity of the SPI, at different pH levels, was determined by
the method of Quinn and Beuchat (1975). The samples were heated at 40 and 60°C
as these are temperatures considered to be moderate and high, respectively.
The apparent viscosity was determined at those two temperatures with a Brookfield
(Model RVT) viscometer equipped with a No. 1 spindle. Apparent viscosity in
cetipoises (cps) was reported as the average of three readings.
Emulsification Properties
Emulsification properties were investigated ssentially according to the
procedure of Yasumatsu et al. (1972) for the two samples. Samples (5
g) were suspended in deionised water (40 mL). The pH was adjusted to pH 7 using
1 N, NAOH or HCL. After stirring for 15 min, the pH was checked again to confirm
that it was still reading pH 7 and the volume made up to 50 mL. Soy oil (50
mL) was added and blended for 3 min (MSE homogenizer, at Maximum speed). The
emulsion was divided between two 50 mL centrifuge tubes and centrifuged (4500
rpm for 20 min). The ratio of height of the emulsified layer to the total height
of fluid was calculated and the emulsification activity expressed at this ratio
x100. Emulsion stability was determined by the same procedure except that the
emulsion was heated at 60°C for 30 min and then cooled under
running tap water for 15 min prior to centrifugation.
Dispersibility
The dispersibility of the SPI and commercial soya protein isolate at different
pH levels were measured according to the method of Karuna et al. (1991)
with minor modifications. Dispersibility was measured by placing 10 g of the
sample in a 100 mL stoppered measuring cylinder, adding distilled water to reach
a volume of 100 mL, stirring vigorously and allowing it to settle for three
hour. The volume of settled particles was subtracted from 100 and the difference
reported as percentage dispersibility. Three replicates of measurements were
carried out and the mean was taken.
Whipping Properties
The whipping properties of dispersion were investigated by the method described
by Lin et al. (1974) with slight modifications. Samples (10 g) were dispersed
in deionised water (200 mL) and the pH adjusted to 7.0 using 0.1 N, NaOH and
HCl. The suspension was homogenized at full speed Moulinex Blender for one minute.
Suspensions were then whipped on speed 7 in a Kenwood chef food mixer for 10
min, using the wire whip attachment. The resulting foam was immediately poured
into a litre measuring cylinder and the foam height and volume of liquid collecting
in the bottom of the cylinder, was measured at intervals.
The percentage foam expansion was calculated according to the method described by Lawhon et al. (1972). Foam volume as a percentage was calculated taking the foam volume at zero time as 100%. Leakage was calculated as volume of liquid collected over volume of liquid before whipping x100. The apparent viscosity was measured in triplicates immediately after whipping using a Brookfield viscometer (Spindle 3, 30 rev min-1). Triplicates whips were prepared for all experiments.
Foaming Capacity
Foam capacity and stability at different pH levels were determined according
to Aruna and Parakash (1993). One hundred milliliters of deionised water at
different pH were separately added to 3 g of defatted sesame protein isolate
and commercial soya protein isolate and the mixture homogenized at 400 rpm for
3 min in a Virtis homogenizer at 25°C and transferred to a measuring cylinder.
The volume of foam was measured and the volume increase was expressed as percentage
foam capacity. The foam stability was determined by measuring the decrease in
volume of foam as a function of time at 15, 30 and 60 min.
Statistical Analysis
The results were subjected to statistical analysis of variance (ANOVA),
using a Statistical Analysis System (SAS Institute, Inc. 2002) The Significant
of difference between means were determined by Duncan’s Multiple Rang Test
(DMRT), where p<0.05 was considered for significant difference.
RESULTS AND DISCUSSION
Solubility
Solubility curves obtained for both SPI and commercial soya protein isolate
were not significantly different (p<0.05), (Fig. 2). Both
isolates showed a minimum solubility in the region of between pH 4-5.5. (10-15%
for SPI and 10-15% for the commercial soya protein isolate) This is some how
higher than the reported isoelectric points (pI) of pH 4.4 and pH 4.8 of sesame
α-globulin, which was prepared by different method (Prakash and Nandi,
1978). Our report obtained was in agreement with previous reports when promine-D
(a protein component of soya bean) was isolated from soya protein (Kinsella,
1979; Wolf, 1970).
