Molecular and Biochemical Parasitology, 40 (1990) 129-136 Elsevier
129
MOLBIO 01316
A rapid D N A assay for the species-specific detection and quantification of Brugia in blood samples Catherine B. Poole 1'* and Steven A. Williams 1'2 XDepartment of Biological Sciences, Smith College, Northampton, MA, U.S.A. and 2Molecular and Cellular Biology Program, University of Massachusetts, Amherst, MA, U.S.A. (Received 1 September 1989; accepted 4 December 1989)
This report describes a new assay for detecting filarial parasites of the genus Brugia in blood samples using labeled DNA probes. The sequences of these DNA probes are based on the HhaI repeat DNA family found in the genus Brugia. These DNA probes are species-specific and can detect the DNA from a single microfilaria in hybridization assays. To adapt this test for use on blood samples collected in the field, complex steps to separate microfilariae from blood cells and to purify parasite DNA were eliminated. We found that the most effective method was to filter blood samples through 5.0 ~.m pore nitrocellulose membranes, lyse the microfilariae on the membranes by proteinase K digestion, denature the parasite DNA with sodium hydroxide, and hybridize with the DNA probe. With this method, individual microfilariae can be visualized and counted on autoradiograms. The assay was evaluated in a mock field study using Brugia malayi-infected jirds and was found to be an efficient and accurate method for quantifying microfilariae in blood samples. Key words: Lymphatic filariasis; Brugia malayi; Brugia pahangi; DNA probe; Species identification; Repeated DNA
Introduction T h e design and evaluation of successful parasite control p r o g r a m s d e p e n d s on the ability of investigators to gather detailed and accurate data in epidemiological surveys. T h e s t a n d a r d diagnostic tools for gathering such data on filarial parasites are severely limited b o t h in terms of their applicability to large-scale surveys and in terms of species identification. Carriers of lymphatic filarial parasites are typically identified by the detection of microfilariae using light microsc o p y in stained b l o o d smears. This m e t h o d is not only tedious, but species or strain identification is
Correspondence address: Steven A. Williams, Department of Biological Sciences, Clark Science Center, Smith College, Northampton, MA 01063, U.S.A. "Present address: New England Biolabs, Inc., 32 Tozer Road, Beverly, MA 01915, U.S.A. Abbreviations: SSC, standard saline citrate; SDS, sodium dodecyl sulfate; i.p., intraperitoneal.
often difficult, if not impossible [1]. T h e s e problems greatly limit the epidemiological value of data collected f r o m stained b l o o d smears. Various biochemical m e a n s of detection and quantification have p r o v e n to be insensitive or t o o complex for use in large-scale surveys [1-5]. Until a new m e t h o d for detection and identification is widely available, m a n y questions regarding the epidemiology, transmission and pathogenicity of lymphatic filarial parasites will r e m a i n unanswered. A cloned D N A p r o b e could p r o v e useful in a diagnostic assay if three criteria are met. First, the D N A p r o b e should be at least as sensitive and reliable as the established m e t h o d ( s ) of detection. Second, the D N A p r o b e must be exquisitely species-specific. T h e r e must be no cross-reactivity with host D N A or with the D N A of closely related species of parasites. Third, the assay based on these D N A p r o b e s must be practical w h e n applied on a large scale in developing nations. T h e first two criteria have clearly b e e n met by our D N A p r o b e s [6,7], and the third criterion is addressed in this paper.
0166-6851/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
130
The Hhal repeat D N A family we have found in Brugia consists of members which are 322 nucleotides in length and make up 10% of the parasite's total D N A (30 000 copies per haploid genome) [6,7]. Due to 89% sequence homology between the repeats of Brugia malayi and those of Brugia pahangi, there is often cross-hybridization when a repeat of one species is used to probe DNA from the other species [6]. Speciesspecificity can be achieved in two ways. First, oligonucleotide probes can be constructed based on a region within the repeat that is only 72% homologous between the two species. Oligonucleotide probes based on this region of divergence are species-specific and sensitive when hybridized under conditions of moderate stringency (66°C, 5 x SSC) [7]. A second approach is to use full-length cloned repeats as hybridization probes. To achieve species-specificity with these probes, more stringent hybridization conditions must be used (70°C, 5 x SSC) [81. This paper first describes the assay we have developed using labeled D N A probes to detect individual microfilariae in the blood of vertebrate hosts. The method was developed using human blood samples to which known numbers of B. malayi or B. pahangi microfilariae had been added. The method has only a few simple steps, does not require any D N A purification procedures, and results in the detection of individual parasites in a species-specific and quantitative manner. To demonstrate the efficacy of this assay for use in field surveys, a mock field trial was conducted using blood samples from jirds infected with B. malayi. Materials and Methods
B. malayi and B. pahangi microfilariae were obtained from John McCall (TRS Laboratory, Athens, GA). These microfilariae were harvested from the peritoneal cavities of i.p.-infected jirds (Meriones unguiculatus) [9], separated from peritoneal cells by density gradient centrifugation [10], and then shipped overnight packed in wet ice. Parasite D N A was isolated by proteinase K (Boehringer Mannheim, Indianapolis, IN) digestion of B. malayi or B. pahangi microfilariae [6]. Uninfected venous blood from one
of the authors was drawn into a vacutainer tube (Becton-Dickenson, Vacutainer Systems, Rutherford, NJ) containing the anticoagulant EDTA.
