Proc. Nati. Acad. Sci. USA Vol. 81, pp. 889-893, February 1984 Medical Sciences

Differentiation of schistosomes by species, strain, and sex by using cloned DNA markers (parasitic trematodes/speciation/ribosomal genes) THOMAS F. MCCUTCHAN*, ANDREW J. G. SIMPSON*, JUDITH A. MULLINS*, ALAN SHER*, THEODORE E. NASH*, FRED LEWISt, AND CHARLES RICHARDSt *Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20205; and tBiomedical Research Institute, 12111 Parklawn Drive, Rockville, MD 20852 Communicated by Richard M. Krause, September 15, 1983

ABSTRACT We have detected species, strain, and sex- vide useful strain distinctions in this system. Interactions be- specific genetic markers for the genus by South- tween parasite and tend to concentrate populations into ern blot analysis of its DNA using cloned DNA segments of the foci, allowing genetic differences that would otherwise be ribosomal gene as probes. Restriction homogenized to persist. analysis of DNA from eight different strains of S. mansoni, from Africa and the Caribbean, revealed that the predominant or major DNA fragment containing the ribosomal gene unit EXPERIMENTAL PROCEDURES was the same in each but that low copy number or minor frag- Enzymes. Restriction endonucleases were purchased from ments containing the gene varied. It was shown that the detec- either Boehringer Mannheim or Bethesda Research Labora- tion of these minor fragments could serve as the basis for both tories. Digestions were carried out in enzyme excess, with strain differentiation and the analysis of individual differences the conditions recommended by the manufacturer. within a strain. Analysis of the parents and progeny ofa genet- DNA Isolation. DNA was isolated from populations of ic cross revealed sex-linked markers and suggested that these schistosomes as described earlier (8). DNA from individual markers are inherited in a Mendelian fashion. from the worms was extracted as follows. Each worm was frozen in species and dry ice and then pulverized with a glass rod. The homoge- were also analyzed. Differences in the length of the major re- nate was thawed into 50 Al of 50 mM Tris-HCl (pH 8.0) con- peating unit of the ribosomal gene served to distinguish each taining 50 mM EDTA and 100 mM NaCl (extraction buffer) species. Furthermore, an array ofminor bands was detected in at 37TC. When thawed, an equal volume of extraction buffer each species, suggesting that strains of S. haematobium and S. containing 1% NaDodSO4 and 10 ,Ag of proteinase K was japonicum could be differentiated in the same manner as S. added and the solution was incubated at 37TC for 1 hr. The mansoni strains. lysate was then extracted twice with an equal volume of phe- nol and once with chloroform. Five micrograms of tRNA Schistosomes are parasitic trematodes and are a major cause (Boehringer Mannheim) or 5 Ag of salmon sperm DNA was of human disease in parts of Africa, Asia, South America, added to the solution and then it was adjusted to 0.3 M with and the Caribbean. There are 17 species of Schistosoma, sodium acetate and precipitated with 2 vol of ethanol. which are separated into three groups based on egg morphol- Parasites. Adult worms and cercariae of S. mansoni