| Fig. 2: | Solubility of SPI and commercial soya protein isolate |
At pH values between 1-3 for SPI the solubility result obtained was 45-65% while for soya protein was 40-63%. Above the pI region which was observed to be above pH 5.5, the solubility for the SPI increased but <40% between pH 5.5-7, for soya protein within that region of pH it was observed showing a solubility profile <50%. The solubility of the SPI rose sharply to values above 40% for the SPI after pH 7 and for soya protein above 45%. We observed an increase in protein solubility below and above the pI regions for both proteins. It might be so because protein solubility in aqueous solutions is dependent on pH. At pH values above and below the isoelectric pH, proteins carry a net charge; electrostatic repulsion and ionic hydration promote solubililization of protein. For most proteins, minimum solubility occurs at their pI region, where electrostatic repulsion and ionic hydration are minimum and hydrophobic interaction between surface nonpolar patches is maximum (Basha and Pancholy, 1982). However, some proteins like whey protein have higher solubility characteristics at their isoelectric pH (pI = 4.8-5.2) because the exposed surfaces of these proteins contain a high ratio of hydrophilic to hydrophobic groups (Zhu and Damodaran, 1994). This could be the reason why the SPI showed higher solubility at the region of its pI or it can also be attributed to the possibility that the SPI underwent considerable denaturation during preparation. Solubility of our product was a good indication of functional potential which could be used in protein supplementation to foods as it did not show significant difference (p<0.05) of solubility characteristic with the commercial soya protein isolate.
Water-and Oil-Holding Capacity
Interactions of water and oil with proteins are very important in food systems
because of their effects on the flavor and texture of foods. Intrinsic factors
affecting water binding of food protein include amino acid composition, protein
conformation and surface polarity/hydrophobicity (Barbut, 1999). However food
processing methods have important impacts on the protein conformation and hydrophobicity.
During the processing of the SPI since the temperature was a little bit moderate
(45°C) it is possible that the proteins were likely denatured at that temperature
exposing more hydrophobic sites, which explained the high water retention of
the SPI then the commercial soya protein that was significantly different (p<0.05),
(Table 1). The decrease in oil absorption capacity could be
due to irreversible denaturation caused by the working temperature (45°C)
for the processing. Present findings were consistent with the results reported
by Bandyopadhyay and Ghosh (2002) in the preparation and characterization of
papain-modified sesame (Sesamum indicum L.) protein isolates. Though
the commercial soya protein showed a higher water holding capacity more than
the SPI but lower in fat absorption (Table 1) but were not
significantly different (p<0.05). SPI had a water-holding capacity of 302
g, similar to that reported by Prakash and Narasinga (1986) and within the range
of the commercial values of protein concentrates which was (300-320 g), as reported
by Lin and Zayas (1987) as SPI was 302 g water absorption while soya protein
was 289 g. It has been reported that the protein concentrate exhibits poor water-binding
capacity compared to that of the isolate (Bandyopadhyay and Ghosh, 2002). This
is likely due to the fact that the protein isolate has greater ability to swell,
dissociate and unfold, exposing additional binding sites, whereas the carbohydrate
and other components of the protein concentrate may impair it (Kinsella, 1979).
In food applications, the water-holding capacity or water-uptake capacity of
a protein is more important than hydration capacity. Water-holding capacity
refers to the ability of a protein matrix, such as protein particles, protein
gels, or muscle, to absorb and retain water against gravity. This water includes
bound water, hydrodynamic water, capillary water and physically entrapped water.
The physically entrapped water, however, is the largest fraction. It imparts
juiciness and tenderness in various foods (Scheraga et al., 1962).
| Table 1: | Water-and oil-holding capacity and bulk density of SPI and commercial soya protein isolate |
| aValues represent Means±Standard Deviation; means values with different letters in the same column are significant at level (p<0.05) | |
Sesame proteins showed a lower oil-holding capacity than soybean flour though not significantly different (p<0.05) but had a higher value than chickpea flour (Marina, 1986). Kinsella (1979) explained the mechanism of fat absorption as a physical entrapment of oil and several authors have related the oil absorption capacity to the nonpolar side chains of the protein as well as to the different conformational features of the proteins (Graham and Philips, 1976). Present results suggested that SPI had both good water-holding and good oil-holding capacity. Though slightly lower than the commercial soya protein isolate; it was seen as a good functional property that could be used in several food formulations that needs to utilize the above functional property.
Bulk Density
SPI produced less dense mixture than the commercial soya protein the difference
was significant (p<0.05, Table 1). Difference in particles size may account
for such difference in bulk density between the two proteins. SPI was passed
through a mesh sieve of 100 mm while the commercial soy protein isolate was
passed through a mesh sieve 80 mm according to the producer of that product.
Several authors have attributed solubility, hydrodynamic properties, hydrophobicity
and microstructure of proteins plays an important role in the bulk density of
any protein isolate (Anon et al., 2001; Kruse et al., 2001) but
some reports have also attributed it to the fact that since protein isolate
is rich in protein it will have low bulk density as there will be little or
small amount of carbohydrate that usually increases the bulk density of most
food product (Krause et al., 2002; Paulson and Tung, 1989). Present
result (0.169 g mL-1) was in consistent with Wang and Kinsella (1976)
they extracted protein from an alfalfa leaf by both chemical (NaCl) and water
but reported that the protein isolate from the water extraction has a lower
bulk density as the SPI.