Filtration of human blood samples through nitrocellulose membranes. Blood samples containing microfilariae were prepared by mixing 150 txl 0.85% NaCI containing either 100 B. malayi or B. pahangi microfilariae with 150 txl of human blood. Each sample was applied to a 5.0 Dxmpore nitrocellulose filter disk (25 mm diameter; Schleicher and Schuell, Keene, NH) on a glass filtration apparatus (25 mm diameter glass filter holder with glass frit; Schleicher and Schuell). A vacuum was applied by water aspiration. Each disk was rinsed with 1.0 ml 0.5 M NaOH by filtration on the same apparatus. This rinse was followed by a rinse with 1.0 ml 1 M Tris-HCl, pH 7.5, and then 1.0 ml 0.5 M Tris-HCl, pH 7.5/1.5 M NaC1. Each disk was then air-dried, sample side up, prior to proteinase K digestion. This procedure filters most of the blood components through the nitrocellulose disks while retaining the microfilariae. Rinsing the disks with NaOH and the Tris solutions removes blood constituents that can cause a high background signal when hybridized with radiolabeled DNA probes.
Proteinase K digestion of microfilariae to release DNA for hybridization. Two pieces of thick filter paper (3MM; Whatman, Clifton, NJ), cut to fit a large 120-mm diameter petri dish, were placed in the cover of the petri dish and saturated with 1 x SSC (0.15 M NaC1, 0.015 M sodium citrate pH 7.0), 4% Sarkosyl (N-Lauroylsarcosine sodium salt, Sigma Chemical Co., St. Louis, MO), and 100 Ixg m1-1 proteinase K (Boehringer-Mannheim, Indianapolis, IN). The air-dried nitrocellulose disks were placed on top of the saturated flter paper, sample side up, taking care that no air bubbles were trapped between the disks and the filter paper. The bottom half of the petri dish was carefully placed over the top half, and the dish was sealed with parafilm to prevent evaporation and drying of the disks during the incubation. The petri dishes were then incubated at 55°C for 1 h. Following the incubation, the disks were removed from the petri dishes and set aside to air dry. This procedure disrupts both B. malayi and
131
B. pahangi microfilariae in situ, exposing the parasite DNA for hybridization. NaOH denaturation of the microfilaria DNA. Following proteinase K digestion, the air-dried nitrocellulose disks were transferred, sample side up, onto sheets of thick filter paper (3MM; Whatman) saturated with 0.5 M NaOH for 2 min. The disks were transferred to a second stack of filter paper saturated with 0.5 M NaOH for an additional 2 min, then sequentially to two stacks saturated with 1 M Tris-HCl, pH 7.5 for 5 min each, and finally to two stacks saturated with 0.5 M Tris-HCl pH 7.5, 1.5 M NaCI for 5 min each. The filters were air-dried briefly and then baked under vacuum at 60°C for 2 h. This procedure denatures the exposed parasite DNA from the double-stranded to the single-stranded form in preparation for hybridization. Hybridization of the nitrocellulose disks with radiolabeled probe DNA. The nitrocellulose disks were prehybridized in plastic hybridization bags (Bethesda Research Laboratories, Bethesda, MD) at 70°C for 2 h in 5 × SSC, 5 x Denhardt's solution [11], 100 ~g m1-1 sonicated salmon sperm DNA, 0.1% sodium dodecyl sulfate (SDS) and 10 mM EDTA. The disks were arranged back-toback in pairs and sealed in place with a tacking iron so that the hybridization solution flowed freely over the disks while they remained fixed in place. This method ensures uniform hybridization for all samples because it prevents the disks from clinging to one another in clumps. Following prehybridization, radiolabeled B. malayi specific probe D N A (MB1) or B. pahangi specific probe D N A (P24R) was added to each bag (1 x 106 cpm m1-1 of hybridization solution). MB1 is an M13mpl8 bacteriophage clone containing a 322 base pair HhaI repeat cloned from B. malayi [7]. When stringent hybridization conditions are used, this clone gives very sensitive, species-specific detection of B. malayi DNA. P24R is an M13mpl8 bacteriophage clone containing a 322-base-pair HhaI repeat cloned from B. pahangi. With stringent hybridization conditions this clone gives very sensitive, species-specific detection of B. pahangi DNA. These probes were labeled by the oligonucleotide random priming method [12,13] using
[a-35S]dCTP (1000 Ci mmol-1; Amersham, Arlington Heights, IL). The labeled DNA was separated from unincorporated [ot-35S]dCTP nucleotides by exclusion chromatography using a G-50 Sephadex column (Pharmacia LKB Biotechnology, Piscataway, N J). The disks were hybridized overnight at 70°C with constant agitation. Following hybridization, the disks were washed three times at 70°C for 30 min each in 2 × SSC/0.1% SDS. The disks were then dried under vacuum at 60°C for 1 h prior to autoradiography.
