Schistosomiasis, a disease of
poverty directly affecting approximately 240 million people worldwide with a
further 700 million at risk of infection, is regarded as one of the most
prevalent neglected tropical diseases (NTDs) (Sady et al., 2015). Schistosomes
are parasitic flatworms (flukes) that are transmitted to humans through direct
contact with freshwater that is contaminated with cercariae, the free-swimming
larval form of schistosomes which develop inside snails as intermediate hosts.
Schistosomal species vary with geographic region, but it is estimated that 92%
of cases of schistosomiasis are found in Africa, according to The World Health
Organisation (WHO). Species that are of widespread clinical importance include S. haematobium, S. japonicum and S. mansoni,
with S. haematobium causing
urogenital schistosomiasis and the latter two causing intestinal
schistosomiasis. There are other species affecting humans, such as S. intercalatum and S. mekongi, but
these have a more restricted distribution (Mahmoud, 2001). S. japonicum is the most geographically restricted out of the three
and is found mainly in China, Indonesia and the Philippines. S. haematobium and S. mansoni are found mainly in sub-Saharan Africa. S. mansoni is also endemic in South
America and the Caribbean and is the most widespread schistosome, affecting
over 50 countries worldwide, highlighting the importance of this species in
schistosomiasis control and prevention (Morgan et al., 2005). Sufferers of
schistosomiasis can be asymptomatic during early stages of infection, but
severe infections may cause Katayama fever, schistosomal myeloradiculopathy,
and pneumonitis. Longer-term infections can potentially lead to more serious
complications that are more threatening, such as the obstructions of
uropathies, hepatic fibrosis and granulomatous cerebral lesions (Kinkel et al.,
In accordance to WHO
recommendations, Kato-Katz preparations with coproscopy methods are routinely
used to identify the presence of eggs in stools when intestinal schistosomiasis
is suspected. Urine examinations are performed when urogenital schistosomiasis
is suspected (CDC, 2013), but it is also possible for a low number of eggs to
be present in the faeces during urogenital schistosomiasis infections for various
reasons. Although microscopy diagnosis is fast, cheap and easy, making it a
gold-standard, particularly in resource-poor countries, it does have its
drawbacks. Microscopy is heavily dependent on the skill of the observing
technician and even though it is highly specific, the sensitivity is suboptimal
(Meurs et al., 2015), especially during light infections where there may be a
low number of eggs. Additionally, sensitivity is further reduced when only a
single Kato-Katz smear is prepared from a specimen, which means that tests need
to be repeated in order to increase sensitivity. This would not be practical
and efficient in laboratories that handle a high number of patient samples, as
is common in resource-poor countries. On a wider scale, this has an impact on
control, prevention and treatment programmes, as low sensitivities will result
in a higher number of false-negative diagnoses resulting in an underestimation
of disease prevalence within populations. It is therefore important, for the
benefit of all levels of healthcare, to develop alternative methods for the
diagnosis of schistosomiasis that is both more sensitive, specific and
efficient than current routine microscopy methods.
The aim of this study was to develop
an alternative, highly sensitive and specific method for the diagnosis of Schistosoma mansoni using molecular,
biochemical and immunological methods. The results will be assessed by
comparing the results of the study’s methods to current tests, including the
Kato-Katz egg count. Methods used include PCR for DNA amplification, Bradford
assay for protein concentration analysis and ELISA techniques with recombinant
proteins for evaluation of the presence of S.
mansoni in sera.
Amplification and identification of DNA using PCR and gel electrophoresis
DNA sample preparation
Two coding sequences of two genes
from S. mansoni were cloned prior to
the study and labelled gene A and gene B. The recombinant plasmids containing
these genes were isolated for use in this study. The plasmids were kept on ice
in tubes labelled ‘A’ and ‘B’ to correspond with gene A and gene B respectively.
Two separate tubes for the PCR reaction were set up and labelled accordingly.