Viscosity
From the results obtained (Fig. 3) it was observed that
SPI was able to form a low viscosity solution even at high temperature (60°C)
the viscosity was still low at pH 12 (210 cps) which is still within an acceptable
range (Schenz and Morr, 1996). As it has been reported that most of the food
items particularly infant foods prepared within the range of 40 and 60°C
as they are considered as moderate and high temperatures, respectively, will
exhibit a viscosity that will be good for infant formulation (Kulkarni et
al., 1991).
| Fig. 3: | Viscosity of SPI at different pH at 40 and 60°C |
At temperature 60°C SPI exhibited a gradual increase as the pH was increased; this was observed because solution conditions, such as pH, ionic strength and temperature affect viscosity of protein solutions. The viscosity of the SPI at 60°C was higher than SPI at 40°C and the difference was significant (p<0.05) because the high temperature helped the denaturation of the protein that increases the viscosity. It was reported that the viscosity of globular proteins generally decreases as the pH and temperature are decreased and increases when the two are increased (Modler and Emmons, 1977). The viscosity of SPI is actually dependent on the high protein concentrate in sesame seed. Present result was in accordance with the results of Dench et al. (1981) they studied selected functional properties of sesame (Sesamum indicum L.) flour and two protein isolates. The low viscosity of SPI may be useful in the development of high protein drinks, juice-based beverages and its supplementation in infant formulation without suffering the adverse consequences of high viscosity (Frokjaer, 1994; Sze-Tao and Sathe, 2000).
Emulsification Properties
Proteins are composed of charged amino-acids, non-charged polar amino acids
and nonpolar amino acids, which make proteins possible emulsifiers. The surfactant
possessing both hydrophilic and hydrophobic properties is able to interact with
both water and oil in food system (Chove et al., 2002). As shown in Fig.
4 SPI was a good emulsifier as the difference showed by the commercial soya
protein isolate was not significantly different (p<0.05) from the SPI. The
two proteins according to Fig. 4 have a higher emulsifier
capacity at pH 2 where 170 mL oil mg-1 was observed, but fell when
the pH was adjusted to 3, 4, 5 and 6 and increased again between pH 7-10 for
both of them. They only dropped slightly at pH 11 and 12. This is so because
protein exhibited low emulsification properties when expose to a very pH (11-14)
(Jung et al., 2005). Present results showed that proteins of both SPI
and commercial protein isolate exhibited good emulsion capacity at pH 2, neutral
and higher in an alkaline region. This could be attributed to the fact that
most proteins of plant origin can be solubilized below the pI region and above
that region as well. It could also be explained that the presence of salts as
during the production of the SPI both HCl and NaOH were use to control the pH
during the process sodium chloride which is salt developed that might contributed
to the denaturation of the protein that increased the emulsion properties of
SPI (Jung et al., 2005). It is possible that the pH values influenced
the emulsion capacity of the SPI which could also be possible for the commercial
soya protein isolate. Denaturation could improve the emulsifying properties
of the proteins due to increased hydrophobic surface and flexibility could also
be possible for the SPI since it was processed under a temperature that might
denature the protein. Present findings were in accordance with the reports of
Gonzalez-Perez et al. (2005). They studied the emulsion properties of
sunflower and reported similar results to SPI.
| Fig. 4: | Emulsification properties of SPI and Commercial soya protein isolate |
Dispersibility
The reconstitution property of SPI, in terms of dispersibility, was significantly
higher at pH 8, 9, 10 than the commercial soya protein (Fig. 5). But when the
pH was adjusted to pH 11 the commercial soya protein isolate exhibited higher
disperseability though the difference was not significant (p<0.05). However
as the pH was increased the dispersibility of commercial soya protein kept on
increasing and the difference was significantly different (p<0.05). Thus
suggesting that when pH is within the range of 8-10 for SPI it will display
a very good dispersibility which will be a very good functional property if
this protein isolate is to be supplemented to foods like weaning foods that
are supposed to disperse upon mixing so that they do not form a mat at the bottom
while preparing them with water for the consumption of babies as weaning food
was shown to have good dispersibility attribute at pH between 8-10 as reported
by Kulkarni et al. (1991). The dispersibility of our product supported
the results of Volkert and Klein (1979) who reported very similar range of pH
for the higher dispersibility of their product. It was however reported that
higher dispersibility enhances the emulsifying and foaming properties of proteins,
which was observed during the making of bread, macaroni and cookies (Kinsella,
1979). That could be the reason why SPI displayed good emulsification properties
as its dispersibility was significantly good for certain foods formulation or
protein supplementation.