Mock field trial .using jird blood samples to evaluate the DNA assay for quantifying microfilariae. Blood was collected from 45 jirds infected with B. malayi. The animals were bled from the retroorbital sinus with a pasteur pipet. 1 ml of blood was transferred from the pipet to a test tube containing 0.5 ml 1.1 mg m1-1 sodium heparin. To count the number of microfilariae in each sample using the concentration/staining technique [14,15], 150 ixl of blood was filtered through a 5.0 I~m pore polycarbonate filter (Nucleopore Corp.) using a glass frit filtration apparatus (Schleicher and Schuell) and a water aspirator. Duplicate filters were prepared from each jird blood sample. Each filter was placed, sample side up, on a glass microscope slide and stained with both Wright's and Giemsa stains [16,17]. The number of microfilariae was counted by examining each filter using light microscopy at 100 × magnification. The number of microfilariae in each jird blood sample was also determined using the D N A assay described above. 150 ~1 of each jird blood sample was mixed with 150 ~1 of 0.85% NaCI. Samples were filtered through nitrocellulose, rinsed, proteinase K digested, denatured and prehybridized exactly as described above. Each sample was prepared in duplicate. The nitrocellulose disks were then hybridized with [a-35S]dCTP-labeled MB1 D N A (1 x 106 cpm m1-1 of hybridization solution). Single-stranded DNA from this clone was labeled by the oligonucleotide random priming method [12,13]. The disks were hybridized, washed and autoradiography performed as described above. Following autoradiography, the number of spots from each disk was counted and recorded.
132 ResuRs and Discussion
The primary objective of this study was to design a DNA hybridization assay that would use labeled DNA probes to detect individual microfilariae in blood samples. Complicated, expensive and time consuming steps to separate parasites from blood or to purify parasite DNA were avoided. Literally dozens of methods were devised, tested and discarded due to a lack of sensitivity, reliability or convenience. The method reported here has proven to be fast, reliable, sensitive, quantitative, and species-specific. The method also has the virtues of being inexpensive and not requiring specialized instrumentation.
The DNA assay for species-specific detection of individual Brugia microfilariae in human blood samples. Previous DNA hybridization work in our laboratory [6] showed that blood components (possibly the heme in erythrocytes) must be removed from nitrocellulose membranes following application of a blood sample. If these constituents are not removed, a high background signal results that can interfere with the detection of filarial DNA in human, cat and jird blood (unpublished data). The method involves first filtering the human blood through nitrocellulose membranes that have a pore size of 5.0 Ixm. When a human blood sample is applied to the filter with modest vacuum pressure (a hand-held vacuum pump is sufficient), red blood cells and some cellular debris pass through the filter, while intact microfilariae remain on the surface of the membrane. In one simple filtration step, most of the components in blood that interfere with DNA hybridization and signal detection are eliminated. The standard nitrocellulose membranes have a pore size that is too small (0.45 ~m) to allow red blood cells and cellular debris to pass. When one attempts to filter blood through such membranes, they rapidly clog and become useless. It should be noted that erythrocytes of both humans and jirds have a diameter of about 5 p~m. Following filtration of the human blood samples through the 5.0 p.m membranes, a NaOH solution is filtered through to rinse the membrane of residual blood components. This rinse is crucial to reducing the background signal caused by
