i i Supervisor: Dr. Terry W. Pearson

ABSTRACT

Procyclic culture forms of Trypanosoma brucei species and antibodies to these parasites were used in developing antibody-detection and antigen-detection assays for diagnosis of African human sleeping sickness. An agglutination assay using live procyclic trypanosomes- the Procyclic Agglutination Trypanosomiasis Test (PATT) was developed for detecting anti-trypanosome antibodies in the sera of trypanosome-infected vervet monkeys and humans. Antibodies to procyclic surface antigens were detected by the PATT in sera of vervet monkeys as early as 7 days post-infection with T. b. rhodesiense. Positive agglutination titres were obtained with sera from monkeys with active, untreated infections and with sera taken soon after successful drag cure. Similar positive agglutination results were also observed using the PATT with sera from T. b.gambiense- infected patients from Cote d'Ivoire and Sudan and with documented sera from T. b. rhodesiense—infected patients from Kenya. No agglutination reactions were observed with preinfection sera from vervet monk, ys, with sera from uninfected Canadians or with sera from Americans working in endemic areas. Together these results confirm the diagnostic value of using procyclic trypanosomes to detect anti-trypanosome antibodies in. human African sleeping sickness.

A double antibody sandwich ELISA using monoclonal antibodies and polyclonal rabbit antibodies to the surface membrane antigens of procyclic trypanosomes was developed. This assay detected circulating trypanosomal antigens in the sera of trypanosome-infected mice and in the sera from parasite-infected patients. However, limited success was obtained with this sandwich ELISA when tested on a larger repertoire of sera from infected humans. Rabbit antibodies made against whole lysates of T. b. rhodesiense procyclics were then employed in an antigen-trapping sandwich ELISA. The results demonstrated the effectiveness of this sandwich ELISA in revealing the infection status of vervet monkeys or humans infected with either T. b. rhodesiense or T. h. gambiense. Trypanosomal antigens were detected in the sera of parasitologically confirmed monkeys and patients but not in preinfection sera nor in control sera from uninfected North Americans. i i i The PATT and the sandwich ELISA exhibited higher sensitivities than the currently employed diagnostic assay for human sleeping sickness, the Card Agglutination Trypanosomiasis Test (CATT), when tested with sera of parasitologically-confirmed humans. The sandwich ELISA was superior to the antibody-detecting PATT and CATT in monitoring trypanocidal drug-treated patients. The overall sensitivity of the PATT and sandwich ELISA was 94.3% and 97.4% and the specificity was 84.5% and 95.5%, respectively. These results thus confirm the diagnostic value of these tests for the diagnosis of human African sleeping sickness.

Identification of diagnostically useful antigens was attempted in order to facilitate the adaptation of these diagnostic assays to a simpler format for field application. Pooled sera obtained from trypanosome-infected patients was used as a probe to detect trypanosome antigens separated by high performance liquid chromatography, immunoaffinity and immunoblotting techniques. Most of the antigens were detected in the higher molecular weight range (>62 Kd). Immunization of mice with the target antigens yielded six trypanosome-specific monoclonal antibodies. In a double antibody sandwich ELISA, these antibodies were successful in trapping circulating parasite antigens in sera from trypanosome-infected mice as early as 3 days post-infection. Some of these antigens have been partially biochemically characterized. Trypanosomal antigens were also detected by these antibodies in the urine of infected mice. The antigen-capture sandwich ELISA using either the selected monoclonal antibodies or the rabbit anti-procyclic whole lysate antibodies gave similar results with sera from trypanosome-infected mice, human sleeping sickness patients and uninfected humans from North America and Kenya. The results showed that these MAbs and their antigens were useful in the diagnosis of African human sleeping sickness. Examiners:

Dr. T erry^i Peaifs^, Supervisor lijche(Department of Biochemistrylijche(Department and Microbiology)

Dr. William W.lCay, Departmental lum ber (Department of Biochemistry and Microbiology)

L . Dr.7Ro'SertWTOl^sm^ E&partmental Member (Department of Biophemis^and Microbiology]

Dr^Miphael J. Ashwood-Smith, Outside Member (Djepapment of Biology)

Dr. Robot D. Burice, Outside Member (Department of Biology)

Dr. Timothy Lee, External Examiner (University of Calgary) V TABLE OF CONTENTS ABSTRACT...... ii

TABLE OF CONTENTS ...... v LIST OF TABLES ...... vii LIST OF FIGURES...... ix ACKNOWLEDGEMENTS...... xiii FOREWORD...... xiv

INTRODUCTION...... 1

CHAPTER 1 Use of Procyclic trypanosomes in an antibody detection assay for African human sleeping sickness ...... 1 Introduction ...... 40 Materials and Methods ...... 42 R esults.,...... 46 Discussion ...... 60 CHAPTER 2 Detection of circulating trypanosomal antigens by double antibody sandwich ELISA using antibodies to procyclic trypanosomes ...... 62 Introduction ...... 62 Materials and Methods ...... 65 Results...... 73 Discussion ...... 92 CHAPTER 3 Serodiagnosis of human African sleeping sickness by detection of anti-procyclic antibodies and trypanosome antigens 96 Introduction ...... 96 Materials and Methods ...... 98 Results ...... 104 Discussion ...... 143

CHAPTER 4 Identification of procyclic trypanosomal antigens that have serodiagnostic potential for human sleeping sickness ...... 149 Introduction ...... 149 Materials and Methods ...... 150 Results...... 166 Discussion ...... 236 V

DISCUSSION...... 243 LITERATURE CITED...... 255 VII LIST OF TABLES

Table 1 Measurement of anti-procyclic surface antibodies in vervet monkey sera before and during infection with T. b. rhodesiense and at various times after treatment with trypanocidal drugs ...... 52 Table 2 Total IgM and IgG levels in vervet monkey sera before and during infection with T. b. rhodesiense and after drug treatment 56 Table 3 Cellular, biochemical, parasitological and serological measurements on sera from African sleeping sickness patients from Daloa, C6te d'Ivoire ...... 57 Table 4 Competitive solid-phase radioimmunometric assay of anti- trypanosome monoclonal antibodies ...... 87 Table 5 Binding of biotin- or enzyme-labeled MAbs to trypanosomal antigens trapped by homologous or heterologous MAbs in double antibody sandwich ELISA ...... 88 Table 6 Comparison of three different enzyme assay systems in indirect ELISA...... 89 Table 7 Stability of antibody coated microtiter plates and nitrocellulose paper in ELISA and dot-blot assays after storage at different time intervals and temperatures ...... 90 Table 8 Detection of antigens or epitopes in bloodstream form and procyclic culture form trypanosomes by ELISA ...... 91 Table 9 Measurement of anti-procyclic antibodies and circulating trypanosomal antigens in vervet monkey sera before and during infection with T. b. rhodesiense and at various times after treatment with trypanocidal drugs ...... 122 Table 10 Detection of anti-trypanosome antibodies and trypanosome antigens in sera of trypanosome-infected Kenyans before drug treatment and at the time of relapse ...... 126

Table 11 Measurement of anti-procyclic surface antibodies and circulating trypanosomal antigens in sera of trypanosome-infected patients before and at various times after trypanocidal drug treatment 127 Table 12 Serological measurements on sera from African sleeping sickness patients from Daloa, Cote d'Ivoire, using the PATT, the CATT and the double antibody sandwich ELISA ...... 129 Table 13 Cellular, biochemical, parasitological and serological measurements on sera from trypanosome-infected Sudanese patients ...... 132 VIII Table 14 Cellular, biochemical, parasitological and serological measurements c n sera from trypanocidal drug treated human sleeping sickness patients from Sudan ...... 134 Table 15 Cellular, biochemical, parasitological and serological measurements on sera from serologically positive (CATT) patients from Sudan ...... 137 Table 16 Cellular, biochemical, parasitological and serological measurements on sera from uninfected Sudanese, uninfected North Americans and human sleeping sickness patients from Daloa, Cote d'Ivoire ...... 139 Table 17 Detection of anti-trypanosome antibodies and circulating trypanosomal antigens in sera of trypanosome-confirmed, trypanocidal drug-treated or serologically CATT positive Sudanese 141 Table 18 Measurement of antibodies and antigens in sera from patients with different parasitic diseases ...... 142 Table 19 Binding characteristics of selected monoclonal antibodies to detergent lysates of Trypanosoma species and Leishmania species. 228 Table 20 Detection of trypanosomal antigens by anti-PCF monoclonal antibodies in trypanosome supernatants and pellets ...... 229 Table 21 The amino acid composition of immunoaffinity purified trypanosomal antigen from MAb # 20 immunoadsorbent 230 Table 22 Amino terminal amino acid sequence of T. b. rhodesiense ViTat 1.1 PCF antigen recognized by monoclonal antibody #20 ...... 231 Table 23 The amino acid composition of immunoaffinity purified trypanosomal antigens from MAb # 148 immunoadsorbent 232 Table 24 Amino terminal amino acid sequence of T. b. rhodesiense ViTat 1.1 PCF antigen recognized by monoclonal antibody # 148 ...... 233 Table 25 Competitive solid-phase radioimmunometric assay of anti- trypanosome monoclonal antibodies...... 234 Table 26 Binding of biotin-labeled MAbs to trypanosomal antigens trapped by homologous or heterologous MAbs in double antibody sandwich ELISA...... 235 ix LISTS OF FIGURES

Figure 1 The life cycle of Trypanosoma brucei species...... 3 Figure 2 Transmission cycles involved in African human sleeping sickness...... 10 Figure 3 Maximum serum dilutions causing agglutination of trypanosomes in the Procyclic Agglutination Trypanosomiasis Test...... 50

Figure 4 The effect of adding different biodnylated anti-7, h. rhodesiense procyclic monoclonal antibodies on binding to trypanosome procyclic water lysates in indirect ELISA 7 7 Figure 5 Optimization of double antibody sandwich ELISA for detection of antigens in trypanosome water lysates or membranes 7 9 Figure 6 Detection of different solubilized extracts from Trypanosoma brucei rhodesiense bloodstream forms and procyclic culture forms using rabbit anti-PCF membrane polyclonal antibodies as 'capture' antibodies and biotinylated anti-TBRPl MAbs (247,346 and 477) as detector antibodies in a double antibody sandwich ELISA...... 8 1 Figure 7 Double antibody sandwich ELISA results and parasitemia data for sera from 7. b. rhodesiense -infected mice...... 8 3 Figure 8 Double antibody sandwich ELISA results and parasitemia data for the second set of sera from 7. b. rhodesiense -infected m ice...... 85 Figure 9 Detection of trypanosomal antigens in water lysates of parasites by double antibody sandwich ELISA ...... 110 Figure 10 Summary of antibody and antigen detection tests with respect to the infection status of vervet monkeys infected with 7. b. rhodesiense...... 112

Figure 11 Anti-trypanosome antibodies and trypanosomal antigens in sera of vervet monkeys No. 47 and No. 49 during 7. b. rhodesiense infection ...... 114 Figure 12 Detection of anti-trypanosome antibodies and trypanosomal antigens in sera from 10 Kenyan patients ...... 116 Figure 13 Measurement of anti-trypanosome antibodies and circulating trypanosomal antigens in sera from patients who relapsed shortly after trypanocidal drug treatment ...... 118 X Figure 14 Detection of anti-trypanosome antibodies and circulating trypanosomal antigens by the PATT and die sandwich ELISA in sera from eight patients who relapsed long after trypanocidal drug treatment ...... 120 Figure 15 Separation of T. b. rhodesiense ViTat 1.1 PCF crude lysate by gel permeation chromatography using a Sephacryl S-200 column ...... 174

Figure 16 Ion-exchange-HPLC profiles of pooled fractions # 40-50 from GPC...... 176 Figure 17 Separation of fractions # A-D from the ion-exchange HPLC column by a reducing 5-15 % gradient SDS-PAGE gel 178 Figure 18 Reverse phase HPLC profiles of pooled fractions (A, B and C) from the ion-exchange HPLC column ...... 179 Figure 19 Antigen profiles detected by pooled HSSS from Daloan patients (Cote d'Ivoire) in various parasite lysates using immunoblotting ...... 181 Figure 20 Antigen profiles detected by pooled HSSS from Kenyan patients in various parasite lysates using immunoblotting 132 Figure 21 SDS-PAGE silver stain profiles of T. b. rhodesiense ViTat 1.1 PCF eluted from an immunoaffinity column coupled to pooled immunoglobulins from HSSS from Kenyanpatients 183 Figure 22 Silver stained SDS-PAGE gel patterns cf extracted gel fractions used for immunization of BALB/c mice ...... i 84 Figure 23 Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmania species using anti-PCF MAb # 20...... 185

Figure 24 Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabiiized Trypanosoma and Leishmania specier using anti-PCF MAb # 236...... 187 Figure 25 Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmania species using anti-PCF MAb # 65...... 189 Figure 26 Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmania species using anti-PCF MAb # 401, 148 and 91 ...... 191

Figure 27 The Trypanosoma specific antigen profile recognized by MAb # 20 by immunoblotting ...... 193 X Figure 28 Detection of antigen in whole lysates of Trypanosoma and Leishmania species by inrimunoblotting using MAb # 148 ...... 194 Figure 29 Detection of antigen in whole iysates cf Trypanosoma and Leishmania species by immunoblotting using MAb # 236 ...... 195 Figure 30 Detection of antigen in whole lysates of Trypanosoma and Leishmania species by immunoblotting using MAb #65 ...... 196 Figure 31 Detection of antigen in whole lysates of Trypanosoma and Leishmania species by immunoblotting using MAb #91 ...... 197 Figure 32 Detection of antigen in whole lysates of Trypanosoma and Leishmania species by immunoblotting using MAb #401 ...... 198 Figure 33 Analysis of antigens Tecognized by the six selected and-FCF monoclonal antibodies after heat, protease and chemical treatm ent...... *...... 199 Figure 34 Enzyme-linked immunosorbent assay of T. b. rhodesiense ViTat 1.1 PCF after passing over Con-A agarose using MAb # 20 , 201 Figure 35 Enzyme-linked immunosorbent assay of T. b. rhodesiense ViTat 1.1 PCF after passing over Con-A agarose using MAb # 65...... 203 Figure 36 Extraction of trypanosomal antigens by various solubilization buffers...... 205 Figure 37 Enzyme-linked immunosorbent assay on T. b. rhodesiense ViTat 1.1 PCF lysates: eluted fractions from the MAb # 20 immunoadsorbent ...... 208 Figure 38 Silver stained SDS-PAGE profile of ELISA positive, affinity purified fractions from the MAb # 20 immunoadsorbent 210 Figure 39 Enzyme-linked immunosorbent assay on T. b. rhodesiense ViTat 1.1 PCF lysates: eluted fractions from the MAb # 148 immunoadsorbent...... 211 Figure 40 Silver stained SDS-PAGE profile of ELISA positive, affinity purified fractions from the MAb #148 immunoadsorbent 213 Figure 41 Binding of different biotinylated anti-trypanosome monoclonal antibodies to trypanosome procyclic water lysates in indirect ELISA...... 214

