J Infect Chemother (2003) 9:25–29 © Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases 2003
DOI 10.1007/s10156-002-0226-2
ORIGINAL ARTICLE
Hisashi Yamaura · Kunioki Araki · Ken Kikuchi
Ichiro Itoda · Kyoichi Totsuka · Takatoshi Kobayakawa
Evaluation of dot-ELISA for serological diagnosis of amebiasis
Received: July 8, 2002 / Accepted: December 7, 2002
Abstract We improved the dot enzyme-linked immuno-
sorbent assay (dot-ELISA) reported by Itoh and Sato,
and assessed the usefulness of this test for the diagnosis
of amebiasis. The sensitivity of dot-ELISA was compared
with that of plate ELISA, the indirect hemagglutination
test (IHA), and the indirect fluorescent antibody test (IFA)
for the diagnosis of amebiasis. Of 37 serum samples from
patients with documented amebiasis, 36 (97.3%) were posi-
tive by dot-ELISA. There was consistency among the re-
sults of dot-ELISA, plate ELISA, and IFA, although the
positive rate of IHA was lower than that of the others
(78.4%; 29 of 37 cases were positive). The specificities of
dot-ELISA and plate ELISA were assessed using a total
of 68 sera, collected from 38 patients infected with seven
different parasites other than Entamoeba histolytica, 10
patients showing diarrhea or liver abscess without para-
sitic infection, and 20 healthy individuals. The two assays
showed no false-positive results. There were no differences
in sensitivity and specificity between dot-ELISA and plate
ELISA. However, the dot-ELISA technique seems to be
more feasible for clinical application than plate ELISA
techniques, because the assay does not require any specific
equipment.
A summary of this paper was presented at the 74th General Meeting of
the Japanese Association for Infectious Diseases (Fukuoka, April
2000).
H. Yamaura (*) · K. Kikuchi · I. Itoda · K. Totsuka
Department of Infectious Diseases, Tokyo Women’s Medical
University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan
Tel. / Fax ⫹81-3-3358-8995
e-mail: yamaura@research.twmu.ac.jp
K. Araki
Department of Microbiology, National Institute of Public Health,
Tokyo, Japan
T. Kobayakawa
Department of International Affairs and Tropical Medicine, Tokyo
Women’s Medical University, Tokyo, Japan
Key words Dot-ELISA · Amebiasis · Entamoeba histoly-
tica · Serological diagnosis
Introduction
In recent years, Entamoeba histolytica infection (amebiasis)
among homosexual men has been increasing in
Japan.1,2 Various serological tests have been used for the
immunodiagnosis of amebiasis.1–3 The indirect hemaggluti-
nation test (IHA) and indirect fluorescent antibody test
(IFA) have been used for the serological diagnosis of intes-
tinal amebiasis and amebic liver abscess at Tokyo Women’s
Medical University. Although IHA was a simple and prac-
tical diagnostic method for amebiasis, the manufacture of
this kit was stopped in 1999. Therefore, we attempted to
develop a simple and reliable diagnostic method to replace
the IHA kit.
The dot enzyme-linked immunosorbent assay (dot-
ELISA) technique has several advantages. The results can
be read with the naked eye and the antigen-dotted nitrocel-
lulose membrane strips can be stored for 3 years at 4°C.
Because of these advantages, dot-ELISA has been used for
the serological diagnosis of parasitic infections.4–7 The use-
fulness of dot-ELISA for the immunodiagnosis of helmin-
thiasis has been reported in Japan.8–10 In addition, the
dot-ELISA has been performed in a commercial medical
laboratory since 1999 for the screening of parasitic infec-
tions. However, dot-ELISA has never been assessed in
detail in Japan for the diagnosis of amebiasis, although
this method has been used to detect amebiasis in other
countries.11,12
The present study aimed to evaluate the usefulness of
dot-ELISA as a diagnostic method for amebiasis. To assess
the sensitivity for amebiasis, the results of dot-ELISA were
compared with those of IFA, plate ELISA, and IHA. More-
over, the cross-reactivity of anti-E. histolytica antibody in
dot-ELISA was examined using a panel of seven different
helminthic antigens. At the same time, cross-reactivity
between the E. histolytica antigen and sera from patients
26
infected with parasites other than E. histolytica was exam-
ined to assess the specificity of dot-ELISA.
Subjects, materials, and methods
Serum samples
Serum samples were collected from 75 patients infected
with parasites, 5 patients each with diarrhea and liver
abscess but without parasitic disease, and 20 healthy volun-
teers. The number of serum samples for each parasitic in-
fection was as follows; 37 for E. histolytica (9 intestinal
amebiasis and 28 amebic liver abscess), 10 for Fasciola
sp., 6 for Giardia lamblia, 6 for Entamoeba coli, 2 for
Endolimax nana, 4 for Paragonimus westermani, 4 for