blood constituents. Following the NaOH, Trisbuffered solutions must be filtered through the membrane. This is necessary to neutralize and prevent disintegration of the nitrocellulose membranes. The microfilariae are then lysed in situ on the nitrocellulose membranes by digestion with proteinase K in the presence of the detergent Sarkosyl. The use of Sarkosyl rather than sodium dodecyl sulfate (SDS) is crucial since SDS was shown to cause severe smearing of the DNA signal. Such smearing prevents the detection of individual microfilariae on autoradiograms. The DNA released by the digested microfilariae is then denatured by placing the membranes on filter paper saturated with NaOH. This treatment gives sharp, distinct signals on autoradiograms with no smearing or blurting of the signal. It also helps to reduce background hybridization by leaching residual blood components from the membrane to the saturated filter paper. The probes used in the hybridization assay are full-length 322-bp HhaI repeats. The clone used to detect B. malayi (MB1) includes a copy of the HhaI repeat cloned from B. malayi microfilariae from Bengkulu, Indonesia. The clone used to detect B. pahangi (P24R) includes a copy of the HhaI repeat cloned from B. pahangi obtained from John McCall (TRS Laboratory, Athens, GA). The copy of the HhaI repeat in MB1 has the DNA sequence that most closely matches the B. malayi HhaI repeat consensus sequence [7]. The copy of the HhaI repeat in P24R has the DNA sequence that most closely matches the B. pahangi HhaI repeat consensus sequence [7]. A variety of hybridization and wash conditions were evaluated for these probes. Hybridization at 70°C in 5 × SSC followed by three stringent washes at 70°C in 2 × SSC gave the strongest signal while maintaining species specificity of the probes. Less stringent hybridization and wash conditions (70°C) the intensity of the hybridization signal was reduced. Blood blots using this experimental protocol gave a strong signal with no cross-hybridization to microfilariae of the other species. Both B. malayi and B. pa-
133
a
b
d
Fig. 1. Species-specific detection of individual Brugiu microfilariae immobilized on nitrocellulose membranes. Human blood containing B. malayi microfilariae was filtered through the membranes shown in (a) and (c). Human blood containing B. pahangi microfilariae was filtered through the membranes shown in (b) and (d). The membranes were rinsed, the microfilariae were lysed, the parasite DNA was denatured and the filters were hybridized as described in Materials and Methods. The filters shown in (a) and (b) were hybridized with the [a-3sS]dCTP-1abe1ed B. malayi probe, MBl. The filters shown in (c) and (d) were hybridized with the [a-3sS]dCTP-1abe1ed B. pahangi probe, R23R. Individual spots representing B. malayi microfilariae were only seen on autoradiograms of the B. maluyi filters hybridized with the B. maluyi probe (a) and not on the B. malayi filters hybridized with the B. pahangi probe (c). Likewise, individual spots representing B. pahangi microfilariae were only seen on autoradiograms of the B. pahangi filters hybridized with the B. pahangi probe (d) and not on the B. pahwgi filters hybridized with the B. malayi probe (b).
hangi microfilariae in human blood were detected as distinct spots on autoradiograms using this method (Fig.1). Nitrocellulose disks were prepared by filtering B. malayi in human blood and others by filtering B. puhungi in human blood. When these filters were hybridized with the B. muluyi specific probe (MBl), individual spots representing microfilariae were seen on the autoradiograms of the B. muluyi filters but not the B. puhungi filters (Fig. la and b). When these filters were hybridized with the B. puhungi specific probe (P24R), individual spots representing microfilariae were seen on the autoradiograms of the B. puhungi filters, but not the B. maluyi filters (Fig. Id and c). With this method, the microfilariae are visualized as sharp, distinct spots on the autoradiogram after only a one- or two-day exposure to the filters. These results demonstrate that the method is equally effective in detecting B. puhungi microfilariae with a B. puhungi specific DNA probe as in detecting B. mufuyi microfilariae with a B. muluyi specific DNA probe. Oligonucleotide probes based on the short, highly divergent region of the HhuI repeat can
also be used in this assay [7]. These probes give equivalent sensitivity and species-specificity but at less stringent hybridization conditions (66°C 5 x SSC) than the full-length repeat probes. Both the full-length repeats used in this study and the oligonucleotide probes used in our previous study have been used to detect microfilariae from a variety of endemic regions in Indonesia (Williams, S.A., Szabo, S.J., McReynolds, L.A., Partono, F., manuscript in preparation). Mock field trial using jird blood samples to evuluute the DNA assay for quantifying microfiluriue.