A PCR master mix was made up using
the reagents and volumes described in Table 1. A suitable primer was selected
for S. mansoni detection and included
in the master mix (Table 1). Final volumes were determined by scaling up the
reaction volumes enough so that there was enough for each reaction, accurate
pipetting, plus excess as spare. The individual reaction volume was therefore
multiplied by six for the master mix volume. 24 ?l of the master mix was
pipetted into each reaction tube, followed by 1 ?l of plasmid A and B in each
corresponding tube, making the total of each reaction 25 ?l. Incubations and running
of the PCR was then followed according to the methods of Espírito-Santo et al.,
1: Components of the PCR master mix.
Master Mix Volume
Nuclease-free distilled water
10X Buffer (with loading dye) with MgCl2
dNTP (2.5 mM)
Primer 1 or 3 (100 ?M)
Primer 2 or 4 (100 ?M)
Taq DNA polymerase (5 U/?l)
Agarose Gel Electrophoresis
A 2% agarose gel was prepared and
assembled according to the manufacturer’s instructions to make up 100ml of gel.
The gel was then set up and ran according to the protocol designed by Lee et
Estimation of protein concentration using Bradford assay
The protocol for the Bradford assay
was adapted from Ernst and Zor (2010) with some amendments. A bovine serum
albumin (BSA) standard assay was prepared by setting up 11 eppendorf tubes labelled
S0-10 according to Table 2, with S0 being used as a blank. The total volume of
protein solution and water equated to exactly 50 ?L in each solution. Following
intermediate steps and incubation times according to the protocol, the standard
absorbances were read using a spectrophotometer at 595nm only, instead of at
450nm as well (as in Ernst and Zor’s protocol), in triplicates and recorded.
2: Volumes of water and undiluted BSA
used in 10 serial dilutions for a standard assay.
The concentration of the undiluted
protein used was 2mg/mL
Volume of undiluted BSA (?L)
Volume of water (?L)
Protein concentration per assay
In order to ensure that absorbances
of the unknown protein concentrations were within the linear range of the
standard curve, proteins A and B were diluted. The volumes used for the
dilutions were performed according to Table 3, and labelled A1-5, B1-5. The
same steps were repeated as for the standard solution absorbances. Analysis and
representations of the recorded data was then performed using Microsoft Excel.
3: Volumes of water and Schistosoma mansoni recombinant proteins
Protein A1-5 are different dilutions
of protein A. Protein B1-5 are different dilutions of protein B.
Volume of undiluted protein sample
Volume of water (?L)
Detection of infection presence using recombinant proteins and ELISA
The protocol for this ELISA was
designed according to Smith et al. (2012) with some amendments, particularly
with the samples used. Samples in this study were collected from a battalion of
UK military participants where a known portion of individuals were exposed to S. mansoni in Uganda and tested positive
for current or past infection. Others had never been to an area with a S. mansoni endemic and had also tested
negative for current or past infections. Samples were screened to confirm
infections by other methods, such as Kato-Katz method, CCA test and serum
antibody test, in order to use as a reference for comparing the effectiveness
of the recombinant proteins A and B in the ELISA test. Wells were prepared by
attaching expressed proteins to detect serum antibodies for proteins A and B.
After preparation, incubation and pipetting of the solutions in the wells, the
plates were read using a plate spectrophotometer with a 450nm filter, and the
optical density readings were recorded. The results were recorded as either a
positive or negative infection.
PCR and Gel Electrophoresis
1. Agarose gel electrophoresis (2%
agarose) image of two recombinant Schistosoma
mansoni gene inserts amplified by PCR from plasmids.
Lane 1 shows a DNA ladder. Lane 2
shows gene B. Lane 4 shows gene A. Lanes 3 and 5 are empty. Annotations show
the approximate size of the DNA fragments.