ogy. Each grouping of species is further differentiated by an- strains were provided by the Biomedical Research Institute atomical, developmental, and host range characteristics. (Rockville, MD). Schistosoma haematobium and Schisto- However, few methods exist for distinguishing parasites soma japonicum used in this study were provided by a Na- within a species, a problem which may be of critical impor- tional Institute of Allergy and Infectious Diseases supply tance. Intraspecies diversity may be the basis of differences contract (Al 02656). Cercariae resulting from genetic crosses in the course and severity of the disease in humans or be of the PRC3 and PRT3 strains (Table 1) were kindly provided related to its epidemiology. Certainly, differences in snail in- by A. Cheever. Cercariae were concentrated by centrifuga- fectivity (1, 2), drug resistance (3), and isoenzyme patterns tion, washed once with phosphate-buffered saline (pH 7.6), (4, 5) among parasites of a single species suggest that consid- and frozen at -70TC. Seven- to eight-week-old adult worms erable genetic variability does exist. Further, pathogenicity were perfused from the mesenteric veins of outbred Swiss of Schistosoma mansoni as measured in mice seems to relate mice by the technique of Duvall and DeWitt (9). Perfusion to genetic differences within the infecting parasite (6, 7). fluid consisted of 0.15 M NaCl/0.025 M sodium citrate. The Whether this variability among individuals of a species re- worms were washed by sedimentation and frozen in dry ice lates to the particular frequencies of relevant genes in a pop- in a minimal volume of perfusion fluid. ulation or to the presence of coexisting but separate subspe- Southern Blot Analysis. DNA fragments were separated by cies is not known. The use of population-specific markers to agarose gel electrophoresis after cleavage by the appropriate trace variants both geographically and through genetic cross- restriction endonuclease, transferred from the gel to nitro- es should help clarify these points. cellulose (10), and hybridized to a 32P-labeled probe. The hy- We report here that potentially useful population-specific bridization was performed in 0.45 M NaCl/0.045 M sodium markers can be found by analyzing DNA both from strains citrate/0.08% bovine albumin/0.08% Ficoll/0.08% polyvi- and from individual parasites by a Southern blot procedure nylpyrrolidone/0.1% sodium lauryl sulfate containing 50 ,g using cloned segments of the ribosomal gene as radiolabeled of denatured and sheared salmon sperm DNA and 100 ng of probes. Although ribosomal genes are probably not linked to denatured 32P-labeled DNA probe (1 x 107 cpm) in a total genes influencing factors like pathogenicity, they still pro- volume of 30 ml for 16 hr at 670C. Filters were then washed for five 30-min periods at 52TC in 0.015 M NaCl/1.5 mM so- The publication costs of this article were defrayed in part by page charge dium citrate/0.1% sodium lauryl sulfate. payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. Abbreviations: rDNA unit, ribosomal DNA unit; kb, kilobase(s). 889 Downloaded by guest on October 1, 2021 890 Medical Sciences: McCutchan et al. Proc. Natl. Acad Sci. USA 81 (1984)