Figure 42 Detection of trypanosomal an'igens in water lysates of parasites by double antibody sandwich ELISA ...... 216 XI Figure 43 ELIS measurement of trypanosomal antigens in sera and urine from T. b. rhodesiense -infected mice...... 218 Figure 44 ELISA measurement of trypanosomal antigens in sera and urine from T. b. rhodesiense -infected mice using rabbit ;ir.ti- TBRP1 whole lysate antibodies ...... 220 Figure 45 Immunoblotting profiles of MAb # 148 on various T. brucei lysates separated on a reducing 10% SDS-PAGE gel 27" Figure 46 Immunoblotting profiles of MAb # 148 on various T. brucei species lysates separated on a non-reducing 10% SDS-PAGE gel ...... 223 Figure 47 Detection of Trypanosomal antigens in parasitologically- confirmed sleeping sickness sera from Kenyan patients using rabbit anti-trypanosome whole lysate antibodies or MAb m ixture...... 224 F gure 48 Detection of Trypanosomal antigens in uninfected human sera from Kenyans (- ve controls) using rabbit anti-trypanosome whole lysate antibodies or MAb mixture ...... 226 xiii ACKNOWLEDGEMENTS

I am grateful to Dr. ferry Peaison for his inspiring supervision throughout this project. My thanks to the staff of the Department of Biochemistry and Microbiology, specifically to Robert Beecroft, Jennifer Duggan, Armando Jardim, Sandy Kielland, dbert Lubossiere and Scott Scholz, for their advice and assistance during my study. I would also like to the thank my Doctoral Dissertation Commits: Drs. Michael Ashwood-Smith, William Kay, Robert Olafson and Robert Burke for their positive input.

My special thanks to Dr. Michael Clarke for his support and encouragement at the commencement of my graduate studies as well as to Chrystal McNabb for her assistance during one 'long' summer. I am indebted to Dr. Jennifer Richardson for kindly providing the anti-TBRPl hybridomas; to my collaborators: Drs. Paul Sayer, Tony Vervoort, Bruce Wellde and Pierre Cattand for providing sera from vervet monkeys and humans used in this study; to the B. C. government and the University of Victoria for financial aid in the form of scholarships; and to the International Development Research Center of Canada and the » UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases for funding tins project.

I am most grateful for the unfailing support of my family: Bill Conconi, Ellie Conconi, Flo Conconi and especially Diarmaid 6 Foighil. Counter-espionage, usually called simply by its initials CE, is a widely misunderstood branch of secret operations. Its pur pose is not to apprehend enemy agents; that is an aim of the security forces. It is the word 'counter' which causes the trouble, since it is generally interpreted to mean 'against'; a defensive operation against the enemy's intelligence operations. Quite to the contrary, CE is an offensive operation by using- or, more usually, attempting to use- the opposition's operations The ultimate goal of all CE operations is to penetrate the opposition's own secret operations apparatus; io become, obviously without the opposition's knowledge, an integral and functioning part of their calculations so far as intelligence is concerned, you know what he knows. You have thereby annulled, in one stroke, the value to him of his secret intelligence about you you are iu position to control his action.*

* C. Felix, The Spy and His Masters. 1 Introduction

African trypanosomiasis is a complex of diseases caused by trypanosomes, a taxon of flagellated protozoan parasites. Trypanosomiasis occurs throughout the tropical regions of Africa and the parasites are cyclically transmitted by tsetse flies ( Glossina spp.; ILRAD, 1986) to a variety cf vertebrate hosts. Different species of trypanosomes infect a great variety of feral mammals and also domesticated species including cattle, sheep, goats, pigs, horses and camels. Some also infect humans. Trypanosomiasis in domesticated African animals is caused by Trypanosoma congolense, T. vivax, T. brucei brucei, T. simiae and T. evansi. In humans, African trypanosomiasis is also known as "sleeping sickness" and the infective organisms are T. b. rhodesiense and T.b. gambiense (Hoare, 1970). Collectively, these diseases have severely hindered the economic and social development of Sub-Saharan Africa (WHO, 1979).

Presently, sleeping sickness is endemic in 36 African countries, coinciding with the geographic range of the tsetse fly vector. Recent estimates suggest that 50 million people are at risk of infection (Maurice and Pearce, 1987). Until 1979, there were approximately 10,000 new cases recorded per annum, but recent serious outbreaks in Cameroon, Sudan i- id Uganda have at least doubled the infection rate (Goodwin, 1985). Present estimates are probably low because of inadequate reporting (WHO, 1986). Sleeping sickness is often fatal if left untreated and has led to the depopulation of many parts of Africa since the beginning of the century (WHO, 1979). The recent Ugandan epidemic and outbreaks in the southern Sudan since the 1970's result from the breakdown in implementation of diagnostic and treatment facilities brought about by civil unrest (Goodwin, 1985). In the past 20 years, however, increased research funding has led to significant progress in the development of diagnostic and therapeutic technology and in our understanding of the complex molecular biology of trypanosome-mammalian host interaction. Nevertheless, no permanent solution, such as the development of a vaccine or the elimination of the vector, can be expected in the near future (WHO, 1986). Because of this, the development of effective diagnostic tests and curative therapies continue to hold the most short-term promise in controlling sleeping sickness in Africa (Maurice and Pearce, 1987). 2 Historical Perspective

Human sleeping sickness was first described in the 14th century by the Arab physician A1 Qualquashaudi (Hoeppli, 1959). However, it was not until the late 19th century that the importance of trypanosomes in this disease were recognized by Lewis and Evans in India and Bruce in Africa, all of whom gave their names to species of trypanosomes (Hoare, 1972; Goodwin, 1985). The causative agents of human African sleeping sickness, T. b. rhodesiense and T.b. gambiense, were first described in 1902 and 1903, respectively ( Hoare, 1972). Endemic foci of sleeping sickness in Western Africa and the Congo River Basin were mapped by European colonizers in the late 19th century. Sleeping sickness has been documented to the east of the African Rift Valley and in the Zambesi Basin only within the last 70 years. This does not necessarily imply that the disease originated in West Africa. The East African form of sleeping sickness caused by T. b. rhodesiense is distinct from the West African form and its relatively diffuse distribution may have prevented early observers from pinpointing endemic foci (Duggan, 1970). More recent human population migrations caused by climatic or socio-economic factors have facilitated the spread of sleeping sickness from its ancient foci to other regions of Africa (Duggan, 1970).

Developmental Cycles Of African Trypanosomes

Human-infecting Trypanosoma brucei are cyclically transmitted by tsetse fly vectors (Glossina spp.) to their mammalian hosts (Ormerod, 1976) (see Fig. 1). The parasites adapt to these distinct environments (fly and mammal) by unde going a series of morphological and metabolic metamorphoses. Three main developmental stages occur in the parasite life cycle. In the mammalian host the parasites occur predominantly in the blood, hence this form is termed the bloodstream stage. The parasite resides in two sites in the tsetse fly: the midgut and the salivary glands, each with a distinct developmental stage, respectively termed procyclic and metacyclic forms. During developmental metamorphoses, the most dramatic changes occur in the mitochondrial system and the surface membrane of the parasite (Vickerman, 1971). Both can be related to the survival mechanisms of the parasite in its different hosts. The parasite switches from a total reliance on glycolysis for its source of metabolic energy in the mammalian host to a cytochrome- mediated metabolism in the insect midgut (Vickerman et al., 1988). Alterations in Fig. 1

The life cycle of Trypanosoma brucei species (Adapted from Roditi and Pearson, 1990). 4

MET ACYCLIC LONG SLENDER BLOODSTREAM FORMS

EPIMASTIGOTE

SHORT STUMPY BLOODSTREAM FORMS

PROCYCLIC 5 molecules of the plasma membrane allow the flagellate to evade the mammalian host's specific and non-specific defense mechanisms (Vickerman and Barry, 1982). Other developmental changes occur in the parasite's endocytotic apparatus (Steiger, 1973) and in its giycosomes (Opperdoes and Borst, 1977). During its life cycle the parasite alternates between proliferative phases when it undergoes multiplication and non-proliferative periods in which it is incapable of cell division. The latter are associated with major transitions in environment, the former with the establishment of the parasite in a newly acquired environment (Vickerman, 1985).

Infection of the mammalian host is initiated by the bite of a parasitized fly vector which results in the deposition of tsetse saliva containing metncyclic trypanosomes into the dermal connective tissue of the mammal (Hoare, 1972). Parasites then enter the draining lymphatic vessels and from there access the host bloodstream where they multiply with a doubling time of circa 6 hours (Seed, 1978) as long slender bloodstream forms. These bloodstream trypanosomes display two significant specializations to their mammalian habitat. Each individual's single mitochondrion is non-functional (Vickerman, 1965) and the parasites obtain their energy solely from glycolysis which occurs in a special organelle called the glycosome (Opperdoes, 1985, ±987). Bloodstream trypanosomes survive in the mammalian host by utilizing r unique immune evasion tactic, antigenic variation (Gray, 1965; Tanner et al., 1980). Each parasite is ensheathed in a 12-15 nm thick surface coat composed of a single form of glycoprotein, the variant surface glycoprotein (VSG), (Vickerman, 1969; Cross, 1975). Upon infection, the host's immune system responds primarily to this surface antigen. However, trypanosomes, as a population, can evade the host's immune surveillance mechanism by successively changing to different and unique surface coats, a phenomenon known as antigenic variation (Turner, 1984, Haduk e t al., 1984). This antigenic switching results in a fluctuating parasitemia that often characterizes a trypanosome infection. Dividing slender parasites predominate in the ascending parasitemia and a particular antigenic type, called the homotype, forms the major part of the population (Van Meirvenne et al., 1975). The parasitemia goes into remission as trypanosomes of the predominent variable antigenic type (VAT) are eventually removed by the specific host immune response to that homotype (Seed, 1977, Hajduk and Vickerman, 1981a). However, the few slender trypanosomes that undergo antigenic variation temporarily evade the host immune response. Some of these cells transform into a non­ dividing morphologically distinct "stumpy" form and the parasitemia declines. This change from slender to stumpy bloodstream trypanosome coincides with the activation of 6 mitochondrial enzymatic pathways (Vickerman, 1965). Stumpy forms may be preadapted for further differentiation into the procyclic stage in the tsetse midgut once they are imbibed when a fly feeds on an infected mammalian host. Repeated syringe passage of bloodstream form T. brucei among mammalian hosts results in the formation of slender forms throughout the infection period (Ashcroft, 1960), and such trypanosome populations are described as being monomorphic. Tsetse flies feeding on rodents infected with monomoiphic strains of T. brucei are not capable of reinfecting other rodents, suggesting that the slender to stumpy differentiation is necessary for trypanosome survival in the insect and, hence, completion of the life cycle (Ashcroft, 1960; Hadjuk and Vickerman, 1981).

Trypanosome-infected blood is ingested by the tsetse into its crop and from there passes to the lumen of the midgut, which is the site of the transformation from stumpy bloodstream to procyclic insect form (Vickerman, 1985). Most ingested slender bloodstream forms are probably unable to develop into procyclics and thus die (Wijers and Willett, 1960). Transformation takes place in the posterior part of the midgut in the endoperitrophic space, i.e., inside the peritrophic membrane that separates the bloodmeal from the midgut epithelium (Evans and Ellis, 1983). Procyclic development is associated with a loss of the VSG coat (Turner et al., 1988), cessation of endocytosis (Steiger, 1973) replication of the kinetoplast-DNA network (Hajduk etal., 1984), a switch from the utilization of glucose to ptoline as a principal energy source (Evans and Brown, 1972; Bowman and Flynn, 1976; Bienen et al., 1981), complete activation of the mitochondrion (Brown et al., 1973) and a change in glycosome morphology (Hart et al., 1984). These changes occur over a 48-72 hour period in the tsetse gut and are accompanied by active division of parasites. Comparable transformations occur in vitro when bloodstream forms are grown in various culture media at 26 °C (Brun and Jenni, 1985), but the time course is often shorter (Brown et al., 1973). Cultivated procyclic trypanosomes resemble their midgut counterparts in their morphology (Steiger, 1973), metabolism (Brown et al., 1973), antigenicity (Honigberg et al., 1976; Richardson et al., 1986), protein patterns (Pearson et al., 1987) and by il.rir capacity to produce mature infections in the tsetse fly (Evans, 1979).

Procyclic trypanosomes penetrate the peritrophic membrane of the tsetse gut to access the ectoperitrophic space (Evans and Ellis, 1983). As the parasites migrate to the proventriculus, they cease cell division and develop into elongate mesocyclic forms. The parasites again cross the peritrophic membrane and migrate via the oesophagus, proboscis 7 lumen and hypopharynx to the tsetse's salivary glands. Mesocyclic trypanosomes then attach to the microvilli of the salivary gland cells using punctate junctional complexes and resume cell division as attached epimastigote forms (Tetley and Vickerman, 1985). These events are followed by a series of morphological transformations: epimastigotes (lacking the VSG coat) lead to uncoated premetacyclics, which give rise to VSG-coated nascent metacyclics and eventually produce VSG-coated, free-swimming metacyclic trypanosomes in the salivary gland lumen (Tetley et al., 1987). Production of the VSG coat in metacyclics is accompanied by the repression of mitochondrial metabolic activity, reversion of glycosomes to form spherical organelles and the cessation of cell division (Vickerman et al., 1988).