Figure 1 shows the PCR products of
genes A and B in the agarose gel electrophoresis which has produced clear,
visible bands with a high resolution. Using the DNA ladder for reference, it
can be seen that gene A and gene B have sizes of 300 bp and 600 bp
Spectrophotometry and Bradford assay
The mean absorbances of the
triplicate absorbances for each protein concentration is shown in Table 4 with
the standard deviations calculated, which were low for all samples. The
absorbances increase as the concentration of BSA increases which applies to all
4. Mean spectrophotometer absorbances
of standard BSA protein concentrations at 595nm.
BSA Conc. (µg/ML)
Mean absorbance (595nm)
2. Graph showing BSA protein
concentrations versus mean absorbance at 595nm.
Mean data points are shown on the
graph together with the equation of the linear trend line and the R2
value. Error bars for each data point have also been included using the
standard deviation of the observed absorbances.
There was a positive correlation
between the absorbances and protein concentrations in the standard calibration,
as shown in Table 4 and Figure 2. An increase of protein concentration resulted
in a proportional increase of protein absorbance at 595nm in all standard
samples. The standard deviations for all the samples, except the blanks, ranged
between 0.0645 and 0.0670, as represented by the error bars in Figure 2. The
linear trend line equation was also calculated to be y=0.0008x + 0.1461 for
which when rearranged for (x), gives x=(y-0.1461)/0.0008.
5. Mean spectrophotometer absorbances
of diluted unknown recombinant protein concentrations at 595nm.
Mean absorbance (595nm)
Using the rearranged linear
equation, it was possible to calculate the unknown protein concentrations of
recombinant proteins A and B using the mean absorbances from Table 5. A
suitable dilution was chosen for each protein to be used in the calculation.
Samples A4 and B4 were chosen since both absorbances were under 1.00 meaning
there was less chance for the absorbance readings to be distorted by the
saturation of the sample. The absorbances were also not low enough so that
small errors and differences in the readings would have less impact when
calculating the concentration according to the dilution factor. After applying
the dilution factor in the calculation, the final undiluted concentration of
sample A4 was found to be 59.3 mg/15 mL whereas sample B4 was 135.5 mg/15 mL.
ELISA results against other tests
Both proteins A and B successfully
detected positive and negative results by using ELISA, however, false positives
and negatives were identified within these results when compared against
Kato-Katz faecal counts. The serum antibody did not produce any false negatives
but did have some false positive results whereas the CCA test did not produce
any false positives or false negatives at all (Table 6).
6. Positive and negative results
acquired using a range of diagnostic methods from 40 participants.
Protein A and B refer to the
recombinant proteins used in the ELISA test.
Serum Antibody Test
Protein A (ELISA)
Protein B (ELISA)
3.3.2 Statistical Analysis
The prevalence, sensitivity,
specificity and negative/positive predictive values were calculated from the
results recorded from the ELISA with a 95% confidence interval (Table 7). The
prevalence was the same for all tests, but the other statistics were varied.
Overall, the CCA test showed the highest sensitivity and specificities with
high positive and negative predictive values, all at 100% for estimated values.
There was a reduction in sensitivity for the ELISA tests compared to non-ELISA
tests, particularly when using protein B which had a 46.7% reduction in
sensitivity compared to the 100% sensitivities in the non-ELISA tests. Protein
A did produce highly sensitive results, but it is still not as sensitive as the
CCA test or serum antibody test. Regarding specificities, protein B had a
slightly higher (+4%) specificity than protein A and the serum antibody test, which
were both the same, but all were lower in specificities than the CCA test. The
positive and negative predictive values for the ELISA were lower than the other
7. Prevalence, sensitivity, specificity
and negative/positive predictive values for two recombinant proteins A and B
compared to CCA and serum antibody test for ELISA.