RESULTS AND DISCUSSION kb Strain Differentiation by Using DNA Markers. Strains or 1 2 3 populations of S. mansoni used for this study were isolated over a period of years from areas in which they are endemic and have been maintained under laboratory conditions since Pr 2 the date of isolation (Table 1). These strains of parasites were originally differentiated on the basis of date and place Pr 1 of isolation, snail infectivity characteristics, or both (Table 1). The method of differentiation described here results from ITS Southern blot analysis of'parasite DNA by using cloned seg- BH ments of the S. mansoni ribosomai gene (Fig. 1) as radiola- FIG. 1. Restriction map of the ribosomal gene unit of S. man- beled probes. With these particular probes we were able to soni, indicating the restriction sites used to generate the DNA identify differences characteristic of particular strains (Figs. probes for this study. Spacer regions between coding areas for ma- 2 and 3) as well as differences among individuals within a ture rRNAs are designated by NTS and ITS. Areas involved in cod- strain (Fig. 4). As described below, DNA from all S. man- ing for mature rRNA are designated by a heavy line (-). Restriction soni that we have tested contains the same predominant or nuclease cleavage sites are indicated by B (BamHI) or H (HindI11). major restriction fragment homologous to these gene seg- Probes 1-3 are designated Prl-Pr3. kb, kilobase(s). ments; in addition, there are minor, variable-sized frag- ments, referred to here as polymorphic forms, which vary 3. In this experiment, aliquots of strain-specific DNA from strain to strain and even from individual to individual. (PRC3, PR2, and PR1) have been digested with varying The pattern of polymorphic forms has remained stable in our amounts of the restriction endonuclease BamHI. Southern laboratory strains over the 3-year period of our studies. blot analysis reveals a common major fragment as well as a Southern Blot Analysis Distinguishes Strains of S. mansoni. series of minor fragments, some of which are strain specific. Variability in the length of a complete rDNA unit was mea- No differences in banding patterns are seen with varying sured by Southern blot analysis and the resulting strain-spe- amounts of enzyme, suggesting that neither incomplete di- cific DNA patterns were compared. DNAs from seven dif- gests nor contaminating nuclease activity is contributing to ferent strains were restricted with the enzyme Bgl I, which the observed polymorphism. Further, aliquots of two strains cuts each ribosomal gene unit only once, yielding unit length were mixed and codigested with BamHI as a further control fragments. Southern blot analysis of the restricted DNA by against artifacts. Southern blot analysis of these reveals the using probe 2 is shown in Fig. 2. All strains have a major expected composite banding pattern. As will be shown be- band at 10 kb, indicating homogeneity within the species among strains both from Africa and Puerto Rico. However, the complement of minor bands varied among strains. PR2 1 2 3 4 5 6 7 and PRC3 (lanes 1 and 2) are clearly distinguishable and have kb minor bands'not found in other strains. PRT3, A20K, and A22A (lanes 3, 4, and 6), a Puerto Rican and two African strains, respectively, are nearly identical. A21Z and A23CAR (lanes 5 and 7) can be distinguished from other strains but contain no bands unique to either strain. South- ern blots like those shown in Fig. 2 were analyzed with probe 3. This probe, which contains only nontranscribed spacer and therefore should not hybridize to contaminating DNA 23- (host, bacterial, etc.), gave identical results. This experiment confirms the schistosome origin of each DNA band. Measurement of unit lengths, which encompass all of the alterations in different parts of the gene, is clearly useful for strain differentiation, but slight variations in size of the entire @41 rDNA unit may be difficult to resolve technically. Thus, 9.6- * strain differences were also determined by measuring poly- morphism in a smaller segment of the ribosomal gene spanned by probe 1, which is a 3.2-kb BamHI fragment in- 6.7- cluding parts of the nontranscribed spacer area and the cod- 4if ing region for the small rRNA. This approach increased the sensitivity of our measure of strain differences. An example of the use of probe 1 for strain differentiation is shown in Fig. i Table 1. Strains of S. mansoni Strain Date isolated Locality PRI 1950 Arecibo, Puerto Rico PR2 1975 Luquillo, Puerto Rico PRC3* 1976 Boqueron, Puerto Rico PRT3* 1976 Boqueron, Puerto Rico NMRI 1940 Puerto Rico FIG. 2. Determination of ribosomal DNA unit (rDNA unit) A20K 1981 Kenya length for seven different strains of S. mansoni. DNA (0.5-1.0 ,g) A21Z 1981 Zaire from each strain was restricted with the enzyme Bgl I and electro- A22A 1981 Zaire phoresed at 2 V/cm through a 0.8% agarose (wt/vol) gel for 48 hr. A23CAR 1981 Central Mrican Republic The DNA was transferred to nitrocellulose paper and hybridized to probe 2. The following strains were tested: PR2, lane 1; PRC3, lane *PRC3 and PRT3 were isolated from the same patient and separated 2; PRT3, lane 3; A20K, lane 4; A21Z, lane 5; A22A, lane 6; and with regard to their infectivity characteristics to snails. A23CAR, lane 7. Downloaded by guest on October 1, 2021 Medical Sciences: McCutchan et aL Proc. NatL Acad. Sci. USA 81 t1984) 891