Paragonimus miyazakii, and 6 for Toxocara species. These
sera were kept at ⫺40°C at the Central Clinical Laboratory
of Tokyo Women’s Medical University or the National
Institute of Public Health until use.
Amebic liver abscess was diagnosed by clinical features
and plate ELISA, and intestinal amebiasis was determined
by these methods and also by the demonstration of tro-
phozoites in fecal samples. The diagnoses of protozoan
infections other than E. histolytica in these patients were
previously confirmed by fecal examination, and those of
helminthic infections were confirmed by plate ELISA and/
or Ouchterlony’s immunodiffusion test.
Antigens
Recombinant Toxocara canis antigen, phosphate-buffered
saline (PBS)-extracted antigens of E. histolytica, P.
miyazakii, P. westermani, Dirofilaria immitis, Anisakis
simplex, Fasciola hepatica, and Spirometra erinacei were
used. The amebic antigen was prepared from E. histolytica
(HM-1: IMSS strain) axenically grown in BI-S-33 medium.13
The protein content of each antigen was measured by a Bio-
Rad protein assay (Bio-Rad, Hercules, CA, USA). Each
antigen was kept at ⫺40°C until use.
Procedure of dot-ELISA
Dot-ELISA was performed as reported by Itoh and Sato,8
with modifications. A nitrocellulose membrane (pore size,
0.45 µm, 3 ⫻ 3 mm with a printed grid; Toyo Roshi, Tokyo,
Japan) was used. The protein content of recombinant T.
canis antigen was adjusted to 0.16 µg/ml and those of the
other antigens were adjusted to 2 mg/ml with PBS. A 0.5-µl
aliquot of each antigen, or 1 : 400 diluted human serum, as a
positive control, was dotted onto nitrocellulose membrane
strips and dried for 30 min at 37°C. The strips were dipped
in a blocking buffer (0.15 M PBS containing 0.05% Tween
20 and 1% bovine serum albumin; BSA/T) for 30 min at
room temperature, then dried again and stored at 4°C
until use. Before examination, the strips were wetted with
BSA/T and put in a plastic tray. Then, 80 µl of 1 : 200 diluted
serum was pipetted onto each strip and the strips were
incubated for 40 min at 37°C. The strips were then washed
three times with a washing buffer (0.05% Tween 20 in
0.15 M PBS; PBS/T). Then 80 µl of peroxidase-conjugated
anti-human immunoglobulin G rabbit serum (ICN Pharma-
ceuticals, Aurora, OH, USA) diluted 1 : 750 was added, and
the strips were incubated for 40 min at 37°C. After a wash-
ing with PBS/T, the strips were immersed in substrate solu-
tion (0.04% 4-chloro-1-naphthol, 0.05% H2O2, and 16%
ethanol in 1/15 M phosphate buffer, pH 7.4) (Sigma Chemi-
cal, St. Louis, MO, USA) for 7 min. To stop color develop-
ment, the strips were immediately washed in distilled water
and dried. The results can be read with the naked eye and
purplish blue spots indicate a positive reaction.
Procedure of IHA, IFA, and plate ELISA
IHA and IFA were performed using an E. histolytica HA
kit (Japan Lyophilization Laboratory, Tokyo, Japan) and
an Ameba-spot IF kit (bioMerieux Japan, Tokyo, Japan),
respectively. The cutoff point was set at a serum dilution of
1 : 80 for IHA and 1 : 100 for IFA. Plate ELISA was per-
formed as reported by Yamasaki et al.14 The test serum was
diluted 1 : 200 before assay. Peroxidase-conjugated anti-
human immunoglobulin G rabbit serum (ICN Pharmace-
uticals) and 2,2⬘-azino-bis (3-ethylbenthiazoline-6-sulfonic
acid) (Sigma Chemical) were used as the conjugate and
substrate reagents, respectively. Absorbance at 415 nm was
measured by an ELISA reader (Bio-Rad; model 550). The
cutoff point was set at three times the optical density (OD)
of the negative pooled serum sample from healthy individu-
als. This value corresponds to greater than the mean plus 4
OD of the value for healthy individuals.
Results
Sensitivity of the four serological tests
The sensitivity of the four serological tests to E. histolytica
antigen was compared using sera from patients with ame-
biasis and healthy individuals, as shown in Table 1. Of 37
sera from patients with amebiasis, 36 (97.3%) were positive
to E. histolytica antigen by dot-ELISA. All the sera from
the 28 patients with amebic liver abscess were positive, but
1 (11.1%) of the 9 sera from patients with intestinal amebia-
sis was negative. The positive rate of IFA and plate ELISA
was also 97.3%. On IHA, 22 (78.6%) of the 28 amebic liver
abscess samples and 7 (77.8%) of the 9 intestinal amebiasis
samples were positive. Thus, the sensitivity of IHA was
lower than that of dot-ELISA and the other tests. On the
other hand, none of the sera from the 20 healthy individuals
showed any false-positive results on any of the four sero-
logical tests.
Amebiasis-positive cases by dot-ELISA
In 36 sera from patients with amebiasis, bluish-purple spots
were observed against diluted human serum as a positive
 27