To evaluate the potential use of the DNA assay in field studies, and to demonstrate the concordance between the number of spots counted on autoradiograms (DNA assay) and the number of microfilariae counted on filters (stain assay), both methods were used to test blood samples from 45 jirds infected with B. muluyi. For each jird blood sample, two filters were prepared for the DNA assay and two for the concentration/staining technique. Concentration/staining was the method shown by Southgate to be the most sensitive of
134
the morphological/staining procedures for detecting microfilariae [18]. The DNA assay and the concentration/staining technique gave very similar results in each of the 45 jird blood samples examined. The graphs in Fig. 2 show that as the microfilaria titers increased, the number of microfilariae detected by each of the methods increased concomitantly. 20 of the samples showed no microfilariae on either of the two concentration/stain filters or on either of the two DNA assay filters. These 20 samples are shown as only a single point at the origin in Fig. 2. When all 45 samples are included in the data set, the correlation coefficient between the DNA assay results and the concentration/staining results is 0.99 (P
129
MOLBIO 01316
A rapid D N A assay for the species-specific detection and quantification of Brugia in blood samples Catherine B. Poole 1'* and Steven A. Williams 1'2 XDepartment of Biological Sciences, Smith College, Northampton, MA, U.S.A. and 2Molecular and Cellular Biology Program, University of Massachusetts, Amherst, MA, U.S.A. (Received 1 September 1989; accepted 4 December 1989)
This report describes a new assay for detecting filarial parasites of the genus Brugia in blood samples using labeled DNA probes. The sequences of these DNA probes are based on the HhaI repeat DNA family found in the genus Brugia. These DNA probes are species-specific and can detect the DNA from a single microfilaria in hybridization assays. To adapt this test for use on blood samples collected in the field, complex steps to separate microfilariae from blood cells and to purify parasite DNA were eliminated. We found that the most effective method was to filter blood samples through 5.0 ~.m pore nitrocellulose membranes, lyse the microfilariae on the membranes by proteinase K digestion, denature the parasite DNA with sodium hydroxide, and hybridize with the DNA probe. With this method, individual microfilariae can be visualized and counted on autoradiograms. The assay was evaluated in a mock field study using Brugia malayi-infected jirds and was found to be an efficient and accurate method for quantifying microfilariae in blood samples. Key words: Lymphatic filariasis; Brugia malayi; Brugia pahangi; DNA probe; Species identification; Repeated DNA
Introduction T h e design and evaluation of successful parasite control p r o g r a m s d e p e n d s on the ability of investigators to gather detailed and accurate data in epidemiological surveys. T h e s t a n d a r d diagnostic tools for gathering such data on filarial parasites are severely limited b o t h in terms of their applicability to large-scale surveys and in terms of species identification. Carriers of lymphatic filarial parasites are typically identified by the detection of microfilariae using light microsc o p y in stained b l o o d smears. This m e t h o d is not only tedious, but species or strain identification is
Correspondence address: Steven A. Williams, Department of Biological Sciences, Clark Science Center, Smith College, Northampton, MA 01063, U.S.A. "Present address: New England Biolabs, Inc., 32 Tozer Road, Beverly, MA 01915, U.S.A. Abbreviations: SSC, standard saline citrate; SDS, sodium dodecyl sulfate; i.p., intraperitoneal.
often difficult, if not impossible [1]. T h e s e problems greatly limit the epidemiological value of data collected f r o m stained b l o o d smears. Various biochemical m e a n s of detection and quantification have p r o v e n to be insensitive or t o o complex for use in large-scale surveys [1-5]. Until a new m e t h o d for detection and identification is widely available, m a n y questions regarding the epidemiology, transmission and pathogenicity of lymphatic filarial parasites will r e m a i n unanswered. A cloned D N A p r o b e could p r o v e useful in a diagnostic assay if three criteria are met. First, the D N A p r o b e should be at least as sensitive and reliable as the established m e t h o d ( s ) of detection. Second, the D N A p r o b e must be exquisitely species-specific. T h e r e must be no cross-reactivity with host D N A or with the D N A of closely related species of parasites. Third, the assay based on these D N A p r o b e s must be practical w h e n applied on a large scale in developing nations. T h e first two criteria have clearly b e e n met by our D N A p r o b e s [6,7], and the third criterion is addressed in this paper.
0166-6851/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
130
The Hhal repeat D N A family we have found in Brugia consists of members which are 322 nucleotides in length and make up 10% of the parasite's total D N A (30 000 copies per haploid genome) [6,7]. Due to 89% sequence homology between the repeats of Brugia malayi and those of Brugia pahangi, there is often cross-hybridization when a repeat of one species is used to probe DNA from the other species [6]. Speciesspecificity can be achieved in two ways. First, oligonucleotide probes can be constructed based on a region within the repeat that is only 72% homologous between the two species. Oligonucleotide probes based on this region of divergence are species-specific and sensitive when hybridized under conditions of moderate stringency (66°C, 5 x SSC) [7]. A second approach is to use full-length cloned repeats as hybridization probes. To achieve species-specificity with these probes, more stringent hybridization conditions must be used (70°C, 5 x SSC) [81. This paper first describes the assay we have developed using labeled D N A probes to detect individual microfilariae in the blood of vertebrate hosts. The method was developed using human blood samples to which known numbers of B. malayi or B. pahangi microfilariae had been added. The method has only a few simple steps, does not require any D N A purification procedures, and results in the detection of individual parasites in a species-specific and quantitative manner. To demonstrate the efficacy of this assay for use in field surveys, a mock field trial was conducted using blood samples from jirds infected with B. malayi. Materials and Methods
B. malayi and B. pahangi microfilariae were obtained from John McCall (TRS Laboratory, Athens, GA). These microfilariae were harvested from the peritoneal cavities of i.p.-infected jirds (Meriones unguiculatus) [9], separated from peritoneal cells by density gradient centrifugation [10], and then shipped overnight packed in wet ice. Parasite D N A was isolated by proteinase K (Boehringer Mannheim, Indianapolis, IN) digestion of B. malayi or B. pahangi microfilariae [6]. Uninfected venous blood from one
of the authors was drawn into a vacutainer tube (Becton-Dickenson, Vacutainer Systems, Rutherford, NJ) containing the anticoagulant EDTA.