a b C entire population (Fig. 4B). The patterns cbntain most of the same bands found in the PR!, PR2, and PRC3 strains (Fig. 2), with some variation. Band d in parasites 1 and 2 is the 1 2 3 4 5 6 7 8 9 10 11 same size as the PRC3-specific band shown in Fig. 2. Para- kb site 3 has a pattern that is much the same as the PR2 popula- tion shown in Fig. 2. Parasites 3-6 have bands (c and f) that are prominent in each DNA analyzed in the experiment shown in Fig. 2. Band e in parasites 1 and 2 appears to be specific to the NMRI strain. Generally, individuals from the NMRI strain are of a small number of types, the number of polymorphic forms per individual varies only slightly with

b *F .. I probe 1, and there may be some linkage between forms. Fu- ture analyses of individuals on a larger scale should provide data that establish characteristics of a population or gene pool. Populations from the wild may prove to be even more diverse than these laboratory strains. 6.7- Analysis of Parents and Progeny of a Genetic Cross. The behavior of these markers during a genetic cross was also tested (Fig. 5). Populations of parents and F1 progeny were compared by DNA analysis as described above. Males from one population were crossed with females of another. DNA 4.2- __ analysis of the F1 population revealed a composite of the parental patterns, with one exception. One band from each strain appears to be sex linked ("SL," Fig. 5) and passed only from the female, which is the heterochromatic partner " in schistosomes (11). Other bands are strain specific and car- ried by either sex. Finally, many bands are common to both 3.2- mm- strains. No anomalies were detected in the transfer of these DNA segments to the F1 generation and they appear to seg- regate in a Mendelian fashion, although individuals of this and future generations should be tested to confirm this point. FIG. 3. Comparison of three strains of S. mansoni by using probe 1. One-microgram aliquots of S. mansoni DNA from the A 1 2 3 4 5 6 B 1 strains PR1, PR2, and PRT3 were digested with 5, 10, and 20 units of BamHI for 2 hr. DNA was electrophoresed at 2 V/cm through a 0.8% agarose gel for 48 hr, transferred to nitrocellulose paper, and kb .Am- kbl hybridized to probe 1 (Fig. 1). (A) Lanes 1-3 represent PR1 with increasing concentrations of restriction nuclease. (B) Lanes 5-7 rep- 6.7- tesent PR2. (C) Lanes 9-11 represent PRC3. Lane 4 contains 0.5 mg -1 of PR1 DNA and 0.5 ,g of PR2 DNA, which have been mixed and 11 digested with 20 units of BamHI. Lane 8 contains 0.5 .zg of PR2 DNA and 0.5 1Lg of PRT3 DNA, which have been mixed and digest- U: ed with 20 units of BamHI. a b 6.7 - low, the measurement of polymorphism in the area spanned c,d by probe 2 was also useful for strain differentiation. The di- a versity detected by these two probes results from variation b in different areas of the ribosomal gene unit and thus these is f* '= c,d probes might profitably be used in conjunction. 4.2- *ejf Individual Parasites of a Single Strain Differ. The pattern 4.2 - ef resulting from analysis of the DNA isolated from an entire population is the composite of the patterns of its individual members. Clearly, a procedure of this type needs to be ap- plied to individual worms if it is to be of maximal epidemio- 3.2 - logical and genetic use. It is important to define the number of different types of patterns found in individuals of a popula- 3.2- tion and whether or not any polymorphic patterns are linked. We have analyzed DNA both from large populations and from individuals bf the NMRI strain. This strain has been I maintained in the laboratory on a large scale and may thus have more individual diversity than othertlaboratory strains. FIG. 4. Comparison of the restriction analysis patterns between We have found that about 1 pg of DNA can be extracted DNA extracted from individual worms and DNA extracted from a from a single adult worm, which is ample DNA for Southern large population of worms. (A) DNAs were extracted from each of blot analysis. More than 60 individual adult worms were ana- six adult male worms. They were restricted with the enzyme lyzed by using probes 1 and 2 and, although many worms BamHI, electrophoresed through a 0.8% agarose gel at 2 V/cm for 14 hr, transferred to nitrocellulose, and hybridized to probe 1. Let- were identical, there were several variant types. Analysis of ters a-f mark the positions of six variable bands. (B) DNA was ex- the DNA from each of 6 adult male individuals as well as tracted from -3,000 adult worms. A 2-pig aliquot of this DNA was from an entire population of >3,000 worms is shown in Fig. restricted with the enzyme BamHI, electrophoresed through a 0.8% 4. At least three patterns are represented in the 6 individual agarose gel at 2 V/cm for 36 hr, transferred to nitrocellulose, and worms (Fig. 4A) and the composite of these three types hybridized to probe 1. Letters a-f mark the positions of six bands would look like the pattern derived from the DNA of the that are comparable in size to the variable bands labeled a-f in A. Downloaded by guest on October 1, 2021 892 Medical Sciences: McCutchan et al. Proc. NatL Acad ScL USA 81 (1984)