Table 1. Comparison of sensitivity among four serological tests

Clinical diagnosis No. of cases No. of antibody-positive cases

Dot-ELISA IFA Plate ELISA IHA

Amebic liver abscess 28 28 (100%) 28 (100%) 28 (100%) 22 (78.6%)

Intestinal amebiasis 9 8 (88.9%) 8 (88.9%) 8 (88.9%) 7 (77.8%)
Total 37 36 (97.3%) 36 (97.3%) 36 (97.3%) 29 (78.4%)
Healthy individuals 20 0 0 0 0
Dot-ELISA, Dot enzyme-linked immunosorbent assay; IFA, indirect fluorescent antibody test;
IHA, indirect hemagglutination test

Fig. 2A,B. Results of dot-ELISA for serum samples showing negative

reaction to E. histolytica antigen on indirect hemagglutination test
Fig. 1A,B. Results of dot-enzyme-linked immunosorbent assay (IHA), A The positions of control and antigens (same as in Fig. 1.).
(ELISA) for serum samples form patients with intestinal amebiasis and B NC, Negative control; PC, positive control for E. histolytica; 1–6,
liver abscess. A the positions of control and antigens. Control, Normal IHA-negative amebic liver abscess samples
human serum; P.m., Paragonimus miyazakii; Fas., Fasciola sp.; Ani.,
Anisakis simplex; D.i., Dirofilaria immitis; P.w., Paragonimus
westermani; S.e., Spirometora erinacei (plerocercoid); r-T.c., recom-
binant Toxocara canis; E.h., Entamoeba histolytica. B NC, Negative
control; 1–5, intestinal amebiasis samples; 6–11, amebic liver abscess
samples

control and E. histolytica antigen. However, these sera

showed no reaction with the antigens of seven helminthic
species. As shown in Fig. 1, there was no significant differ-
ence in the intensity of spots between intestinal amebiasis
and amebic liver abscess. In six cases judged as negative
by IHA, clear spots against E. histolytica antigen were
recognized (Fig. 2).