The developmental cycle within the tsetse vector takes 3-5 weeks to complete (Vickerman et al., 1988). When innoculateti into the new environment of the mammalian host by the feeding tsetse fly, metacyclic trypanosomes transform into proliferating long slender bloodstream forms and reestablish the mammalian infection (Evans and Ellis, 1983). Whether or not the trypanosome life cycle is at all stages programmed in one direction, or if certain stages can be omitted, are still open questions. However, accumulating evidence on the transcription of the maxicircle component of kinetoplast DNA during the slender > stumpy > procyclic form transformations suggest that this developmental sequence is unidirectional. Stuart (1987) and Michelotti and Hajduk (1987) found that the 9S and 12S mitochondrial ribosomal RNA levels are 30-fold lower in slender forms than in stumpy bloodstream parasites. Transcripts from 3 other mitochondrial genes: cytochrome B and subunits I and II of cytochrome oxidase are undetectable in slender bloodstream trypanosomes but are present in stumpy forms at a similar level to that of procyclics. However, the stumpy transcripts are not translated until metamorphosis to the procyclic stage occurs. These results imply that stumpy forms are preadapted to the tsetse midgut. Recent identification of developmentally regulated proteins that appear during the course of differentiation from bloodstream to procyclic stages provide additional evidence that the trypanosomal developmental cycle may be indeed unidirectional (Richardson et al., 1988; Roditi etal., 1989b; Varner et al., 1989).

There is considerable interest in identifying the signals that trigger the large number of developmental stage transformations in the trypanosome life cycle. According to Czichos and co-workers (1986), a 10 *C temperature drop (37 *C to 27 *C) and the addition of 3 mM cis-aconitate to minimal essential medium (plus 15% heat-inactivated horse serum) are sufficient to repress the synthesis of the VSG coat and induce the in vitro transformation of bloodstream to procyclic trypanosomes. A role has been implicated for lectins in inducing developmental changes in the life cycle stages that parasitize the tsetse vector (Maudlin and Welbum, 1987,1988). Black et al. (1985) suggested that mammalian host factors may prompt the slender to stumpy transformation of bloodstream trypanosomes. The recent identification of an epidermal growth factor receptor homologue in T. brucei has raised the possibility that growth factor interactions similar to those found in mammaliancells are involved in cell growth regulation in these parasites (Hide et al., 1989).

It was long believed that trypanosomes replicated solely by binary fission (Vickerman, 1985). However, the existence of a sexual stage in the T. brucei life cycle was implicated by isoenzyme studies of natural parasite populations (Gibson et al., 1980; Tait, 1980,1983), by direct analysis of DNA complexity vborst et al., 1980) and content (Borst et al., 1982), by restriction site polymorphisms (Gibson et al., 1985) and by general protein gene product polymorphisms (Anderson et al., 1985). Genetic recombination in T. brucei was confirmed by the production of hybrid progeny when two genetically marked clones were co-transmitted through the tsetse vector (Jenni et. al., 1986; Paindavoine etal., 1986; Wells et al., 1987; Sternberg et al., 1988; Pearson and Jenni, 1989). In contrast to the malaria parasite, sexual fusion does not appear to be an obligate part of the trypanosome's development in the insect vector (Sternberg et al., 1989).

Although human tryy .osomiasis is a vector-borne disease, the transmission cycles are subject to a variety of interacting factors (biotic as well as abiotic) which influence the dynamics of transmission. Owing to the multiplicity of determinants that apply to the host, vector, and parasite, transmission cycles are inevitably complex and subject to environmentally imposed changes. In addition to the trypanosome-infected person that can serve as parasite reservoir, domestic and wild animals can also be reservoir hosts of both T. b. gambiense and T. b. rhodesiense The zoonotic nature of endemic T. b. rhodesiense. disease has long been recognized. Heisch et al. (1958), in an ethically questionable experiment, isolated T. b. rhodesiense from a naturally infected bushbuck which on inoculation into humans produced the sleeping sickness disease. Human pathogenic T. b. rhodesiense has also been isolated and identified from domestic cattle, sheep, dogs and goats (Rickman and Robson, 1970; Hawkin, 1975; Herbert etal., 1980). With the advent of new methods of characterizing trypanosomes, such as isoenzyme electrophoresis and 9 DNA analysis, parasites similar or identical to T. b. gambiense have been demonstrated in domestic animals (dogs, cattle, sheep and pigs) and game animals (kob and hartebeest) in areas endemic for human sleeping sickness (Maurice and Pearce, 1987). The prevalence of these animal trypanosome infections is difficult to estimate, since animal parasitemaia can be low with long parasitemic phases. Wild animals and some domestic animals such as cattle and pigs (Mehlitz et al., 1982) are likely important animal reservoirs for human pathogenic trypanosomes because they are usually said to be 'trypanotolerant', i.e. they do not develop the clinical signs of trypanosomiasis (Olubayo, 1978). The transmission cycles in T. b. gambiense and T. b. rhodesiense diseases thus involve man-fly-man, game- fly-man and domestic animal-fly-man cycles (Fig. 2), further complicating the epidemiological control of African human sleeping sickness.

Pathology and Immunology of African Sleeping Sickness

There are two distinct forms of African sleeping sickness in humans, each due to a subspecies of Trypanosoma brucei (Apted, 1970a). The Gambian form is caused by T.b. gambiense and the Rhodesian or East African form is caused by T.b. rhodesiense. The subspecies are morphologically indistinguishable. However, their presence in an infection can be inferred by the geographic location of the disease and by its clinical course (De Raadt and Seed, 1977). Gambian sleeping sickness is distributed mainly in Western and Central Africa and is a chronic disease which patients normally survive for 2-3 years without treatment. By contrast, the Rhodesian form is found predominantly in Eastern and Southern Africa and is an acute disease usually ending in death within 2-6 months of infection (Brown, 1983). In the absence of chemotherapy both forms of African sleeping sickness are generally fatal in humans (Poltera, 1985).

An infection can be divided into three stages according to its clinical manifestations. Initially, the parasites remain localized at the site of the tsetse bite (initial stage) producing a chancre. Subsequently, the trypanosomes become widely distributed by the host's bloodstream circulation (systemic stage) and the host displays symptoms of recurrent parasitemia. Finally the parasite invades the central nervous system and other organs (advanced stage) inducing heart and nervous system irregularities in the host (Molyneux et al., 1984; Shapiro and Pearson, 1986). 10 Fig. 2

Transmission cycles involved in human African Sleeping sickness (Adapted from Maurice find Pearce, 1987). 11

Hranan-tishu; oaa cycle

% ' $ U ' r* * ^ 0 '* - r ' V '‘ ’S' , * ' H '"" " a . , v < v V*' ' V j t J l ^ *• * ’ Vi J v**’ s * • / ,} * • 12 The first sign of a trypanosome infection is often a chancre or lesion at the site of the tsetse bite (Fairban and Godfrey, 1957: Basson et al., 1977). The chancre represents a combination of an acute inflammatory response and an immune reaction to the locally proliferating trypanosomes. It exhibits marked oedema and an intense cellular reaction which appear 5 15 days after the tsetse bite and lasts for approximately 2 weeks. Infiltration of polymorphonuclear leukocytes, small lymphocytes, lymphoblasts and macropha ges into the infection focus comprise the initial host reaction (Shapiro and Pearson, 1986). The frequency with which this chancre occurs in infections has been the subject of some controversy. It is more commonly observed in infected Europeans than in Africans and also in T.b. rhodesiense infections than in T.b. gambiense infections (Duggan and Hutchinson, 1966; Gelfand, 1966).

Trypanosomes that escape the initial host response at the chancre will migrate via local draining lymph nodes and the lymphatic system into the bloodstream (Ssenyonga and Adam, 1975). Antigenic variation is used by the bloodstream trypanosomes to evade the host's immune surveillance. This results in fluctuating parasitemia accompanied by symptoms such as intermittent fevers, headaches, joint pains, splenomegaly and lymphadenopathy (Apted, 1970a; Greenwood and Whittle, 1980). Although swelling of the posterior cervical lymph nodes is fairly characteristic of a T.b. gambiense infection, most other symptoms occur only periodically, coinciding with the increase of parasite numbeis in the circulation (Molyneux et al., 1984).

As the infection progresses, par asites migrate from the vascular system into the interstitial fluid of many organs, especially the heart and the central nervous system (W€ry et al., 1982). Dysfunction of hean, liver, lungs, kidneys and the endocrine system is frequently observed at this, the advanced stage of the infection (Apted, 1970a, 1980; Poltera etal., 1976,1977; Bafort and Schmidt, 1983; Wellde etal., 1989a). Typical histopathological features include vasculitis of the capillary vessels (Poltera and Sayer, 1983) and organ lesions which are associated with granular immunoglobulin deposits and with a marked cellular infiltration (Poltera and Cox, 1977). Lesions in the central nervous system (CNS) progressively lead to a variety of clinical manifestations: headaches, irritability', tremors, ataxia, convulsions, personality changes, somnolence, pronounced wasting and coma (Poltera, 1985). Once the trypanosomes access the CNS, death is inevitable if the patients do not receive prompt chemotherapy, which at this late stage is not always effective. 13

During the progress of the disease a state of immunodepression deveiops. As a result, the cause of death is sometimes a secondary infection such as pneumonia (Brown, 1983; Shapiro and Pearson, 1986). Profound immune system changes occur in trypanosome-infected patients. The immune system plays important roles in both the attempted control of the parasite and the pathogenesis of the disease state. An intense immunoproliferative reaction develops in most lymphoid organs during the initial and systemic stages. This is followed by, in the murine system at least, a general immunosuppression in the advanced stage which affects antibody responses as well as T- cell mediated immune responses to both the parasite and non-parasite-relaied antigens (Mansfield, 1981; Bancroft and Askonas, 1982). Hosts succumb in the long term to either very heavy parasite loads or else to secondary infections to which they become more susceptible due to their weakened irrunune function. The extent of immune disruption varies with the virulence of the trypanosome clone (Sacks et al., 1980).

Trypanosomes stimulate an intense proliferation of B- and T-cells, null cells and macrophages during the early course of an infection. B-cell proliferation occurs in the lymph nodes, bone marrow and spleen, resulting in a striking increase in levels of circulating IgM (Clarkson, 1976). Trypanosome whole lysates or membrane fractions can also elicit this reaction (Clayton et al., 1979). Some of the IgM production is non-parasite specific and may result from activated host macrophages producing factor(s) mitogenic for lymphocytes (Sacks et al., 1982). Polyclonal activation and the parasite-specific antibody response have been implicated in some aspects of the pathogenesis of trypanosomiasis. Antibodies (Abs) may play a role in causing the haemolytic anemia during the systemic stage of infection and in the immune-complex dysfunction and autoimmunity characteristic of advanced infections (Mansfield and Kreier, 1972; Lambert et a., 1981). Studies with nude mice, incapable of IgG production, indicate that the IgM isotype alone is effective in controlling parasitemia (Campbell et al., 1978). Antibodies produced against the surface antigen coat of trypanosomes are responsible for the elimination of these parasites by a combination of complement-mediated lysis (Murray and Urquhart, 1977) and antibody- dependent phagocytosis (Greenblatt et al., 1983; Ngaira et al., 1983).

During the infection, antibodies are also produced against the, trypanosome invariant plasma membrane components, nuclear and cytoplasmic constituents. A pathological effect is suspected for these Abs. Infected athymic mice do not produce Abs to trypanosome 14 non-variant molecules and do not exhibit the tissue damage and immune-complex symptoms produced in thymic-intact controls (Mansfield, 1990). In addition to parasite- specific Abs, auto-antibodies against the host's self-antigens such as complement C3, fibrinogen-fibrin degradation products (Boreham and Facer, 1974) and tissue antigens such as liver, thymus and brain (Mansfield and Kreier, 1972; Poitera et al., 1980) are also generated during polyclonal B-cell activation. Heterophile, or anti-eiythrocyte Abs and anti-DNA Abs have also been reported in cases of human trypanosomiasis (Parratt and Herbert, 1976). Although the precise contribution of tnese diverse autoantibodies to pathogenesis is unclear, they undoubtedly play some role in part of the disease syndrome, including haemolytic anemia and coagulation impairment (Parratt and Herbert, 1976).

Infection of the central nervous system (CNS) during the advanced stage of ;he disease is usually associated with a significant increase in the cerebrospinal fluid of IgM levels, lymphocytes and immune complexes formed by parasite variant antigens and Abs produced against them (Green wooc und Whittle, 1973; Lambert et al., 1981). If deposited in the brain, the immune complexes can potentially damage capillaries and lead to localized oedema (Lambert et al., 1981). Autoimmune responses in the CNS have been suggested to occur during trypanosome infections, but the inconsistent evidence for anti-neuronal anti) xxiies in experimentally infected mice (Pentreath,1989) suggests that other factors are involved in neuronal damage. Astrocytes are one of the most numerous cell types in the CNS and are closely associated with the CNS tissue/blood system interface (McCarron et al., 1985). Expressionof astrocyte surface-exposed Class II major histocompatibility antigens (Ia-determinants) is induced by activated T-cells that cross the blood/brain barrier during an infection (Wekerle et al., 1987). Pentreath (1989) hypothesizes that la-induced astrocytes can serve as antigen-presenting cells. Once trypanosomes invade the CNS, astrocytes may interact with the activated T-cells to produce cytokines. These cytokines may, in turn, induce the proliferation of lympho cytes, astrocytes or microglia, and provoke a concomitant R-cel1 response and antibody secretion. The cerebral manifestations of the disease may depend on the inflammation sites, the inteiaetions between parasites, antibodies and astrocytes, a id on the levels and types of mediator release at these sites (f entreath, 1989).

Although levels of circulating IgM remains high auring a chronic infection and some IgM against novel trypanosome antigens continues to be produced (Luckins and Mehlitz, 1976; Sacks and Askonas, 1980), the ability to mount effective, specific antibody 15 responses to new trypanosomal antigens decreases with time (Oka et al., 1984). Various mechanisms ha /e been proposed to explain the immunosuppression phenomenon in advanced stages of infection. Evidence for the involvement of suppressor cells (Pearson et al., 1978; Jayawardena et al., 1978; Kar etal., 1981, Yamamoto etal., 1985), soluble suppressor substances (Tizard et al., 1978) and differential macrophage activity (Paulnock et at., 1988) have been presented. Theories of "clonal exhaustion" (Askonas et al., 1985) or "down-regulation" via an anti-idiotypic network (Sacks, 1984) have also been proposed to account for the immune system hyporesponsiveness observed during trypanosome infections. Interestingly, drug-cure of an infected immunosuppressed host does lead to rapid recovery of immune responsiveness (Roelants et al., 1979; Clayton et al., 1980). Indeed, immunosuppression in these hosts can be reversed within a few days post­ treatment (Askonas, 1985).