Data has been expressed in decimal
percentages. Lower and upper limits have been expressed with a 95% confidence
Serum Antibody Test
Protein A (ELISA)
Protein B (ELISA)
Using the results of the gel
electrophoresis and existing reference information, it can be deduced that the
size of the S. mansoni genes A and B
are the same size given in the reference data, thus possibly confirming the
identity of the genes. The existing data states that the protein products of
the genes were 100 and 200 amino acids in length for proteins A and B
respectively. Since each amino acid is coded by a codon (3bp in length), the
size of the DNA fragments should be 300bp for A and 600bp for B. However, the
results obtained from this PCR may not be conclusive evidence that the genes
are confirmed to be the same as the reference data. This is due to the lack of
a positive and negative controls in the method. A suitable positive control,
such as a known plasmid, should have been included in the electrophoresis in
order to affirm that the primer is working. Furthermore, to decrease the cost
of the procedure, dilutions of the reagents and primers that were too low to
pipette accurately should have been performed instead of multiplying the total
volume. This will improve the cost-efficiency of the PCR, but ultimately does
not affect the final results.
The low standard deviations of the
triplicates indicated The R2 value for the standard calibration
curve was 0.9683 indicating that the data has a strong correlation and nearly
all the data points fit the model. This is valid evidence to suggest that the
absorptions and concentration increase proportionally which can be seen by the
positive correlation in Figure 2, consistent with the Beer-Lambert law (Dean,
2014). The correlation coefficient (square root) of the R-squared value was also
found to be 0.984, indicating a positive relationship between the two variants.
This meant that the protein concentrations extrapolated from the standard curve
were done confidently.
For the ELISA test, the accuracy of
the results and the methods used are questionable. This is reflected in the
sensitivities and specificities produced by the ELISA results as the use of the
proteins produced many false positive and false negative results, as also shown
in the positive and predictive values. However, the use of Kato-Katz as a
reference point to compare against the sensitivities of the ELISA may not be
ideal. Since the sensitivities of the Kato-Katz smear method is limited (Meurs
et al., 2015), it may not be suitable to use it as a control in order to assess
the sensitivities of other tests, particularly in light infections where eggs
may not be present.
The use of ELISA is undoubtedly
useful in diagnosis, given how immunogenic schistosome infections are, but the
lack of sensitivity would mean that two or more assays should be run in
parallel in order to improve sensitivity. The question of logistics is also
raised with these techniques. Use of PCR for DNA amplification and ELISA allows
for a high number of samples to be tested in one go, making it relatively
efficient, but the high setup costs of the equipment and resources needed, such
as a reliable electrical source, may not be ideal in resource-poor environments
or in-field (Lodh et al., 2013). Currently, alternative methods are both more
sensitive and more efficient than those used in this study. Based on the
results from this study, the CCA test is a much more sensitive and specific
test that is more accurate in the diagnosis of S. mansoni. Furthermore, it is much more portable than other tests.
However, it may not be useful in detecting other species of schistosomes which
do not release circulating cathodic antigens (CCA), such as S. haematobium. The use of serological
tests to detect antibodies also has further drawbacks including the ability to
differentiate between past and present infections as well as the inability to
measure the intensity of the infection (Gomes, Enk and Rabello, 2014).
Currently, there is ongoing research
into alternative coproscopy methods which can potentially replace the Kato-Katz
smear as a gold-standard, which would improve sensitivity testing. Such methods
utilise sedimentation, centrifugation, fluctuation and miracidium hatching
which are more sensitive, but is more laborious as a result (Gomes, Enk and
Rabello, 2014). There are other non-microscopy based methods being developed
for schistosome detection which are similar to those in this study. In a
literature review by Gomes et al. (2014), which compared the sensitivities and
specificities of current advancements in schistosomiasis diagnosis, highlighted
the promising loop mediated isothermal amplification (LAMP) method which is a
low-tech PCR. The method showed high sensitivities of 96.7% and specificities
of 100% and could be used as an efficient diagnostic tool in resource-poor
countries since the amplification of DNA only requires a simple water bath and
a dye for visualisation.
Despite the clear and significant
potential for PCR, Bradford assay and ELISA as a sensitive alternative to
microscopy methods, microscopy still remains the current gold-standard despite
its suboptimal sensitivity. A refined method and further work into this
research area to further improve sensitivity is required in order to
appropriately assess the validity and application of these techniques.