Specificity of dot-ELISA
Nonspecific reactions with the E. histolytica antigen were
examined using sera from patients infected with parasites
other than E. histolytica and patients with nonparasitic
infections. In the sera from ten patients with fascioliasis,
although cross-reactivity with Paragonimus sp. and/or
nematode antigens was observed in several sera, no positive
spot against E. histolytica antigen was recognized (Fig. 3).
All the sera from four patients with Paragonimus
westermani infection, four with Paragonimus miyazaki
infection, and six with toxocariasis were negative for E.
histolytica antigen. In the sera from six patients each with G.
lamblia and nonpathogenic E. coli, no spots were seen for
E. histolytica and the other seven antigens (Fig. 4). No
cross-reactivity with the antigens mentioned above was
observed in the sera from the two patients with Endolimax
Fig. 3A,B. Results of dot-ELISA for serum samples from patients with
fascioliasis. A The positions of control and antigens (same as in Fig. 1.).
B NC, Negative control; PC, positive control for E. histolytica; 1–9;
fascioliasis samples
nana infection and the five patients each with diarrhea and
liver abscess but no parasitic infections.
Discussion
Because the symptoms in patients with amebiasis some-
times progress rapidly, early diagnosis is imperative. From
this viewpoint, several serological tests have been used to
facilitate the diagnosis. The present study was undertaken
to examine the usefulness of dot-ELISA, a technique con-
28
Fig. 4A,B. Results of dot-ELISA for serum samples from patients with
E. coli infection and giardiasis. A The positions of control and antigens
(same as in Fig. 1.). B NC, Negative control; PC, positive control for
E. histolytica; 1–6, E. coli infection samples; 7–10, giardiasis samples
ventionally used for the serological diagnosis of helminthi-
asis, for the diagnosis of amebiasis. We modified the
method of dot-ELISA reported by Itoh and Sato8 so that
the results can be obtained within 2 h if a stock antigen-
dotted nitrocellulose membrane strip is used. Moreover, we
confirmed that the same reagents, except for the substrate,
used in plate ELISA could be used in dot-ELISA. It was
therefore easy to perform dot-ELISA and to compare the
results with those of plate ELISA. Assessment of the use-
fulness of serological diagnostic techniques requires the
validation of sensitivity and specificity. We therefore com-
pared the sensitivity of the modified dot-ELISA technique
with that of IFA, plate ELISA, and IHA. The anti-
E. histolytica antibody detection rate was 97.3% by dot-
ELISA, and it was the same as the rate by IFA and plate
ELISA. When dot-ELISA and IHA were compared, the
antibody detection rate was lower in IHA (78.4%), thereby
corroborating a previous report that dot-ELISA is more
sensitive than IHA.11
Among cases of amebiasis, the antibody titers detected
by IFA and IHA have been reported to be higher in cases of
amebic liver abscess than in cases of intestinal amebiasis.15,16
However, the intensity of the spots against E. histolytica
antigen did not differ markedly between intestinal amebia-
sis and amebic liver abscess. Furthermore, there were some
cases in which the antibody titer measured by plate ELISA
did not correlate with the intensity of the spots obtained
by dot-ELISA. This discrepancy does not seem to be a
problem, as a similar finding has also been reported in
paragonimiasis.
Diarrhea in overseas travelers is suspected of reflecting
not only amebiasis but also protozoiasis such as Giardia
lamblia, Cryptosporidium parvum, and Cyclospora
cayetanensis infections. In cases of liver abscess in which
antibiotics are ineffective and parasitic disease is suggested,
one should consider the possibility of amebic liver abscess
or fascioliasis.17 The presence or absence of eosinophilia is
useful in distinguishing amebic liver abscess from fasciolia-
sis. However, double infection with E. histolytica and other
parasites may occur in patients from developing countries,
and in such cases, it is impossible to rule out amebiasis even
when eosinophilia is present. To assess the specificity of
dot-ELISA, we checked for cross-reactivity by testing 37
serum samples from patients with amebiasis against E.
histolytica and seven helminthic antigens. Positive spots
were recognized only against E. histolytica antigen. Because
no cross-reaction with G. lamblia infection, fascioliasis, or
nonparasitic diseases was recognized, we confirmed that the
specificity of dot-ELISA is high. This test, therefore, seems
to be useful for differential diagnosis between amebiasis
and other parasitic diseases.
In the present study, the sensitivity and specificity of dot-
ELISA were the same as those of plate ELISA, but dot-
ELISA is less costly than plate ELISA, because only one
nitrocellulose membrane is needed for each sample. There-
fore, dot-ELISA seems to be more suitable than plate
ELISA for the screening of amebiasis, which often tests
samples from a small number of patients. An attempt to
measure antibody titer by dot-ELISA using various serum
dilutions has been reported in another country.11 Plate
ELISA is more reliable than dot-ELISA to examine anti-
body titers, although dot-ELISA is simpler to use for the
qualitative analysis of antibodies. If the nitrocellulose mem-
brane strips adsorbed with E. histolytica and F. hepatica
antigens are stored in a refrigerator, they can be used for
differential diagnosis immediately. Yamasaki and Araki18
tested the validity of helminthic antigen-dotted nitrocellu-
lose membrane strips and found that they were stable for
3 years at 4°C. Antigen-dotted nitrocellulose membranes
stored at room temperature were also stable for up to 10
days, although some investigators have reported that the
membranes were valid for 3 months at room temperature or
37°C.11 A desirable shelf life for E. histolytica antigen-
dotted membrane strips would be about 3 years at 4°C.
Therefore, further studies on the stability of these nitrocel-
lulose membrane strips are necessary. If these membrane
strips are widely available to hospitals, universities, and
research institutes, reliable serological diagnosis of amebia-
sis will be possible anywhere. Because the prevalence of
amebiasis has been increasing as a result of the recent in-
crease in the numbers of male homosexuals and foreign
travelers, dot-ELISA will become more useful.
Dot-ELISA has many advantages, including no
requirement for special equipment, simplicity of the proce-
dure, low cost, possibility to judge with the naked eye, and
high sensitivity and specificity. The results of the present
study demonstrate that dot-ELISA is as a useful serological
diagnostic method for amebiasis.
Acknowledgments The authors are indebted to Professor Tsutomu
Takeuchi and Dr. Seiki Kobayashi, Department of Tropical Medicine
and Parasitology, Keio University School of Medicine, for supplying
E. histolytica antigen, and to Dr. Hiroshi Yamasaki, Department of
Parasitology, Juntendo University School of Medicine, for supplying
recombinant Toxocara canis antigen for this study.
References
1. Takeuchi T, Kobayashi S, Asami K, Yamaguchi N. Correlation of
positive syphilis serology with invasive amebiasis in Japan. Am J
Trop Med Hyg 1987;36:321–4.