Trypanosomal Antigens

The antigenic profile of African trypanosomes varies throughout their complex life cycle and involves many different parasite molecules. One group of antigens, the VSGs have received much research attention due to their central role in antigenic variation in the mammalian host. A considerable amount of knowledge has accumulated about VSG structure, synthesis and molecular genetics (Turner, 1982; Donelson, 1988). In contrast, non-VSG antigens in these parasites have been less well studied. Although their existence has long been known, only a few non-VSG antigens have recently been biochemically characterized. For example, there are a Trypanosoma brucei species-specific 30-40 kDa glycoprotein, procyclin (Roditi et al., 1987; Richardson et al., 1988) and a giycosyl- phosphatidylinositol-specific phospholipase C of Trypanosoma brucei with an apparent molecular mass o f 37-40 kDa (Bulow and Overath, 1986; Hereld et al., 1986; Fox et al., 1986). Expression of these molecules in the bloodstream and procyclic stages of trypanosomes is developmentally regulated (Richardson et al., 1986; Roditi et al., 1989; Carrington et al., 1989). 16 a) Variant Surface Glycoproteins

Franke (1905) was the first to infer immunological shifts in African trypanosomes parasitizing mammalian hosts. He observed that monkey blood, removed 2 weeks post­ infection, contained a substance that was capable of lysing the initial trypanosome innoculum but was ineffective against parasites collected later in the course of the infection. It was subsequently demonstrated that this variation was an inherent property of individual trypanosomes and not a reflection of innoculum heterogeneity (Ritz, 1916). We now know that the systemic stage of a trypanosome infection is characterized by a relapsing parasitemia in which each parasitemia peak consists predominantly of parasites with a distinct VSG coat (Vickerman and Luckins, 1969). The potential number of different VSGs expressible by individual trypanosome clones is >100 (Capbem et al., 1977) and, based on gene-counting estimates, the total repertoire may approach 1,000 (Van der Ploeg et al., 1982).

The identity of a particular bloodstream trypanosome, as determined by it's VSG, is called the "variable antigen type" (VAT). A cloned trypanosome can potentially produce a great number of VATs which constitute its VAT repertoire (WHO, 1986). In the first parasitemic wave, the mammalian host is exposed to VSG coats synthesized by the metacyclic trypanosomes in the tsetse salivary glands (Hajduk and Vickerman, 1981a). Monoclonal antibody screening of metacyclic trypanosomes in single tsetse flies revealed that they are antigenically heterogeneous, but exhibit a more limited antigenic range than bloodstream stages (Esser et al., 1982). A cloned T. b. rhodesiense stock may contain up to 16 different metacyclic VATs (Esser and Schoenbechler, 1985), all of which are also expressible as bloodstream form surface antigens (Barry, 1986).

In the mammalian host, the entire cell surface of bloodstream trypanosomes is covered by a 12-15 nm-thick coat (Vickerman, 1969; Vickerman and Luckins, 1969) consisting of circa 107 molecules of the membrane form of a variant surface glycoprotein (mf VSG; Cross, 1975; Cardoso de Almeida and Turner, 1983; Ferguson et al., 1988). Cross (1975,1984) developed techniques for the large ssale isolation of pure VSGs , allowing their biochemical and structure-function analyses. VSGs have a molecular mass of circa 60,000, the carbohydrate content ranges from 7-10% by weight (Johnson and Cross, 1977) and the polypeptide chains contain 450-500 amino acids (Turner, 1985). Although the carbohydrates are heterogeneous in structure, it is the diversity in amino acid 17 sequence that is responsible for antigenic variation (Turner, 1988). The N-terminal 350 amino acids of distinct VSG sequences differ markedly from each other (Rice-Ficht et al., 1981). However, some conservation in cysteine placement and some conservative amino acid changes within the N-terminal 30 residues have been observed (Olafson et al., 1984). Sequence homologies are also present at the C-terminal 50-100 VSG residues (Rice-Ficht et al., 1981; Holder and Cross, 1981). Two different forms of carbohydrate are present in all VSGs; a N-linked oligosaccharide(s) attaching to the nascent polypeptide chain (Strickler and Patton, 1980; McConnell et al., 1983) and a glycan that is covalently attached to the C-terminus of the protein via an ethanolamine residue (Holder, 1983). In addition, the C-terminal glycan is attached to a dimyristoyl phosphatidylinositol moiety which anchors the VSG to the plasma membrane (Ferguson et al., 1985,1988). The diacyl glycerol of the membrane-attached VSG (mf VSG) may be cleaved by a glycophosphatidylinositol-specific phospholipase C (Hereld et al., 1986). This results in the release of soluble form VSG from the plasma membrane (Gumett et al., 1986) and the exposure of an epitope at the C-terminus conser ed carbohydrate portion, also called the cross-reacting determinant (CRD), that is responsible for the immunological cross- reactivity observed among most VSGs (Holder and Cross, 1981).

Monoclonal antibodies (MAbs) have been produced against purified VSGs since 1980 and have become valuable tools in investigating VSG epitopes and their relationship to the trypanosome cell surface (Pearson et al., 1980; Miller et al., 1984a,b; Hall and Esser, 1984; Pinder etal., 1987; Clarke etal., 1987). Although all of the MAbs were specific for their immunizing VSGs, only a few of the MAbs produced bound to living trypanosomes or, more critically, neutralized infectivity (Turner, 1988). Many of the MAbs can bind to trypanosomes only after chemical fixation (Turner,1988). These data suggest the presence of cryptic antigenic determinants with variant specific sequences that are not surface-exposed (Pearson et al., 1981). The non-cryptic, surface-exposed epitopes are located within the N-terminal domain and are topographically assembled (Clarice et al., 1987).

The molecular mechanisms involved in trypanosome antigenic variation have been studied extensively using recombinant DNA technology. It is now known that the process of antigenic variation is regulated at the transcriptional level (Donelson and Rice-Ficht, 1985; Boothioyd, 1985; Borst, 1986). Genes encoding VSGs are scattered within the T. brucei genome at both intrachromosomal (Van der Ploeg et al., 1982) and telomeric 18 positions (DeLange and Borst, 1982; Williams et al., 1982). Only one VSG gene is transcribed at one time and this expressed VSG gene is usually located at one of several telomeric expression sites (Donelson and Rice-Ficht, 1985). Taree mechanisms have been proposed for the antigenic switch: telomeric activation, reciprocal telomeric exchange and gene conversion (Donelson, 1987). In the telomeric activation model, transcription is switched from one expression site to another with no apparent change in the genomic environment of these VSG genes (Williams et al., 1979; Myler et al., 1984a; Bernards, 1984). Telomeric exchange involves reciprocal recombination occurring between two telomeric VSG genes resulting in the placement of a previously inactive VSG gene into an active expression site (Pays et al., 1985). The third proposed mechanism, gene conversion, involves duplication of a telomeric or intrachromosomal basic copy (BC) VSG gene which is then reciprocally translocated with the p reviously expressed VSG gene at the active expression site (Myler et al., 1984b; Pays, 1985,*.

Despite the recent advances in our understanding of V SG genes, very little is known about the molecular mechanisms that control transcription initiation of a VSG gene. Nor do we know what triggers the switch from the tra” "cription of one VSG gene to another (Donelson, 1988). It was hypothesized that host antibodies might induce the antigenic switches in bloodstream trypanosomes (Dtvle, 1977). This was dismissed when antigenic variation was observed in the absence of specific antibodies during in vitro culture (Doyle et al., 1980). The release of soluble form VSGs from their membrane-attached anchor is catalyzed by a glycosylphosphatidylinositoi (GPI) - specific phospholipase C (PLC). This VSG lipase is a membrane-associated enzyme which cleaves the GPI- membrane anchor of the VSG molecule, forming diacylglycerol and a 1,2-cyclic phosphate on the inositol ring (Cardoso de Almeida and Turner, 1983; Fer^ -son and Cross, 1984; Ferguson et al., 1985,1988). In cell lysates of bloodstream trypanosomes the lipase is recovered in the particulate fraction and detergents are required for its solubilization (Cardoso de Almeida and Turner, 1983). Purification of this enzyme has allowed its further biochemical characterization (Hereld et al., 1986; Biilow and Overath, 1986; Fox et al., 1986). It is now known that the lipase is a non-glycosylated protein of 37-40 kDa which is highly specific for the GPI-moiety; phosphatidylinositol is only slowly cleaved and other common phospholipids do not serve as substrates. In contrast to other phospholipase C's, this GPI-specific lipase does not require Ca++ for its activity (Biilow and Overath, 1986). About 30,000 copies of this VSG lipase are present per cell, each molecule can convert circa 100 molecules of membrane form VSG to soluble form VSG 19 per minute. The GPI-specific lipase is undoubtedly responsible for the rapid release of VSG from the plasma membrane of bloodstream trypanosomes under certain circumstances (England et al., 1988). However, it is still unclear if the enzyme plays an obligatory role in the differentiation of bloodstream to procyclic forms (Biilow et al., 1989a). Using monoclonal antibodies to undetermined epitopes, Biilow et al. (1989b) have localized the enzyme, not on the plasma membrane, but predominantly on the peripheral face of intracellular vesicles of obscure function. The expression of the VSG lipase’s developmenially regulated, with bloodstream trypanosomes containing high levels of both GPI-specific mRNA and VSG lipase activity (Biilow and Overath, 1985). Procyclic forms, however, yield no detectable GPI-specific lipase activity and contain significantly reduced mRNA levels for this enzyme (Carrington et al., 1989).

b) Procyclin

Procyclin is an immunodominant species-specific glycoprotein found on the surface of Trypanosoma brucei (Richardson et al., 1986,1988; Roditi et al., 1989b). It is an unusual molecule of which approximately 40 % of the protein sequence is a glutamic acid- proline (glu-pro) dipeptide repeat (Roditi et al., 1987; Mowatt and Clayton, 1987; Richardson etal., 1988). Eight procyclin genes have now been identified in the T. brucei genome. These genes are arranged as four unlinked pairs of tandem repeats (Mowatt and Clayton, 1987). At least two closely related versions of procyclin are expressed (Roditi et al., 1987; Mowatt and Clayton, 1987; Richardson et al., 1988; Mowatt and Clayton, 1988). Both forms consist of a 31-amino acid N-terminal domain which contains a glycosylation site, followed by a (Asp-Pro )2 (Glu-Pro)22-29 sequence and a 23-amino acid C-terminal hydrophobic domain which might serve as a membrane anchor (Clayton and Mowatt, 1989). Although affinity-purified procyclin from T.b. rhodesiense procyclic culture forms has an apparent molecular mass of 30-40 Kda when run on SDS-PAGE gels, the gene nucleotide sequences predict a 11-15 Kda protein (Roditi et al., 1987; Mowatt and Clayton, 1987; Richardson et al., 1988). The discrepancy between these estimates is probably due in part to the anomalous migration of proline-rich proteins on polyacrylamide gels (Ferguson et al., 1984; Young et al., 1985) and in part to glycosylation of the polypeptide (RicharJ -on et al., 1988). The relative abundance of procyclin in procyclic 20 trypanosomes is estimated to be 0.7% of the total cell protein, or 6X106 molecules per cell (Mowatt and Clayton, 1989).

All ten monoclonal antibodies that have been produced against live T. b. rhodesiense procyclic trypanosomes can bind to procyclin (Richardson et al., 1986,1988; Roditi et al., 1989a). Some of the epitopes recognized by these antibodies have been localized using synthetic peptides which correspond to three different regions of procyclin. Four of these monoclonals bind to the glutamic acid-proline repeats while one antibody recognizes the N-terminal fragment containing amino acids 1-20 fragment and another binds to fragments having amino acids 21-35 (Richardson et al., 1988). Interestingly, although procyclin could be used to raise antibodies which bound to synthetic dipeptide repeats, the converse was not achievable (Roditi et al., 1989a). Both Roditi et a/.(1987) and Mowatt and Clayton (1987) were unable to induce antibodies to dipeptide repeats even when the dipeptides were coupled to several different protein carriers. Improper antigen presenation of the synthetic repeats may account for the failure to induce antibodies specific for the procyclin molecule.

Expression of procyclin in trypanosomes is developmentally regulated. Transformation of bloodstream trypanosomes into procyclic forms is accompanied by a rapid increase in procyclic-specific mRNA soon after cell differentiation commences. Within a few hours of these developments, a rapid increase in procyclin occurs (Roditi et al., 1989b). Two-color flow cytometry has been utilized to investigate the temporal, cell- surface expression of both procyclin and VSG. Results show that during bloodstream > procyclic transformation, loss of the VSG coat is followed by the gradual appearance of procyclin at the cell surface (Roditi et al., 1989b). The biological function of the procyclin molecule is presently unclear, although it could conceivably play a role in the trypanosome's survival in the tsetse midgut (Richardson et al., 1988) or in determining tropism within the tsetse fly (Roditi and Pearson, 1990). 21 c) Other Non-YSG Antigens

It has long been recognized that infected mammalian hosts produce antibodies against both VSG and non-variant, common parasite antigens (Gray, 1960; De Raadt, 1974b), the identities of most of these common antigens are still poorly established. Methods such as complement fixation (Schoenaers et al., 1953), agglutination (Binz, 1972), ELISA (Luckins, 1977) and immunofluorescence techniques (Sadun, 1963; Mehlitz, 1979; Shapiro and Murray, 1982) have determined the presence of antibodies to non-variant trypanosome antigens in infected mammals. Other techniques, including double immuno-diffusion (Gray, 1961; Shapiro and Murray, 1982), immunoelectrophoresis (Le Ray, 1975), counter-electroimmunophoresis (Poupin et al., 1976; Taylor and Smith, 1983), and immunoprecipitation (Shapiro and Murray, 1982; Gardiner et al., 1983), allow a crude enumeration of the different trypanosomal antigens recognized by host antibodies. These studies identified 5-7 different non-variant antigens. However, few of these were further characterized by immunoprecipitation or immunoblotting (Shapiro and Murray, 1982; Gardiner et al., 1983; Burgess and Jeirells, 1985). In one study, serum from a mouse chronically infected with T.b. gambiense recognized 7-8 antigens with molecular weights of 50-120,175 and 300 kDa, all of which could be iodinated on the surface of procyclic trypanosomes (Gardiner et al., 1983). Burgess and Jerrells (1985) also identified a low molecular weight doublet invariant antigen of approximately 22 kDa, using sera from 12 Kenyan patients infected with T.b. rhodesiense . However, none of these antigens were characterized.