Filtration of human blood samples through nitrocellulose membranes. Blood samples containing microfilariae were prepared by mixing 150 txl 0.85% NaCI containing either 100 B. malayi or B. pahangi microfilariae with 150 txl of human blood. Each sample was applied to a 5.0 Dxmpore nitrocellulose filter disk (25 mm diameter; Schleicher and Schuell, Keene, NH) on a glass filtration apparatus (25 mm diameter glass filter holder with glass frit; Schleicher and Schuell). A vacuum was applied by water aspiration. Each disk was rinsed with 1.0 ml 0.5 M NaOH by filtration on the same apparatus. This rinse was followed by a rinse with 1.0 ml 1 M Tris-HCl, pH 7.5, and then 1.0 ml 0.5 M Tris-HCl, pH 7.5/1.5 M NaC1. Each disk was then air-dried, sample side up, prior to proteinase K digestion. This procedure filters most of the blood components through the nitrocellulose disks while retaining the microfilariae. Rinsing the disks with NaOH and the Tris solutions removes blood constituents that can cause a high background signal when hybridized with radiolabeled DNA probes.
Proteinase K digestion of microfilariae to release DNA for hybridization. Two pieces of thick filter paper (3MM; Whatman, Clifton, NJ), cut to fit a large 120-mm diameter petri dish, were placed in the cover of the petri dish and saturated with 1 x SSC (0.15 M NaC1, 0.015 M sodium citrate pH 7.0), 4% Sarkosyl (N-Lauroylsarcosine sodium salt, Sigma Chemical Co., St. Louis, MO), and 100 Ixg m1-1 proteinase K (Boehringer-Mannheim, Indianapolis, IN). The air-dried nitrocellulose disks were placed on top of the saturated flter paper, sample side up, taking care that no air bubbles were trapped between the disks and the filter paper. The bottom half of the petri dish was carefully placed over the top half, and the dish was sealed with parafilm to prevent evaporation and drying of the disks during the incubation. The petri dishes were then incubated at 55°C for 1 h. Following the incubation, the disks were removed from the petri dishes and set aside to air dry. This procedure disrupts both B. malayi and
131
B. pahangi microfilariae in situ, exposing the parasite DNA for hybridization. NaOH denaturation of the microfilaria DNA. Following proteinase K digestion, the air-dried nitrocellulose disks were transferred, sample side up, onto sheets of thick filter paper (3MM; Whatman) saturated with 0.5 M NaOH for 2 min. The disks were transferred to a second stack of filter paper saturated with 0.5 M NaOH for an additional 2 min, then sequentially to two stacks saturated with 1 M Tris-HCl, pH 7.5 for 5 min each, and finally to two stacks saturated with 0.5 M Tris-HCl pH 7.5, 1.5 M NaCI for 5 min each. The filters were air-dried briefly and then baked under vacuum at 60°C for 2 h. This procedure denatures the exposed parasite DNA from the double-stranded to the single-stranded form in preparation for hybridization. Hybridization of the nitrocellulose disks with radiolabeled probe DNA. The nitrocellulose disks were prehybridized in plastic hybridization bags (Bethesda Research Laboratories, Bethesda, MD) at 70°C for 2 h in 5 × SSC, 5 x Denhardt's solution [11], 100 ~g m1-1 sonicated salmon sperm DNA, 0.1% sodium dodecyl sulfate (SDS) and 10 mM EDTA. The disks were arranged back-toback in pairs and sealed in place with a tacking iron so that the hybridization solution flowed freely over the disks while they remained fixed in place. This method ensures uniform hybridization for all samples because it prevents the disks from clinging to one another in clumps. Following prehybridization, radiolabeled B. malayi specific probe D N A (MB1) or B. pahangi specific probe D N A (P24R) was added to each bag (1 x 106 cpm m1-1 of hybridization solution). MB1 is an M13mpl8 bacteriophage clone containing a 322 base pair HhaI repeat cloned from B. malayi [7]. When stringent hybridization conditions are used, this clone gives very sensitive, species-specific detection of B. malayi DNA. P24R is an M13mpl8 bacteriophage clone containing a 322-base-pair HhaI repeat cloned from B. pahangi. With stringent hybridization conditions this clone gives very sensitive, species-specific detection of B. pahangi DNA. These probes were labeled by the oligonucleotide random priming method [12,13] using
[a-35S]dCTP (1000 Ci mmol-1; Amersham, Arlington Heights, IL). The labeled DNA was separated from unincorporated [ot-35S]dCTP nucleotides by exclusion chromatography using a G-50 Sephadex column (Pharmacia LKB Biotechnology, Piscataway, N J). The disks were hybridized overnight at 70°C with constant agitation. Following hybridization, the disks were washed three times at 70°C for 30 min each in 2 × SSC/0.1% SDS. The disks were then dried under vacuum at 60°C for 1 h prior to autoradiography.