PRC3 M F A 1 2 3 4 B 1 2 3 x x F M PRT3 kb kb

a b c d kb

23.9 - I

6.7 - 9.6- *_2 9.6 - I. - SL 4.2 - 6.7 - 6.7 - * % 4 a -p 4.2-9_

FIG. 6. Comparison of rDNA unit length among three different FIG. 5. Analysis of the parents and F1 progeny of a cross be- species of schistosomes. (A) DNA (0.3-1 tLg) from S. mansoni, two tween two strains of S. mansoni. One microgram of 5. mansoni strains of S. haematobium, and S. japonicum was restricted with the DNA from PRC3, PRT3, and two sets of F1 progeny was restricted enzyme Bgl I, electrophoresed at 2 V/cm through a 0.8% agarose to completion with BamHI. The DNA fragments were electropho- (wt/vol) gel for 36 hr, transferred to nitrocellulose, and hybridized resed through a 0.8% agarose gel at 2 V/cm for 36 hr, transferred to to probe 1. Lane 1, S. mansoni DNA; lane 2, S. haematobium DNA nitrocellulose paper, and hybridized to probe 3 (Fig. 1). Lanes: a, from Ghana; lane 3, S. haematobium from Egypt; and lane 4, S. PRC3; b, F1 progeny resulting from PRC3 males (M) being crossed japonicum DNA. (B) DNA (0.3-1 ,tg) from S. japonicum and two with PRT3 females (F); c, F1 progeny resulting from PRC3 females strains of S. haematobium was restricted with the enzyme BamHl, being crossed with PRT3 males; d, PRT3. SL, sex linked. electrophoresed at 2 V/cm through a 0.8% agarose (wt/vol) gel for 36 hr, transferred to nitrocellulose, and hybridized to probe 1. Lane The sex-linked bands must be investigated further, because, 1, S.japonicum DNA; lane 2, S. haematobium from Egypt; and lane 3, S. haematobium from Ghana. if this type of linkage is common, its detection would provide a procedure to distinguish the sex of cercariae before infec- add several important features to parasite differentiation: (i) tion of the vertebrate host. the polymorphic patterns detected by the analysis of DNA Comparison of Schistosome Species. We were also interest- from different stages of the life cycle are exactly alike (un- ed in whether or not this type of approach could be used to published data); (ii) polymorphic patterns of ribosomal genes distinguish species and differentiate strains within other spe- are not expected to vary either with age or with the environ- cies groups. We have analyzed DNA from each of the three ment of the parasite; (iii) the different forms of the gene vary main species of human schistosomes S. mansoni, S. haema- so widely that measurement of this single parameter usually tobium, and S. japonicum for the size of the ribosomal gene suffices to distinguish any two randomly chosen populations; and for in the repeating unit length (Fig. 6A) polymorphism (iv) certain polymorphic forms are sex linked as well as area of the nontranscribed spacer (Fig. 6B). The unit length strain specific, allowing sex to be readily defined; (v) pattern analysis was done with the restriction nuclease Bgl I, as was differences between species are different than, and much shown in Fig. 3. Restriction mapping of the genes from each greater than, differences found among strains of a single spe- species has confirmed that BgI I cuts only once per unit cies. length in each (data not shown). Two features are apparent. This approach must be extended to use of schistosome (i) The major unit length fragment produced by restriction is DNA from eggs and from infected snails gathered from the different in each species; differences in these three species field. Perhaps the most important ramification of this work is then are reflected in a change in the size of the major gene that it will permit epidemiologists to distinguish different par- unit; and (ii) DNA from each species contains an array of asites of a single species and trace their origin. Such studies minor polymorphic forms, which are of potential use in should allow one to test whether or not differences in host strain differentiation. The measurement of polymorphism in range, pathogenicity, or drug resistance are related to the the nontranscribed spacer confirms this point (Fig. 6B). subspecies variation of the parasite. These results indicate that our rDNA probe will be useful for strain differentiation in other schistosome species. We thank Dr. Francine Eden and Dr. Allen Cheever for critical Usefulness and Potential Applications of the DNA Marker review of the manuscript. Approach. This work clearly establishes that schistosomes 1. Basch, P. F. (1976) Exp. Parasitol. 39, 150-169. can be distinguished from each other by using molecular 2. Richards, C. S. (1975) J. Parasitol. 61, 233-236. probes. We have found species-, strain-, and sex-specific 3. Yarinsky, A. (1978) Proc. Int. Conf. 1, 259- DNA differences in laboratory strains. This approach will 269. Downloaded by guest on October 1, 2021 Medical Sciences: McCutchan et al. Proc. NatL. Acad. Sci. USA 81 (1984) 893

4. Fletcher, M., Woodruff, D. S., LoVerde, P. T. & Asch, H. L. 7. Kassim, 0. O., Cheever, A. W. & Richards, C. S. (1979) Exp. (1980) in The Mekong Schistosome, eds. Bruce, J. I., Sorn- Parasitol. 48, 220-224. mani, S., Asch, H. L. & Crawford, K. A. (Whitmore Lake, 8. Simpson, A. J. G., Sher, A. & McCutchan, T. F. (1982) Mol. Michigan), Malacological Review, Suppl. no. 2, pp. 113-122. Biochem. Parasitol. 6, 125-137. 5. Fletcher, M., LoVerde, P. T. & Woodruff, D. S. (1981) Am. J. 9. Duvall, R. H. & DeWitt, W. B. (1976) Am. J. Trop. Med. Hyg. Trop. Med. Hyg. 30, 406-421. 16, 483-486. 6. Anderson, L. A. & Cheever, A. W. (1972) Bull. W. H. 0. 46, 10. Southern, E. M. (1975) J. Mol. Biol. 98, 503-517. 233-242. 11. Short, R. B. & Menzel, M. Y. (1960) J. Parasitol. 46, 273-287. Downloaded by guest on October 1, 2021