Several investigators have employed mouse infection-cure regimens to identify the common antigens that elicit the host's immune responses (Campbell et al., 1981; Parish et al., 1985). In most of these experiments, VSG-specific antibodies were obtained (Pearson et al., 1980; Miller et al., 1984a, b; Hall and Esser, 1984; Pinder et al., 1987; Clarke et al., 1987), although monoclonal antibodies (MAbs) to internal non-VSG antigens were selected by screening hybridoma supernatants with acetone-fixed, or air-dried trypanosomes (Campbell et al., 1981). None of the non-variant antigens recognized by the MAbs were characterized. A 31 kDa protein, shared among bloodstream and procyclic trypanosomes of T. congolense, has been identified using MAbs produced using these immunization protocols (Parish et al., 1985). Shapiro and Murray (1982) attempted to correlate the type of trypanosomal antigens recognized with the clinical course of the disease. Sera from 18 cattle actively infected with T. brucei and from 14 drug-cured cattle were used to identify 22 the trypanosomal antigens by immunoprecipitation and SDS-pclyacrylamide gel techniques. Antibodies from the infected cattle bound to 8 protein antigens of molecular weights 20,40, 45,100,110,150,180 and 300 kDa. Only three of these antigens: 110, 150 and 180 kDa, were recognized by sera from all the recovered cattle. The role of the immune response to these antigens in host control of the disease remains obscure. Recently, a trypanosome peptidase of 60 kDa molecular weight has been found in the plasma of mice infected with T. brucei (Knowles et al., 1987) and of heifers infected with T. congolense (Knowles et al., 1989). Whether hosts respond immunologically to this trypanosome peptidase is unknown. However, it is thought that the peptidase could be involved in the pathology of the disease.

Immunization with homogenates of African trypanosomes or with purified trypanosomal components has also been used to identify non-variant antigens. While some of these studies only demonstrate the presence/absence of antigens commonly shared between the bloodstream and procyclic developmental stages (Weitz, 1960; Seed, 1963; Clarkson and Awan, 1969; Stanley et al., 1978, Beat et al., 1984), others have attempted to identify the antigens present in different trypanosomal species using immunoelectrophoresis (Le Ray et al., 1973; Le Ray, 1975; Marcus and Schwarting, 1976), lectin-blotting (Frommel et al., 1987; Balber and Ho, 1988) and 2-dimensional gel electrophoresis (Anderson et al., 1985). Immunoelectrophoresis studies have revealed that the antigenic contents of T. b. brucei and T.b. rkodesiense stocks were indistinguishable. Immunization with procyclic culture forms elicited antibodies to 34 detectable components, 5 of which were not present in bloodstream forms. Conversely, immunization with bloodstream trypanosomes elicited antibodies to 23 detectable components and only the VSGs appeared to be bloodstream-specific (Le Ray et al., 1973; Le Ray, 1975; Marcus and Schwarting, 1976). Lectin-blotting studies detected 21 concanavalin A (Con A)-binding glycoproteins that are shared among the procyclic and bloodstream forms of T.b. brucei and T.b. gambiense. Additionally, 2 bloodstream-specific, con A-binding glycoproteins (one of which was a VSG, the other being a 81 kDa non-variant glycoprotein) were detected, together with an 84 kDa, procyclic-specific glycoprotein (Frommel et al., 1987). These various glycoproteins have differential distributions within the two phases of the detergent Triton X-114 (Balber and Ho, 1988). Biochemical characterization and cellular localization of these antigens have not yet been accomplished. 23 Other investigators have attempted to characterize immunogenic trypanosomal components by establishing their location in the parasite cell body. Early studies used hyperimmune sera prepared in rabbits innoculated wi th T.b. brucei homogenates and localized trypanosomal antigens in different subcellular fractions (Brown and Williams, 1962,1964). Two antigens were detected in trypanosomal nuclei using immunoelectrophuresis. Recently McLaughlin (1982) has analyzed the immunogenic components of semi-purified African trypanosomal particle fractions. About 7-10 different antigens were detected by cross-immunoelectrophoresis using hyperimmune rabbit sera against different cellular fractions. Most of the antigens appeared to occur at the parasite flagellar pocket (McLaughlin, 1982). Of these, two principal glycoproteins (60 and 66 kDa) and four minor antigens (35-50 kDa) were distinguished (McLaughlin, 1984,1987). Immunization of mice with flagellar pocket membrane fragments yielded up to 60% protection against a challenge infection with T.b. rhodesiense bloodstream forms (McLaughlin, 1987; Olenick et al., 1988). Gardiner et al., (1983) analyzed 14 procyclic trypanosome membrane antigens using hyperimmune rabbit sera against a crude parasite membrane fraction. An 83 kDa surface membrane disposed, non-variant glycoprotein was identified in both T. brucei species and T. vivax, but despite stimulating the production of high titres of specific antibodies, no protection was afforded against tsetse-transmitted challenge with either parasite (Rovis et al., 1984).

Some African trypanosome enzymes are sufficiently different from host enzymes to be immunogenic. Antisera have been prepared against a few purified trypanosome enzymes (Risby et al., 1969). Monoclonal antibodies have also been produced that bind to cytoskeletal components of Trypanosoma brucei. These include a 55 kDa tubulin (Gallo and Anderson, 1983; Gallo et al., 1988); both high (320 kDa) and low (41 kDa) molecular weight microtubule-associated proteins (Schneider etal., 1988a, 1988b); spectrin-like proteins (Schneider et al., 1988c); 68 and 72 kDa paraflagellar rod structural proteins (Gallo and Schrevel, 1985) and a 60 kDa protein that interacts with microtubules and membranes (Stieger and Seebeck, 1986; Seebeck etal., 1988). More detailed analyses of trypanosomal cytoskeletal antigens may provide new leads for chemotherapeutic attack on African trypanosomiasis. 24 Diagnosis Of Infectious Diseases, including Parasites

Diagnostic principles for infectious diseases include the etiologic diagnosis of disease by isolating and identifying the infectious agent and the demonstration of immunological responses (e.g. antibody or skin reactivity) in the patients. The rational selection of drug therapy is often based on these results (Jawetz et al., 1987). Although the clinical syndrome is important in providing information about the infectious agent and, sometimes, in rendering a tentative diagnosis, conclusive results still rely on the direct demonstration of pathogens or their derived products in samples obtained from patients (Gibbons et al., 1985). Because pathogens can grow or die depending on their environment, and can be found in different anatomic sites, body fluids and tissues during the course of an infection, the selection, timing and method of collection and handling of diagnostic samples are crucial for the accuracy of many diagnostic tests (Jawetz et al., 1987).

Demonstration of the infectious agent usually involves microscopical examination of fresh, often stained samples, or preparation of cultures under conditions favoring growth of a wide variety of organisms, including those suspected on clinical grounds. Further biochemical characterization of the isolated microorganisms may be required for a definitive identification (Duerden et al., 1987).

Serological tests based on antigen-antibody interaction have gained increasing popularity for diagnosis of infectious diseases. Depending on the reagents used, these tests can detect either microbe-specific antibodies or microbial antigens. Detection of serum antibodies against specific pathogens is based on distinguishing the elicited host immune response to the infectious agent. Because the acquired serum antibodies may persist for months, or years, serological demonstration of pathogen-directed antibodies indicates exposure (by infection or vaccination) at some time in the past, but may have no bearing on current infection status. Using this diagnostic technique, a significant increase in pathogen-specific antibody concentration in sequential blood samples taken at 10-20 day intervals is often necessary to confirm an active current infection (Jawetz et al., 1987).

Detection of specific microbial antigens in diagnostic samples requires the use of pathogen-specific antibodies. Immunoassay specificity and sensitivity depends on the quality of the antibodies used as reagents. There are two basic detection reagents: 25 polyclonal sera and monoclonal antibodies. Conventional immunodiagnostic tests have utilized antibodies produced by hyperimmunization of animals with the molecules of interest (Hum and Chantler, 1980). Preparations used for immunization may range from crude pathogen extracts to highly purified molecules. The resultant antisera are polyclonal, containing a mixture of antibodies produced against a variety of antigenic determinants in the innoculum. Antiserum activity cannot be precisely controlled due to the fact that each is a sum of the activities of its constituent antibodies. Polyclonal antiserum activity and specificity varies with the preparation used, the animal innoculated and the time of bleeding (Allen, 1985).

The advent of monoclonal antibody technology (Kohler and Millstein, 1975) has provided the technology to produce a continuous supply of antibodies with a defined antigen specificity and of a consistency that polyclonal antibodies cannot match. The most critical aspect of applying MAbs to the diagnosis of infectious pathogens is the identification and selection of hybridomas that are producing antibodies of the desired specificity and required biochemical characteristics for a particular application (Allen, 1985). Careful selection of MAbs allows the preparation of a customized reagent of defined specificity that reacts only with the antigens of interest. Test specificity can also be diversified by mixing several MAbs, each binding to a different epitope. Consequently, MAbs are gradually replacing the use of polyclonal Abs in many serodiagnostic tests (Brooks and York, 1985).

A large number of serodiagnostic tests that detect antigen-antibody reactions have been developed for infectious diseases. They include immunofluorescence (Coons et al., 1942; Collins, 1988), enzyme-linked immunosorbent assays (ELISA; Voller et al., 1979; McLaren etal., 1981), agglutination tests (Christensen etal., 1973; McCarthy, 1985) and precipitation tests (Oudin, 1946; Ouchterlony, 1949; Asahi etal., 1977). These tests have been routinely used for infections caused by bacteria, viruses and parasites (Jawetz et al., 1987; Duerden etal., 1987).

Recently, DNA hybridization techniques using specific gene probes have also been employed for identifying infectious agents that do not grow rapidly and for diagnosing infections where the pathogens are not easily cultured. Using these molecular probes it is possible to identify enterotoxin-producing Escherichia coli in stool samples without going through laborious subculturing and toxin assays (Molsely et al., 1980,1982). Similarly, 26 rapid detection and quantification of human cytomegaloviruses (Chou and Merigan, 1983) and other viruses such as hepatitis B (Tompkins, 1985) can be accomplished through DNA hybridization techniques. Recently, the development of in vitro DNA amplification by the polymerase chain reaction (PCR) technique (Saiki et al., 1985,1988; Mullis and Faloona, 1987) promises to greatly increase the utility of specific DNA probes for diagnosis of infectious diseases. The major problem associated with DNA hybridization in diagnosis is its limited sensitivity, which can usually be attributed to either low levels of target sequence in the pathogen DNA sequence, to poor quality of this DNA, or to the miniscule amounts of pathogen DNA obtained in the diagnostic samples. The PCR technique can remove these limitations by detecting a DNA sequence that is only a very minor component in a complex mixture, e.g., one copy per 105 human cells (De Bruijn, 1988). PCR is thus several orders of magnitude more se isitive than conventional hybridization techniques (Saiki et al., 1985) and has been used for positive identification of patients infected with KTV (Kwok etal., 1987,1989) and with Chagas' disease (Sturm, 1989). The joint application of non-isotopic labelling of nucleic acid and PCR techniques will likely expand the use of DNA probes to the point where they can be applied routinely to diagnosis of various infections.

Diagnosis of tropical parasitic diseases has historically been difficult due to complex parasite life cycles. Clinically, ‘Le protean nature of these diseases allows them to mimic a variety of infectious processes. Because these diseases are often chronic, nnmunosuppression and secondary infections frequently develop. These factors further complicate the clinical profile of the pathogens and clinicians must rely heavily on laboratory diagnostic methods to support a clinical diagnosis. Classically, this is obtained by direct observation of parasite ova, cysts, larvae and other stages in faeces, blood, and biopsy material (Volier and de Savigny, 1981). In some cases, this type of technique can be effective in pinpointing the infection type, e.g., acute malaria (Volier and Houba, 1980). However, many parasites that live in tissue or blood may occur in low densities. The appearance of diagnostic stages may be sporadic and some parasites may not be demonstrable without surgery, e.g., trichinosis, toxoplasmosis, hydatid disease, toxocarosis etc. (Fleck and Moody, 1988). In others, infections may have long prepatent periods, e.g., filariasis. Sensitive diagnosis is also needed for infections of low intensity which may become life-threatening during immunosuppression, e.g., strongyloidiases (Volier and de Lavigny, 1981). In such cases serology provides the most attractive technical option. In other instances, e.g., epidemiological surveys, immunological 27 methods can be very cost-effective when comp ared to conventional parasitological techniques (Lobel and Kagan, 1978).

A large variety of immunodiagnostic tests have been developed for tropical parasitic diseases, yet few are reliable enough for routine use (Volier and de Savigny, 1981). Most problems stem from the immunological complexities of host-parasite interactions. Many parasites have a variety of life history stages within the human host, resulting in a dynamic parasite antigenic profile during the course of an infection. Some stages may be relatively nonimmunogenic and others may share antigens with different organisms. Many tests still use heterogeneous antigens which cross-react with unrelated organisms (WHO, 1975). Within endemic tropical parasite regions, the requiremenv for immunodiagnostic specificity is often increased by the prevalence of multiparasitism in infected hosts (Houba, 1980).

A number of recent advances have significantly improved our knowledge of some tropical parasite antigens, thereby facilitating the development of more effective diagnostic tests for these pathogens. Elucidation of reliable in vitro culture systems have allowed not only the mass production of parasite antigens, but also their biochemical characterization and, in some cases, insights into their function, divorced from the host (Terry, 1985). Among the important milestones are the culture of schistosomes (Clegg, 1965), the in vitro production of procyclic (Bran and Schfjnenberger, 1979) and bloodstream trypanosomes (Hirumi et al., 1977) and the culture of the bloodstream stages of the malaria parasite, Plasmodium falciparum (Trager and Jensen, 1976). Monoclonal antibody technology has also been of great value in the identification, purification and characterization of key parasite antigens involved in host-parasite interactions. Production of monospecific antibodies have undoubtedly improved both the specificity and sensitivity of immunodiagnostic tests for parasitic diseases (Bidwell and Volier, 1981; Gottstein et al., 1985,1987; Feldmeier et al., 1985; Walls and Schantz, 1986; Lai et al., 1987; Muller, et. al., 1989). Application of recombinant DNA technology all s the mass production of specific antigens for study, vaccination trials (Miller et al., 1986; Egan et al., 1987; Patarroyo et al., 1988) and for diagnostic purposes (Klinkert et al., 1988; Affranchino et al., 1989; Muller et al., 1989). Recently, the\ ">lymerase chain reaction (PCR) has also been utilized for the diagnosis of Chaga's disease (Sturm et al., 1989), leishmaniasis (Barker et al., 1986) and malaria (Delves, 1989). Adoption of these tests to an inexpensive, simple format suitable for performance by relatively untrained personnel in Third World field conditions is often difficult. However, this is the challenge that must be 28 met if these tests are to achieve routine application in the diagnosis of tropical parasitic diseases.