Whipping Properties
Various whipping properties of the SPI products were compared with commercial
soya protein isolate. According to Table 2, it was observed
that for foam expansion, SPI had a higher foam expansion. For the foam volume
when the foam was left to stand for some time to investigate the of SPI and
commercial soya protein isolate at different pH as shown on top of each bar
stability of the foam, at 15 min the SPI was higher but at 30 min the commercial
soya protein isolate displayed a higher foam volume and when the time was increased
to 60 min the soya protein dropped but was still higher than the SPI. For the
foam leakage, the same time was also used and it was observed that the leakage
was like the foam volume where at 15 min, the SPI was higher but as the time
was increased, it kept on dropping and the soya protein was higher. According
to present results product could only have good whipping properties at a very
short time if that protein isolate was to be used in foods that should have
good whipping properties like in the preparation of cakes. Both foam volume
and leakage were measured as indices of foam stability. Increases in foam leakage
were not always accompanied by corresponding decrease in foam volume (Yasumatsu
et al., 1972) it can be due to adherence of the foam to the sides of
the vessel. Apparent viscosity of foam was taken as an estimate of foam strength.
| Fig. 5: | Mean scores±standard deviations which are not significant at level (p<0.05), for dispensability |
| Table 2: | Whipping properties of SPI and commercial soya protein isolate |
| aValues represent Means±Standard deviation; means values with different letter(s) in the same column are significant at level (p< 0.05) | |
When compared with SPI from defatted sesame flour, the flour gave the poorest expansion, stability and the weakest foam. The protein content of the defatted flour is approximately half that of the isolate. It might be the reason why the whipping properties of the flour are poor. So it is advisable to isolate the protein for better whipping quality than both the defatted sesame flour and the commercial soya protein isolate but the product should not be left to stand for a very long time (not more than 15 min.).
The two proteins formed soft foams which were self-supporting and which rapidly coarsened and broke up as it was left for a longer time to stand. The colour of the foam rapidly darkened from white and speckled cream to yellow for the SPI and brown for the commercial soya protein isolate.
Sesame protein isolate showed excellent foam expansion which exceeded that of soya protein though the difference was not significantly different (p<0.05). Stability in terms of foam volume was similar to the commercial soya protein isolate foams. Leakages for the SPI and the commercial soya protein isolate were similar as both of them reduced as they were left to stand for some time but SPI showed greater leakage than the commercial soya protein isolate. Lawhon et al. (1972) compared the foam expansion and viscosities of a number of oilseeds, present results corroborated with their report. SPI was prepared according to Fig. 1 and spray-dried, the isolate was obtained in the powdered form. Viscosity of sesame isolate was consistently lower than commercial soya protein isolate. The SPI in this study performed well in comparison with the commercial soya protein isolate in the case of foam expansion, foam volume and foam leakage at 15 min.
Foaming Capacity
The foam capacity (FC) of SPI in (Fig. 6) was pH-dependent
and was found to be lowest at pH within the region of it pI (4-5.5). The lowest
FC was attributed to the protein behavior at its isoelectric point. Beyond pH
5.5, FC significantly increased, especially at pH 10 and 11. The higher FC at
the above two pHs was likely due to the increased net charges on the protein,
which weakened the hydrophobic interactions but increased the flexibility of
the protein, this phenomenon (increase net charge) is applicable from pH 7-11
and decreases when the pH is increased to 12-14 for oil seed protein as reported
by Fligner and Mangino (1991). The decrease of foam capacity after pH 11 needs
to be investigated further. When the net charge increased up to pH 11, this
allowed the protein to diffuse more rapidly to the air-water interface to encapsulate
air particles and then enhance the foam formation (Aluko and Yada, 1995). The
decrease of foam after pH 11 could be attributed that the solution might have
more net charges that have been release during the process, that could not permit
more foam increase as it could have reach a maximum as the water was not increased
to give room for more net charges to be released to entrapped more water molecules
surround air droplets, this could be the reason for the drop in FC after pH
11, but this could be research up to ascertain the reason for such phenomenon.
The profile of FC against pH for the protein isolate was more or less similar
to that of its nitrogen solubility against pH. Present result on the FC supported
the results of Sosulski and Fleming, (1977) but was different from the result
of Fidantsi and Doxastakis (2002). SPI could be a good protein supplementation
to food that caters for good foamingpropertiesforexample salad dressing.
| Fig. 6: | Foaming capacity of SPI at different pH treatment |
CONCLUSIONS
In conclusion, total protein isolate from sesame seed was found to have very good functional properties that could be used in the formulation of different food as protein supplements. This is so because a commercially prepared soya protein isolate which was used to compare with the SPI did not show much significant difference (p<0.05). According to the findings the sesame protein isolate showed good functional properties that confirmed that the procedure used in its production was good enough and it could be used in protein supplementation in various food systems particularly for developing countries where protein deficiencies remain a major health problem for children.
ACKNOWLEDGMENTS
The authors wish to thank the Governments of Sierra Leone and People’s Republic of China for financially supporting this study.