Mock field trial .using jird blood samples to evaluate the DNA assay for quantifying microfilariae. Blood was collected from 45 jirds infected with B. malayi. The animals were bled from the retroorbital sinus with a pasteur pipet. 1 ml of blood was transferred from the pipet to a test tube containing 0.5 ml 1.1 mg m1-1 sodium heparin. To count the number of microfilariae in each sample using the concentration/staining technique [14,15], 150 ixl of blood was filtered through a 5.0 I~m pore polycarbonate filter (Nucleopore Corp.) using a glass frit filtration apparatus (Schleicher and Schuell) and a water aspirator. Duplicate filters were prepared from each jird blood sample. Each filter was placed, sample side up, on a glass microscope slide and stained with both Wright's and Giemsa stains [16,17]. The number of microfilariae was counted by examining each filter using light microscopy at 100 × magnification. The number of microfilariae in each jird blood sample was also determined using the D N A assay described above. 150 ~1 of each jird blood sample was mixed with 150 ~1 of 0.85% NaCI. Samples were filtered through nitrocellulose, rinsed, proteinase K digested, denatured and prehybridized exactly as described above. Each sample was prepared in duplicate. The nitrocellulose disks were then hybridized with [a-35S]dCTP-labeled MB1 D N A (1 x 106 cpm m1-1 of hybridization solution). Single-stranded DNA from this clone was labeled by the oligonucleotide random priming method [12,13]. The disks were hybridized, washed and autoradiography performed as described above. Following autoradiography, the number of spots from each disk was counted and recorded.
132 ResuRs and Discussion
The primary objective of this study was to design a DNA hybridization assay that would use labeled DNA probes to detect individual microfilariae in blood samples. Complicated, expensive and time consuming steps to separate parasites from blood or to purify parasite DNA were avoided. Literally dozens of methods were devised, tested and discarded due to a lack of sensitivity, reliability or convenience. The method reported here has proven to be fast, reliable, sensitive, quantitative, and species-specific. The method also has the virtues of being inexpensive and not requiring specialized instrumentation.
The DNA assay for species-specific detection of individual Brugia microfilariae in human blood samples. Previous DNA hybridization work in our laboratory [6] showed that blood components (possibly the heme in erythrocytes) must be removed from nitrocellulose membranes following application of a blood sample. If these constituents are not removed, a high background signal results that can interfere with the detection of filarial DNA in human, cat and jird blood (unpublished data). The method involves first filtering the human blood through nitrocellulose membranes that have a pore size of 5.0 Ixm. When a human blood sample is applied to the filter with modest vacuum pressure (a hand-held vacuum pump is sufficient), red blood cells and some cellular debris pass through the filter, while intact microfilariae remain on the surface of the membrane. In one simple filtration step, most of the components in blood that interfere with DNA hybridization and signal detection are eliminated. The standard nitrocellulose membranes have a pore size that is too small (0.45 ~m) to allow red blood cells and cellular debris to pass. When one attempts to filter blood through such membranes, they rapidly clog and become useless. It should be noted that erythrocytes of both humans and jirds have a diameter of about 5 p~m. Following filtration of the human blood samples through the 5.0 p.m membranes, a NaOH solution is filtered through to rinse the membrane of residual blood components. This rinse is crucial to reducing the background signal caused by
blood constituents. Following the NaOH, Trisbuffered solutions must be filtered through the membrane. This is necessary to neutralize and prevent disintegration of the nitrocellulose membranes. The microfilariae are then lysed in situ on the nitrocellulose membranes by digestion with proteinase K in the presence of the detergent Sarkosyl. The use of Sarkosyl rather than sodium dodecyl sulfate (SDS) is crucial since SDS was shown to cause severe smearing of the DNA signal. Such smearing prevents the detection of individual microfilariae on autoradiograms. The DNA released by the digested microfilariae is then denatured by placing the membranes on filter paper saturated with NaOH. This treatment gives sharp, distinct signals on autoradiograms with no smearing or blurting of the signal. It also helps to reduce background hybridization by leaching residual blood components from the membrane to the saturated filter paper. The probes used in the hybridization assay are full-length 322-bp HhaI repeats. The clone used to detect B. malayi (MB1) includes a copy of the HhaI repeat cloned from B. malayi microfilariae from Bengkulu, Indonesia. The clone used to detect