Diagnosis of African Sleeping SicK-^ss

Diagnosis of African human sleeping sickness has historically been difficult for the following reasons: 1) the clinical symptoms of both the Gambian and Rhodesian forms are ambiguous; 2) parasitemia fluctuation during the systemic stage of infection renders classical parasitological techniques unreliable, 3) antigenic variation of bloodstream stages complicates the development of sensitive serodiagnostic tests with standardized reagents.

Clinical symptoms in early stages of African human sleeping sickness include fever, general malaise, headache, joint pains, dizziness and local oedemea, and may not be very conspicuous, particularly in the mild Gambian form. These symptoms can also be misinterpreted, the fever as a flu virus or malarial infection, and the chancre that forms at the site of the tsetse bite is often mistaken for a boil (Brown, 1983). Swelling of the cervical lymph nodes is indicative of Gambian sleeping sickness, but these symptoms are occasionally absent in the Rhodesian form. In the early weeks of the illness many patients show no outward signs and have no conspicuous symptoms of sleeping sickness. It may ; ot be until the advanced stage of infection, with the development of tremors, unsteadiness, somnolence and other signs of central nervous system degeneration, that the distinguishing disease characteristics become clearly apparent (Apted, 1970a). Currently available drug treatment for advanced stages of infection is toxic and relatively ineffective (W&y et al., 1982).

Non-Specific Diagnostic Tests

A number of non-specific diagnostic tests, including the formol-gel reaction, erythrocyte sedimentation rate test, elevation of IgM levels and changes in cerebrospinal fluid have been used to identify patients infected with African sleeping sickness. The 29 formol-gel reaction is performed by adding one drop of 40% formalin to 1 ml of serum. A solid white gel appearing within one hour indicates a positive reaction. It was originally used for the diagnosis of Kala-azar (visceral leishmaniasis), but a high incidence of positive results in sleeping sickness patients (96-99%) attracted several investigators to evaluate tins test (Dye, 1926; Hope-Gill, 1938). The reaction is mediated by hypergammaglobulinemia and is non-specific for diagnosis of individual diseases. Not every patient w ithin early trypanosome infection gives a positive formol-gel test and in more advanced cases the test may be negative (Apted, 1970a). The erythrocyte sedimentation rate is often markedly increased in early trypanosome infections. It becomes lower as the disease advances and, after successful drug therapy, drops to within normal limits (Hollins and Lewis-Faning, 1947; Gall et al., 1957; Franco, 1960). As with me formol-gel reaction the results of the erythrocyte sedimentation rate assay results depend on high serum gamma globulin levels which are commonly, but not exclusively, found in sleeping sickness patients. Several studies using gel-diffusion techniques revealed a dramatic increase in serum IgM in trypanosome infected patients (Cunningham et al., 1967; Binz, 1972a; Whittle et al., 1977). This technique has been used for epidemological surveys of sleeping sickness (Binz et al., 1968; WHO, 1976). Dried blood samples were collected on filter paper and sent to laboratories in order to estimate IgM levels (WHO, 1979). Increased levels of low molecular mass (7S) IgM in sera and especially cerebrospinal fluid (CSF) of infected patients were shown to correlate well with the progress of the disease (Mattem, 1964, 1968; Whittle et al., 1977). However, increased IgM levels in serum are characteristic, but not specific, for African human sleeping sickness and the very early stages of trypanosome infection may not be detected by this method (Frezil et al., 1974). At advanced stages of sleeping sickness, infection of the nervous system is indicated by anomalies of thr CSF, or even earlier, of the cisternal fluid. Total protein and gamma globulin levels are considerably increased. A highly significant development is the presence of IgM, plasma cells and increased lymphocyte count'; in the CSF. "Normal" CSF has been arbitrarily defined as having not more than 5 lymphocytes/mm3 and 37 mg/100 ml of protein (as determined by a dye-binding protein assay) and an absence of trypanosomes (WHO, 1983). 30 Parasite Detection

Definitive diagnosis of African human sleeping sickness depends on demonstrating the parasite in chancre fluid, lymph fluid, blood, bone marrow or CSF (Paris etal., 1980). In early stages of the infection, parasites may be microscopically detected in smears of peripheral blood samples or in aspirates from enlarged lymph nodes. Later, they may be found mainly in CSF (Houba, 1980). In wet blood film preparations, live parasites are conspicuous by the agitation of surrounding erythrocytes caused by their flagellar movements, although care should be taken to avoid confusion with microfilariae, malaria or spirochaetes (WHO, 1979). Visualization of parasites by light microscopy can often be obscured by high numbers of erythrocytes and aerolysin toxin which lyses erythrocytes has been suggested as an aid for microscopic detection of trypanosomes in whole blood (Pearson et al., 1982). An alternative method is the examination of a Giemsa-stained thick blood film (Baker, 1970). Parasites can be concentrated from heparinized blood by haematocrit centrifugation (Woo, 1970), capillary concentration using glycerol (Walker, 1972), the darkground/phase contrast buffy coat method (Murray et al., 1977) and the miniature anion-exchange/centrifugation technique (Lanham and Godfrey, 1970; Lumsden et al., 1979), to facilitate their microscopic detection. In the haematocrit centrifugation method, trypanosomes are concentrated by centrifugation of heparinized blood in microhaematocrit capillary tubes and they become localized in the buffy coat where they can be viewed by light microscopy (Woo, 1970). Glycerol is added to whole blood to facilitate the separation of parasites from erythrocytes in the capillary tube concentration method (Walker, 1972). Darkground/phase contrast microscopy enhances the visualization of the parasites in the buffy coat (Murray et al., 1977). The effective sensitivity of these methods is approximately 5x10* trypanosomes/ml (Paris et al., 1982).

The most sensitive method for detecting small numbers of trypanosomes in blood is the DEAE cellulose anion-exchange column method (Lanham and Godfrey, 1970). Large volumes of blood can be processed using this technique, which allows trypanosomes to be selectively eluted and then concentrated by centrifugation or by membrane filtration. A miniaturized version of this system has been developed for field use (Lumsden etal., 1979) and its sensitivity is reported to be approximately 5xl03 trypanosomes/ml (Paris et al., 1982). P ecently, a quantitative buffy coat analysis (QBCA) has been developed for haematoparasites (Levine, 1989). Parasites are concentrated in the buffy coat with a specialized QBC tube and they can be visualized by fluorescence staining with acridine 31 orange. Preliminary studies using rats experimentally infected with T. b. rhodesiense indicate that a single parasite per tube (110 |il blood sample) may be detected (Levine etal., 1989).

Because of the cyclical fluctuations in parasitemia typical of the systemic stage of infection, repeated blood examination over a period of several days may be required for the microscopic detection of bloodstream trypanosomes. Parasitemia levels are generally higher in Rhodesian than in Gambian form infections and in advanced stages of the latter form, the parasitemia tends to remain below microscopically detectable levels (Van Meirvenne and Le Ray, 1985). If microscopic examination fails to locate bloodstream trypanosomes in patients suspected of infection with this parasite, inoculation of lethally irradiated laboratory test animals with the patient's blood sample is frequently used as a diagnostic backup (Houba, 1980a). T.b. rhodesiense easily infects mice and rats whereas T. b. gambiense does not grow well in these animals. Negative results therefore have no decisive value for detection of the latter parasites. In vitro culture methods can also be used for parasitological detection and both trypanosome subspecies can be grown at 25 °C, where they transform into dividing procyclic forms (Brun and Jenni, 1985). The suitability and reliability of many of the parasitological techniques in diagnosis of infections under field conditions is often questioned (Paris et al., 1980) and most simply do not meet the necessary requirements.

Serodiagnostic Tests for African Sleeping Sickness

Several serodiagnostic tests, based on detecting specific anti-trypanosomal antibodies produced by the infected host, have been developed (Aiyedum et al., 1976; Vervoort et al., 1983; Van Meirvenne and Le Ray, 1985). They include complement fixation (Schoenaers et al., 1953; Pautrizel et al., 1960), immunofluorescence (W6ry et al., 1970; Volier, 1977; Magnus et al., 1978a), precipitation (Taylor and Smith, 1983), enzyme-linked immunosorbent assay (ELISA) (Volier et al., 1975; Van Knappen et al., 1977) and direct and indirect agglutination tests (Ross, 1971; Aiyedum and Amodu, 1973; Magnus et al., 1978b; Bon£ and Charlier, 1975). Some of these tests have been applied in endemic areas with considerable success (WHO, 1981). However, their applicability has 32 been hampered by the abscence of standardized diagnostic reagents that yield reliable and reproducible results (Van Meirvenne and Le Ray, 1985).

Complement fixation tests were among the first serodiagnostic techniques introduced for African human sleeping sickness at the beginning of this century (Houba, 1980a). This method is sensitive, highly specific and is generally positive in patients with early infections (7-15 days), particularly in the Gambian form. Its sensitivity declines, however, as the disease progresses, particularly in advanced stages when the central nervous system becomes infected (Pautrizel et al., 1960). Despite its apparent value, the complement fixation assay has not been widely used in clinical or field practice, probably due to inherent technical difficulties involved in preparing suitable antigen (Apted, 1970). Another problem is that anti-complementary activity of sera from many patients limits routine applicability (Houba, 1980a). An immunofluorescent complement fixation test was developed by Perie et al. (1975) but was not widely adopted due to its technical inferiority to the indirect fluorescent antibody assays which were by then available (Houba, 1980a).

The indirect fluorescent antibody (IFA) assay (W6ry et al., 1970) uses air-dried films of T.b. rhodesiense or T.b. gambiense as a source of antigen for the detection of anti­ trypanosomal antibodies which may be anti-VSG or which may bind to non-variant antigens. IFA assays typically exhibit a 90% accuracy rate in detecting patients with active trypanosome infections. Patient sera become positive for IFA two weeks after infection, but also remain positive long after drug cure due to the persistence of anti-trypanosomal antibodies in the patient's blood. Testing of cerebrospinal fluid, rather than serum, with the IFA assay may give a higher fidelity to the patient's infection status (Nozias et al., 1975). IFA tests have been employed for mass surveys of Gambian (W6ry et al., 1970; Fr6zil et al., 1977) and Rhodesian (Wellde, 1989d) sleeping sickness. The assay can be performed on dried blood samples collected on filter paper. Infected blood smears or freeze-dried trypanosomes (Magnus et al., 1978a) may be used as an antigen source. Disadvantages of this test include the cross-reactivity observed with other protozoan infections (mainly malaria), the requirement of a fluorescence microscope which is usually not feasible in field conditions, and somewhat subjective reading of the endpoints (Volier, 1977; Magnus et al., 1978a).

An ELISA assay developed by Volier et al. (1975) used soluble antigen prepared from T.b. brucei by sonication and centrifugation to coat wells of a microtitre plate. 33 Circulating anti-trypanosomal antibodies in the patient's serum is detected following incubation on the plate and subsequent development with alkaline phosphatase-labelled anti-y globulin. ELISA is a sensitive diagnostic tool, both for Rhodesian (Volier et al., 1975,1976,1977) and Gambian sleeping sickness (Van Knappen et al., 1977). Maximum sensitivity and specificity however, largely depends on the nature of the antigen preparation (Roffi et al., 1979). Wide cross-reactivity was observed in sera from patients with Chaga's disease and leishmaniasis (Houba, 1980a), although promising results have been obtained with some purified VSG antigens (Vervoort et al., 1978).

Agglutination tests have been applied for many years as serological diagnostic aids to detect antibodies in African sleeping sickness patients by direct agglutination of bloodstream parasites. Early procedures used suspensions of formalin-fixed bloodstream trypanosomes as antigen sources and met with little success. Later assays employed living trypanosomes and yielded more promising results (Houba, 1980a). Although several modifications of these assays were developed, all suffered from low specificity and this assay method is at present mainly employed for research on antigenic variation and similar studies (Gray, 1967).

The capillary tube parasite haemagglutination assay (CTH) uses a saline extract of T.b. gambiense coupled by glutaraldehyde treatment to type O negative (Rh-) erythrocytes (Bond and Charlier, 1975). Treated erythrocytes are freeze-dried and stored in capillaries. Diluted patient serum is drawn into the capillary containing the antigen-sensitive cells and gently centrifuged, the capillary is then inverted, placed at a 45* angle and the results read after 15 minutes. In the absence of anti-trypanosomal antibodies (i.e. a test negative for infection), the pellet disintegrates and the erythrocytes spread by gravity along the wall of the tubes as a streamer 10-15 mm long. If antibodies that bind to the parasite are present, the pellet remains intact (WHO, 1986). The sedimentation rate of erythrocytes is proportional to the antibody titre. This test has been modified for use in haemagglutination trays and has been commercially developed as "Cellognost" by Behringwerke AG and as "Testryp" by Smith-Kline Ltd.. Good results were obtained when the CTH assay was used to diagnose T.b. gambiense infections in West Africa (Bond and Charlier, 1975), but a majority of sera from infected T.b. rhodesiense patients were negative (Houba, 198Ca). This result may be due to the unsuitability of the T.b. gambiense antigens used, for detection of anti-T.6. rhodesiense antibodies present in the sera of patients infected with the Rhodesian form of parasite. Improvements are therefore needed in tins test before it is of 34 general applicability in diagnosing both types of human African trypanosomiasis (Van Meirvenne and Le Ray, 1985).

Ross (1971) described a capillary tube flocculation test for practical field diagnosis of human sleeping sickness. Serum samples are mixed with a small amount of a stable suspension of sonicated T.b. brucei or T.b. rhodesiense particles in capillary tubes and the degree of agglutination produced (a function of complementary antibody titre) is read either by naked eye against a dark background or with a stereomicroscope. The simplicity of this test, in addition to its reported high specificity and sensitivity, led to its recommendation for routine diagnostic use and seroepidemiological studies (Binz, 1972b; Aiyedum and Amodu, 1973). However, in a comparative study of trypanosome diagnostic tests performed by the World Health Organization, this assay did not correlate well with other tests (WHO, 1986) and has not been widely used.