2. Takeuchi T, Okuzawa E, Nozaki T, Kobayashi S, Mizokami M,
Minoshima N, et al. High seropositivity of Japanese homosexual
men for amebic infection. J Infect Dis 1989;159:808.
3. Yamaura H, Shirasaka R, Sato J, Odagiri Y, Iseki M, Kimata I,
et al. Application of indirect hemagglutination test using glutaral-
dehyde-fixed chicken red blood cells to serological diagnosis of
amebiasis. Jpn J Parasitol 1990;39:468–74.
4. Pappas MG, Hajkowski R, Hockmeyer WT. Dot enzyme-linked
imunosorbent assay (dot-ELISA): a micro technique for the rapid
diagnosis of visceral leishmaniasis. J Immunol Methods
1983;64:205–14.
5. Araujo FG. A method for demonstration of antibodies to Trypa-
nosoma cruzi by using antigen-coated nitrocellulose paper strips.
Am J Trop Med Hyg 1985;34:242–5.
6. Zimmerman GL, Clark CRB. Separation of parasite antigen by
molecular exclusion, anion exchange, and chromatofocusing utiliz-
ing FPLC protein fractionation systems. Vet Parasitol 1986;20:217–
28.
7. Boctor FN, Stek MJ Jr, Peter JB, Kamal R. Simplification and
standardization of dot-ELISA for human Schistosomiasis mansoni.
J Parasitol 1987;73:589–92.
8. Itoh M, Sato S. Multi-dot enzyme-linked immunosorbent assay for
serodiagnosis of trematodiasis. Southeast Asian J Trop Med Public
Health 1990;21:471–4.
9. Okamoto M, Miyake Y, Shouji S, Fujikawa T, Tamaki A, Takeda
Z, et al. A case of severe Paragonimiasis miyazakii with lung and
skin lesions showing massive egg production in sputum and feces.
Jpn J Parasitol 1993;42:429–33.
29
10. Maruyama H, Noda S, Nawa Y. Emerging problem of parasitic
diseases in southern Kyushu, Japan. Jpn J Parasitol 1996;45:192–
200.
11. Kumar S, Band AH, Samantaray JC, Dang N, Talwar GP. A dot
enzyme-linked immunosorbent assay for detection of antibodies
against Entamoeba histolytica. J Immunol Methods 1985;83:125–
33.
12. Kanwar JR, Vinayak VK. Comparative efficacy of plate ELISA
and dot- ELISA to detect antiamoebic antibodies in clinical
patients. Trop Geogr Med 1991;43:261–5.
13. Diamond LS, Harlow DR, Cunnick CC. A new medium for the
axenic cultivation of Entamoeba histolitica and other Entamoeba.
Trans R Soc Trop Med Hyg 1978;72:431–2.
14. Yamasaki H, Araki K, Lim PKC, Zasmy N, Mak JW, Taib R, et al.
Development of a highly specific recombinant Toxocara canis
second-stage larva excretory-secretory antigen for immunodiagno-
sis of human toxocariasis. J Clin Microbiol 2000;38:1409–13.
15. Ambroise-Thomas P, Truong TK. Fluorescent antibody test in
amebiasis. Am J Trop Med Hyg 1972;21:907–12.
16. Parija SC, Kasinathan S, Sambasive Rano R. Rapid indirect
haemagglutination (rapid IHA) using sensitized chick cells for
serodiagnosis of amoebiasis at primary health center level. J Trop
Med Hyg 1989;92:221–6.
17. Kodama K. Human fascioliasis: comparison of a fasciolicidal effect
of bithionol and praziquantel (in Japanese). J Jpn Assoc Inf Dis
1997;71:1162–7.
18. Yamasaki H, Araki K. Immunodiagnosis of parasitic infection (in
Japanese). Rinsho Kensa (J Med Technol) 1999;13:1617–24.