B. pahangi (P24R) includes a copy of the HhaI repeat cloned from B. pahangi obtained from John McCall (TRS Laboratory, Athens, GA). The copy of the HhaI repeat in MB1 has the DNA sequence that most closely matches the B. malayi HhaI repeat consensus sequence [7]. The copy of the HhaI repeat in P24R has the DNA sequence that most closely matches the B. pahangi HhaI repeat consensus sequence [7]. A variety of hybridization and wash conditions were evaluated for these probes. Hybridization at 70°C in 5 × SSC followed by three stringent washes at 70°C in 2 × SSC gave the strongest signal while maintaining species specificity of the probes. Less stringent hybridization and wash conditions (70°C) the intensity of the hybridization signal was reduced. Blood blots using this experimental protocol gave a strong signal with no cross-hybridization to microfilariae of the other species. Both B. malayi and B. pa-
133
a
b
d
Fig. 1. Species-specific detection of individual Brugiu microfilariae immobilized on nitrocellulose membranes. Human blood containing B. malayi microfilariae was filtered through the membranes shown in (a) and (c). Human blood containing B. pahangi microfilariae was filtered through the membranes shown in (b) and (d). The membranes were rinsed, the microfilariae were lysed, the parasite DNA was denatured and the filters were hybridized as described in Materials and Methods. The filters shown in (a) and (b) were hybridized with the [a-3sS]dCTP-1abe1ed B. malayi probe, MBl. The filters shown in (c) and (d) were hybridized with the [a-3sS]dCTP-1abe1ed B. pahangi probe, R23R. Individual spots representing B. malayi microfilariae were only seen on autoradiograms of the B. maluyi filters hybridized with the B. maluyi probe (a) and not on the B. malayi filters hybridized with the B. pahangi probe (c). Likewise, individual spots representing B. pahangi microfilariae were only seen on autoradiograms of the B. pahangi filters hybridized with the B. pahangi probe (d) and not on the B. pahwgi filters hybridized with the B. malayi probe (b).
hangi microfilariae in human blood were detected as distinct spots on autoradiograms using this method (Fig.1). Nitrocellulose disks were prepared by filtering B. malayi in human blood and others by filtering B. puhungi in human blood. When these filters were hybridized with the B. muluyi specific probe (MBl), individual spots representing microfilariae were seen on the autoradiograms of the B. muluyi filters but not the B. puhungi filters (Fig. la and b). When these filters were hybridized with the B. puhungi specific probe (P24R), individual spots representing microfilariae were seen on the autoradiograms of the B. puhungi filters, but not the B. maluyi filters (Fig. Id and c). With this method, the microfilariae are visualized as sharp, distinct spots on the autoradiogram after only a one- or two-day exposure to the filters. These results demonstrate that the method is equally effective in detecting B. puhungi microfilariae with a B. puhungi specific DNA probe as in detecting B. mufuyi microfilariae with a B. muluyi specific DNA probe. Oligonucleotide probes based on the short, highly divergent region of the HhuI repeat can
also be used in this assay [7]. These probes give equivalent sensitivity and species-specificity but at less stringent hybridization conditions (66°C 5 x SSC) than the full-length repeat probes. Both the full-length repeats used in this study and the oligonucleotide probes used in our previous study have been used to detect microfilariae from a variety of endemic regions in Indonesia (Williams, S.A., Szabo, S.J., McReynolds, L.A., Partono, F., manuscript in preparation). Mock field trial using jird blood samples to evuluute the DNA assay for quantifying microfiluriue.
To evaluate the potential use of the DNA assay in field studies, and to demonstrate the concordance between the number of spots counted on autoradiograms (DNA assay) and the number of microfilariae counted on filters (stain assay), both methods were used to test blood samples from 45 jirds infected with B. muluyi. For each jird blood sample, two filters were prepared for the DNA assay and two for the concentration/staining technique. Concentration/staining was the method shown by Southgate to be the most sensitive of
134
the morphological/staining procedures for detecting microfilariae [18]. The DNA assay and the concentration/staining technique gave very similar results in each of the 45 jird blood samples examined. The graphs in Fig. 2 show that as the microfilaria titers increased, the number of microfilariae detected by each of the methods increased concomitantly. 20 of the samples showed no microfilariae on either of the two concentration/stain filters or on either of the two DNA assay filters. These 20 samples are shown as only a single point at the origin in Fig. 2. When all 45 samples are included in the data set, the correlation coefficient between the DNA assay results and the concentration/staining results is 0.99 (P