Magnus et a/.(1978b) introduced a Card Agglutination Trypanosomiasis Test (CATT) which employs a formalin treated suspension of bloodstream form trypanosomes stained with Coomassie Blue. The suspension is mixed with dilutions of test sera on plastic-coated cards and allowed to react for two minutes under rotation. Sera containing antibodies to surface trypanosome antigens agglutinate the parasites as blue particles that are macroscopically visible. Successful CATT assays take advantage of the fact that there are very few serodemes of T.b. gambiense which infect humans (Gray, 1974). These serodemes share the same set of predominant variant antigen types (VATs) that are expressed early in the course of all T. b. gambiense infections (Gray and Luckins, 1976). The corollary to this is that all infected individuals will also produce antibody to the same group of VATs. In the CATT test, trypanosome clones expressing predominant VATs are isolated and produced in bulk by infecting laboratory rodents (Van Meirvenne et al., 1977). The parasites are then fixed and stabilized, and freeze-dried for field use (Magnus et al., 1978a). CATT tests are now commercially available as field kits (CATT; RTT, Smith-Kline Ltd.). Promising results have been obtained for diagnosis of T.b. gambiense infections and an overall sensitivity of >90% has been reported. About 5% of false-positive results can generally be expected but this may be higher in certain circumstances, e.g., most sera of patients with visceral leishmaniasis cross-react with non-variant trypanosomal antigens (Van Meirvenne and Le Ray, 1985). However, the CATT test has had very limited success in identifying T.b. rhodesiense infections. This important discrepancy is probably due to the markedly greater repertoire of VATs produced by the latter subspecies during its 35 bloodstream phase (Turner, 1985). Another expression of the difference in the VAT repertoires of the two trypanosome subspecies is seen in the response of patients, infected with the respective parasite forms, to hypersensitivity reaction assays. Injection of sonicated bloodstream antigens of T. b. gambiense into the skin of Nigerian sleeping sickness patients resulted in the production of a hypersensitive skin reaction within 24 hours (Houba, 1980a). However, skin testing of patients with T. b. rhodesiense antigens was unsuccessful in East Africa (de Raadt, 1974b).

One of the major obstacles to developing reliable diagnostic tests for African human trypanosomiasis stems from the complex nature of the major surface antigens of bloodstream stages, particularly those of T.b. rhodesiense (Vervoort et al., 1983). These antigens include both non-variant and variant forms, the most notable of the latter being the VSGs. Infected hosts produce a spectrum of antibodies to the trypanosomal antigens. Because of their cell surface distribution and pronounced immunogenicity, VSG antigens are typically employed as reagents for serodiagnostic tests. Indeed, carefully selected VSG antigens can be powerful diagnostic tools, as is exemplified by the success of the CATT test in pinpointing patients producing antibodies to T. b. gambiense (Magnus et al., 1978b). The complex nature of VSG expression, however, means that the degree of specificity attained by these tests is krgely dependent on the selection of appropriate VSGs (Vervoort et al., 1983). This is a serious constraint, particularly in T.b. rhodesiense infections where the bloodstream parasites can express an undetermined large number of different VSGs. Unlike T.b. gambiense, which expresses a shared restricted repertoire of predomir int VSG types among its 2-3 serodemes, there is no evidence that T.b. rhodesiense shares predominant VSG types among its numerous serodemes (Turner, 1985). As a result, development of an effective serodiagnostic test for the Rhodesian form of African human sleeping sickness has yet to be achieved.

A major drawback of all currently available serodiagnostic assays for the African trypanosomiasis is that they are designed to detect anti-trypanosomal antibodies. Diagnostic inaccuracies stem from lapses in synchrony between the production of pathogen-directed antibodies and the actual course of the disease. There are two critical periods during an African sleeping sickness infection when currently available tests are of little value. The first is a 2-4 week time lag between the initiation of a trypanosome infection and the production of sufficient pathogen-directed antibodies to yield positive results in current assays. This initial refractory period is especially dangerous in cases of 36 T.b. rhodesiense infection, where parasites can access the central nervous system and then become life threatening, even with chemotherapy, within one month of infection (Van Meirvenne and Le Ray, 1985). A second refractory period for current serodiagnostic tests occurs after drug-treatment of infections. Patients continue to produce significant titres of anti-trypanosomal antibodies for months or years after being cured (WHO, 1979). Because they are antibody-targeted, current tests cannot distinguish recently drag-cured patients from those with active infections (Turner, 1985). This is important because a substantial number of drag-treated patients subsequently redevelop active parasitemias, either due to relapse, car to reinfection (Evans, 1981). In an effort to minimize the incidence of trypanosomiasis relapse, patients often receive prolonged treatment with anti­ trypanosomal drags. Unfortunately, these drags have severe side effects and sustained treatment is frequently counterproductive in terms of the patient's overall health (Williamson, 1976; Gutteridge, 1982).

Although some of the available serodiagnostic tests for African human sleeping sickness, notably the CATT assay, have been invaluable aids in controlling this disease, there is still much room for improvement. A sensitive and specific immunodiagnostic test that allows prompt detection of both the onset and termination of either a T.b. gambiense or T.b. rhodesiense infection is urgently needed.

Control Of African Sleeping Sickness

Numerous control measures of varying ambition and complexity have been attempted in the control of this disease (WHO, 1986). They include vaccine development, vector eradication and treatment and surveillance strategies, all aimed at reducing parasite transmission rates. Recent advances in our knowledge of the impressive complexity of parasite surface antigen variation during the bloodstream stage have severely dampened the hope of developing an effective vaccine for human African sleeping sickness (WHO, 1986).

Vector control measures aim to interrupt the completion of the parasite's life cycle. Elimination of the tsetse fly from target areas has been attempted by ground and aerial 37 spraying of insecticides, including DDT, Dieldrin and Endosulfan. These programs initially succeeded in dramatically reducing tsetse numbers, but the appearance of insecticide-resistant flies, negative effect on non-target organisms and the questionable cost-effectiveness of this control measure have made the complete removal of the tsetse vector impractical (WHO, 1986). Minimization of human-tsetse contact is probably a more realistic goal that can be achieved by avoidance of tsetse prone areas and by the placement of traps and screens in residential areas to reduce local tsetse densities (WHO, 1986). Elimination of feral mammals would reduce trypanosome reservoirs but such a drastic measure is politically unacceptable (Brown, 1983).

Due to the inherent difficulties in identifying rational targets for effective immunological prophylaxis and to the impracticality of eradicating the tsetse vector, control of human African sleeping sickness is largely dependent on the development of comprehensive surveillance and treatment capabilities (WHO, 1986). The importance of surveillance has been recently underlined by endemic sleeping sickness outbreaks in Cameroon and in Southern Sudan and Uganda, where such measures were interrupted by prolonged civil unrest (Goodwin, 1985). Due to the limitations of parasitological tests, serodiagnostic assays for trypanosomiasis are of paramount importance for both mass surveys and individual case detection (Van Meirvenne and Le Ray, 1985).

A variety of trypanocidal drugs are available to treat infected patients. Presently, chemotherapy of African sleeping sickness centers on three key drugs: pentamidine for chemoprophylaxis; suramin for the treatment of early stages of the diseases; and melarsoprol (Mel B) for the treatment of the advanced stages when trypanosomes are present in the CNS (Apted, 1970b, 1980; Williamson, 1976; Gutteridge, 1985). The mode of action of these drugs is still poorly understood. Pentamidine appears to interfere with trypanosomal DNA synthesis (Newton, 1974), while suramin may inactivate glycolytic enzymes (Fairlamb and Bowman, 1980; Hart et al., 1989). Melarsoprol, like all anti­ trypanosomal arsenicals, is believed to work pri warily by disruption of energy generation in trypanosomes. Trivalent arsenicals generally have a high affinity for sulphydryl groups, which form the active sites of many enzymes, especially kinases. Because bloodstream trypanosomes ultilize pyruvate kinase as the key enzyme for net energy production, inhibition of this enzyme activity by melarsoprol explains its anti-trypanosomal effect (Flynn and Bowman, 1974). However, all three trypanocidal drugs have serious shortcomings (Gutteridge, 1985). Pentamidine, introduced in the late 1940s (King et al., 38 1937), can be used against early T. b. gambiense infections, but it is unreliable in its effect on the Rhodesian form (Brown, 1983). Although there are relatively few side effects when pentamidine is given intramuscularly, hypoglycemia and some renal damage have been reported (Gutteridge, 1983). In addition, trypanosomes resistant to this drug have appeared (Maurice and Pearce, 1987). Suramin, introduced in the early 1920s (Kleine and Fischeer, 1922), is extremely effective in combating the initial stages of both Rhodesian and Gambian form infections (Williamson, 1976). However, it is not suitable for large- scale field use because it has to be administered intravenously and is associated with some serious side effects including shock, peripheral neuropathy, kidney damage and optic atrophy. These symptoms can be fatal in some cases (Gutteridge, 1985). Both pentamidine and suramin do not penetrate into the cerebrospinal fluid and so they cannot be used for the advanced-stages of human sleeping sickness (Maurice and Pearce, 1987). Melarsoprol, developed in the 1930s (King et al., 1937), is the only drug that is effective against all stages of the disease. However, due to its pronounced toxicity to patients, it is usually administered only in the advanced stages of infection when alternative drugs are of limited value. Adverse reactions including fever, chest and abdominal pains, disturbances of heat and smell sensation and reactive encephalopathy have been observed with Mel B treatments (Gutteridge, 1985). Mortality due to toxic side effects varies from 5-10% of Melarsporol-treated patients (Brown, 1983). Recently, two other drugs, Nifurtimor (a 5- nitrofuran) and Difluoromethylomithine (DFMO), were found suitable for treating T.b. gambiense infections, including the advanced stages of infection (Bacchi et al., 1980; Jennings, 1987). Overall, drug treatment is relatively ineffective for the later stages of African human sleeping sickness. Early diagnosis is therefore of paramount importance for obtaining a cure, especially in T.b. rhodesiense infections. Effective diagnosis also serves to pinpoint infected persons who unwittingly act as parasite reservoirs and contribute to the rapid spread of the disease. Systematic diagnostic surveys, combined with prompt drug treatment of infected individuals, will decrease the parasite reservoir in humans and reduce the incidence of human-to-human transmission. 39 Thesis objective

The objective of my thesis research was to develop effective serodiagnostic tests for both the Gambian and Rhodesian forms of African human sleeping sickness. Non-variant antigens shared among bloodstream and procyclic stages of the two Trypanosoma brucsi subspecies were targeted as being the most widely conserved parasite molecules available to host immune responses, and therefore, being of most potential diagnostic value. Research was performed to utilize these antigens, and antibodies produced against them, to develop simple, comprehensive diagnostic tests that would detect the presence of either parasite-directed antibodies or antigens of both parasite subspecies. 4 0

Chapter 1. Use of Procyclic trypanosomes in an antibody detection assay for African human sleeping sickness.

Introduction:

The diagnosis of infectious diseases has traditionally been carried out by microscopic detection of the pathogenic organisms in patients' tissues or exudates (Allen, 1985) or by the cultivation of the infecting agent in an in vitro system or laboratory animal (Yolken, 1982). Such methods are often not definitive as the number of pathogens may not be sufficiently high for detection and some pathogens can not be cultivated in generally available tissue culture or animal systems, e.g. hepatitis A and B virus; rotavirus ( Almeida et al., 1971; Purcell et. al., 1975; Kapikian et al., 1976; Fleck and Moody, 1988). Alternatively, the infectious agents can be identified by a specific antigen-antibody reaction (Rytel, 1979). Because most infectious agents have definable antigens against which antibodies can be produced, and since antigen-antibody union can be completed and measured in a relatively short period of time, it is theoretically possible to develop rapid immunoassays for determining the causative agents. Presently, antibody-detection assays are the most widely used diagnostic tools for infectious diseases, e.g., routine screening of donated blood for viral hepatitis virus (Overby, 1985) and HTV virus (Ghrayeb, et al., 1989). However, in order for these antibody-detection assays to work effectively, three conditions have to be met: 1) The host must produce an antibody response to the infectious agent; 2) assay reagents must include r athogen antigens that are recognized by the host antibodies and 3) sufficient signal must be generated by the antigen-antibody binding to allow detection.

Detection of pathogen-specific antibodies in human sera, although not necessarily indicative of active infection, has proven to be extremely useful for diagnosis of sleeping sickness caused by T. b. gambiense (WHO, 1981). The most successfully applied antibody detection assay for sleeping sickness is the Card Agglutination Trypanosomiasis Test (CATT) which depends on agglutination of fixed, stained bloodstream T. b. gambiense bearing a particul? < . ariant surface glycoprotein (VSG) (Magnus et al., 1978). Most humans parasitized with T. b. gambiense produce antibodies to this variable 4 1 antigenic type (VAT) indicating that it is usually expressed during an infection. Although the CATT test has been used successfully in the field for mass serodiagnosis of T. b. gambiense (Van Nieuenhove et al., 1983), it is not useful for diagnosis of T. b. rhodesiense infections. Antibodies to the particular T. b. gambiense VSG used are not produced in T. b. rhodesiense infections, nor has a frequently occurring VAT been found in the latter subspecies (WHO, 1981). Presently, there is no effective serodiagnostic test for Rhodesian form infections.

It has long been recognized that trypanosome-infected mammalian hosts produce antibodies against both VSG and non-variant, common parasite antigens (Gray, 1960; De Raadt, 1974a). Procyclic culture form trypanosomes are devoid of surface VSG and are indistinguishable from tsetse fly midgut procyclics by a variety of criteria (Brown et al., 1973; Honigberg et al., 1976; Richardson et al., 1986; Pearson et al., 1987). Procyclic culture forms and bloodstream forms of African trypanosomes share many antigens (Reviewed by Shapiro and Pearson, 1986), some of which are found on the surface of procyclic forms and are shown to elicit antibody responses in experimentally infected animals (Gardiner et al., 1983). The cell-surface expression of these non-variant antigens potentially make procyclic stages ideal probes for the detection of anti-trypanosome antibodies in host sera. Procyclic culture trypanosomes may therefore be useful for serodiagnosis of African sleeping sickness.

In this chapter procyclic culture forms of three T. brucei subspecies: T. b. rhodesiense, T. b. gambiense and T. b. brucei were tested for detection of antibodies produced in response to infections with T. b. rhodesiense and T. b. gambiense. We used sera from vervet monkeys experimentally infected with a strain of T. b. rhodesiense (EATRO 1989) which was originally isolated by Fink (Fink and Schmidt, 1979), and which on injection into vervet monkeys led to a disease that closely resembles human infections with T. b. rhodesiense in its clinical manifestations and in parasitological, haematological, immunological and histological parameters (Schmidt and Sayer, 1982a). Procyclic culture form agglutination assays were used to detect antibodies made to the shared non-variant antigens during infection. In addition, documented sera from 39 T. b. gambiense sleeping sickness patients were assayed. The patients were hospitalized in Daloa, Cote d’Ivoire, at the clinic of the WHO applied research center, Project de Recherches sur la Trypanosomiases (PRTC). 4 2 Materials and Methods

Parasites

The organisms used to infect vervet monkeys were T. b. rhodesiense KETRI2537 or KETRI2545, derived from EATRO 1989 (Fink and Schmidt, 1979). These parasites cause similar disease manifestations (Schmidt and Sayer, 1982a) and late phase encephalitis (Schmidt and Sayer, 1982b) that are characteristic of T. b. rhodesiense human sleeping sickness. Procyclic culture forms (PCF) were established from cloned bloodstream populations of T. b. rhodesiense ViTat 1.1 (Richardson etal., 1986), T. b. gambiense TREU 1285 (U2) (Gray, 1972) and T. b. brucei LUMP 1026 (Bienen er al., 1983) using the methods of Brun and Schonenberger (1979). Promastigotes of Leis’nmania major (NIH Seidman strain) were obtained from Dr. Neil Reiner, Vancouver and were employed as a negative control test organism. The parasites were grown at 26 °C in SM medium (Cunningham, 1973) containing 10% foetal bovine serum (FBS) and 50 pg/ml gentamycin.

Monkeys, Infection and Treatment

Vervet monkeys ( Cercopithecus aethiops) were maintained at KETRI (Kenya Trypanosomiasis Research Institute, Muguga, Kenya) for studies funded by the UNDP/World Bank/WHO Special Programme for the Chemotherapy of Human African trypanosomiasis. The infection of monkeys, their subsequent drug treatment, and the collection of blood samples for serology and haematotology were performed by P. Sayer, S. Gould, J. Waitumbi, and A. Njogu at KETRI.

Prior to infection, all monkeys had undergone quarantine for > 3 months duration. During this time the monkeys became adapted to cage life and the presence of humans. They were examined clinically and repeated checks were made for evidence of disease, including tuberculosis, intestinal protozoa and helminths, and various viral diseases, including Marburg disease, Rift Valley Fever, Ebola and Congo haemorrhagic fever.

The monkeys were infected by intravenous injection of 103 T. b. rhodesiense suspended in phosphate buffered saline containing 1% glucose (PSG). Capillary blood samples were taken daily from the ear tip and examined for trypanosomes. Every two 4 3 weeks (and occasionally, weekly) after infection the monkeys were anesthetized and examined clinically. During this examination, blood was collected for serology and haematology; electrocardiography was performed and cerebrospinal fluid collected.

The monkeys were treated with the trypanocidal drug, Suramin, as detailed in the Results section, during days 28-42 of infection. A number of monkeys required drug treatment a second or third time following relapse. Other monkeys were not treated until signs of encephalitis became apparent four to six months after the infection. Serum and CSF samples were collected at various intervals and stored at -20°C.

Patients and Laboratory Examination

Blood samples were obtained by venipuncture and cerebrospinal fluid (CSF) by lumbar puncture from 39 T. b. gambiense -infected patients hospitalized at the PRTC. The sera and CSF were examined for the cellular, biochemical, parasitological and serological parameters described in Table 3. All patients showed positive parasitemias by at least one technique and thus all were confirmed sleeping sickness cases. The lymph node biopsy was parasitologically positive in 27 patients. With the exception of the IgM and IgG quantitation, all procedures including the cellular, biochemical and parasitological analyses were performed by P. Cattand in Daloa at the PRTC. Detection of anti-trypanosome antibodies in these patient's sera was performed in Victoria on iyophilized sera shipped from Daloa.

Parasitological Techniques

Lymph nodes of patients exhibiting cervical adenopathy were punctured and biopsy material examined microscopically for trypanosomes. Lumbar punctures were used to collect CSF from patients for laboratory analyses. The presence of trypanosomes in CSF and blood from patients was determined microscopically according to methods described in the WHO manual (WHO, 1983). The double centrifugation technique (DC) was used on CSF (Cattand etal., 1987,1988), whereas the presence of trypanosomes in blood was determined by microscopic observation after concentration of parasites using the microhaematocrit centrifugation technique (MHCT) (Woo, 1971) or the mini-anion 4 4 exchange centrifugation technique (m-AECT) (Lumsden et al., 1979). The presence or absence of microfilariae in blood samples was systematically recorded.

Quantitative Analysis of IgM and IgG

Levels of IgM and IgG in vervet monkey and patient's sera were determined by radial immunodiffusion (Mancini et al., 1965) using commercially prepared plates (Diffugen, TAGO, Inc., Burlingame, CA). The immunoglobulin (Ig) levels in control sera from two uninfected North Americans were also determined. Human IgM and IgG standards (Tago, Inc., Burlingame, CA) were used as references for quantifying the immunoglobulin levels in these sera. However, these may not be ideal controls for detecting elevated monkey immunoglobulin levels. Nevertheless, relative immunoglobulin levels were accurately determined. Some sera were diluted 1/5 in saline in order to obtain accurate readings.

Immunofluorescence

Antibodies specific for the surface of T. b. rhodesiense ViTat 1.1 procyclic culture forms (PCF) were measured by indirect immunofluorescence. Procyclic trypanosomes or Leishmania promastigotes were centrifuged (1000 x g, 15 mins, 4°C) from culture medium, washed once (1000 x g, 15 mins, 4°C) with PSG and 10% heat-inactivated FBS and adjusted to 1x10s cells/ml in PSG/10% FBS. Parasite suspension (20 ill) and human or vervet sera diluted in PSG/10% FBS (50pl) were added to 12x75 mm glass test tubes. The tubes were incubated on ice for 1 hr., washed three times with 2.0 ml of PSG/10% FBS and 20 ill of fluoiesceinated, affinity-purified goat anti-human IgM or IgG (TAGO, Inc., Burlingame, California) were added. After 1 hr. incubation on ice, the parasites were washed thrice as above, resuspended in 50 ill of PSG/10% FBS and 10 ill amounts used to prepare slides for light microscopy. Immunofluorescence was observed using a Zeiss Standard binocular microscope fitted with an epi-fluorescence attachment and a Zeiss 63/1.25 oil immersion objective. 4 5 Agglutination Tests

Sera were assayed for the presence of antibodies to surface antigens of PCF of T. b. rhodesiense ViTat 1.1 or promastigote culture forms of Leishmania parasites by simple agglutination assays. The organisms were grown to log-phase, washed twice with PSG/10% FBS by centrifugation (1000 x g, 10 mins, room temperature) and adjusted to 5 x 10 7/ml in PSG/10% FBS. Within 5 mins, 25 pi amounts of the suspension were added to flat-bottom microtitration plate wells (Titertek, 96 well, Flow Laboratories) containing 25 |xl of vervet monkey or patient's sera dilutions (1/10-1/320 in doubling dilutions). Central wells contained PSG/10% FBS or sera from mice infected with T. b. rhodesiense ViTat 1.1 bloodstream forms, sera from healthy, uninfected North Americans, or sera from patients infected with other parasites such as Toxoplasma and Leishmania. The plates were shaken for 10 sec. to mix their contents and incubated for 5 mins at 26 °C. Agglutination was observed using an inverted microscope at 200x magnification and scored as 4+ (total agglutination-all organisms in one or two large aggregates) to 1+ (some agglutination- organisms in small aggregates of 3-5 organisms). Since, within about 20 mins., small amounts of aggregates were often observed in the PSG/10% FBS controls, it was important to perform the test quickly and to refer periodically to the negative controls.

In separate agglutination experiments, sera from trypanosome-infected patients from Daloa (patient Nos. D359, D367, D370, D373, D376 and D377) were tested at dilutions of 1/4 - 1/2048 in PSG. Controls included PSG/10% PBS, sera from healthy, uninfected North Americans and sera from mice infected with T. b. rhodesiense ViTat 1.1 bloodstream forms. Procyclic culture forms of the three Trypanosoma brucei subspecies, T. b. rhodesiense, T. b. gambiense and T. b. brucei, and promastigotes of Leishmania major were used as test organisms in a separate agglutination assay.

Other Assays

Total protein concentrations in CSF from T. b. gambiense-infected patients were determined using a dye binding assay (Bradford, 1976) with a commercially available reagent kit (Bio-Rad protein assay, Bio-Rad Laboratories, Richmond, CA). The total number of leukocytes in CSF was determined microscopically using a Nageotte counting chamber (WHO, 1983). The count was recorded as the number of cells/mnA 4 6 Results

Vervet Monkeys

S era were from a total of twelve vervet monkeys which were bled prior to infection, during infection with T b. rhodesiense bloodstream forms or after treatment of the infected monkeys with trypanocidal drugs. Sera were numbered according to the number of the monkey and the sampling sequence, e.g. serum 128-1 is the first sample from monkey no. 128. For ease of presentation, these monkey sera and controls were numbered consecutively as VI to V47 (Vervet monkey series) which were referred in the text The assay results are shown in Table 1.

Agglutination of procyclic culture forms of T. b. rhodesiense ViTat 1.1 was correlated to the infection status. No agglutination was observed in sera taken from monkeys prior to infection (sera nos. V3, V8, V16, V32) or sera taken long after drug treatment (days 105-933; sera nos. VI, V2, V13, V15, V24, V35, V41). In addition, T,io agglutination was seen in one serum (no. VI4) taken at 29 days since it was parasitologically confirmed or with two CSF samples, one taken during an active infection (33 days) and one taken 27 days after drug treatment of the animal (nos. V43 and V42, respectively).

Low agglutination titres (1/10-1/80) were found in sera from early untreated infections (days 7-56; sera nos. V5, V9, V I1, V12, V20, V27, V28, V33, V38, V39) and with sera taken 14 and 27 days after drug treatment of infected monkeys (sera nos. V22 and V23, respectively). Preinfection sera (nos. V19, V25, V26,V36) showed titres of 1/20,1/20,1/80 and 1/10, respectively.

High agglutination titres (1/160-1/320) were associated with active untreated infections (sera nos. V6, V7, V10, V17, V21, V28, V34 and V37) and were observed as eariy as day 14 of infection (serum V10). High titres were also seen in two monkeys after drug treatment (sera nos. V18, V29, V30 and V31). Both of these animals later relapsed. Only in one case (serum no. V40) did a high agglutination titre persist long after drug treatment (132 days since parasites were last observed). This animal became sero-negative by 156 days post-infection. 4 7 Control sera from uninfected north Americans (sera V44 and V45) or from normal uninfected BALB/c mice (serum V46) did not agglutinate procyclic trypanosomes, whereas sera from BALB/c mice infected for 21 days with T. b. rhodesiense Vitat 1.1 bloodstream form trypanosomes (serum no. V47), had agglutination titres of 1/320. All sera were also tested in agglutination assays with Leishmania major culture form promastigotes, however, no agglutination titres above 1/10 were found (data not shown).

Immunofluorescence examination showed that specific IgM levels correlated with the agglutination titres in most but not all sera. Specific IgG levels measured by immunofluorescence on living T, b. rhodesiense procyclic forms correlated poorly with agglutination titres. None of the sera tested showed any immunofluorescence reaction with the surface of L. major promastigotes.

Total IgG and IgM levels were determined by radial immunodiffusion using anti­ human Ig immunodiffusion plates and human Ig as standards. The results are summarized in Table 2. IgM levels increased dramatically in infected animals and remained elevated after drug treatment The total IgM levels therefore correlated with the agglutination titres. Mean IgG levels showed little change during infection and after drug treatment Although a few animals showed increased IgG levels, there was no obvious correlation of IgG levels with agglutination titres.

Trypanosome-Infected humans

Blood and CSF from 39 sleeping sickness patients from West Africa were examined by cellular, biochemical, parasitological and serological techniques. Results for these tests are shown in Table 3. Patients are referred to by their hospitalization file numbers with "D" added before each number (i.e. "Daloa" patients). Controls were numbered consecutively, from Del toDc5.

Of the 39 patients sampled, trypanosomes were microscopically detected in the blood of 30 individuals. Six samples could not be tested with the microhaematocrit centrifugation technique (MHCT), 24 of the remainder were positive and 9 were negative. All but 9 blood samples were tested with the mini-anion exchange centrifugation technique (m-AECT), yielding 27 positive and 3 negative samples. It should be noted that all 4 8 parasitologically positive blood samples when tested by the MHCT were also positive by the m-AECT. The three analytical parameters used for the demonstration of "late stage" sleeping sickness: presence of parasites and elevated protein and cell titres, were those adopted by the members of a WHO expert committee (WHO, 1986). "Normal" CSF has been arbitrary defined as having not more than 5 lymphocytes/mm3, 37 mg/100 ml of protein as determined by a dye-binding protein assay and an absence of trypanosomes (WHO, 1983). The presence of parasites was demonstrated in 18 CSF samples. Trypanosomes could not be detected in either the blood or the CSF of two patients (Nos. D435 and D439). These patients were, however, positive for parasites in lymph node biopsy material. Microfilariae were observed in the blood of four patients (Nos. D421, D445, D502 and D504). Thirteen CSF samples had a protein level >40 mg/100 ml and 21 CSF samples had >5 cells/mm3, thereby exceeding the generally accepted background upper limits for CSF normality. Radial immunodiffusion was used to determine the total IgG and IgM levels. The results showed an increase in both Ig levels, particularly in the levels of IgM, in sera from patients infected with trypanosomes (Table 3). Mean IgM levels in control sera were 0.85 mg/ml whereas in sleeping sickness sera, the mean level was 9.34 mg/ml. Mean IgG levels were 11.9 mg/ml in control sera and 20.59 mg/ml in sleeping sickness patients' sera.

Anti-trypanosome antibodies were detected using procyclic trypanosome agglutination assays in all sera from patients infected with T. b. gambiense. Although high agglutination titres (1/64 to 1/2048) were observed in most sera from trypanosome-infected patients, five of the sera (Nos. D376, D377, D378, D421 and D433) showed low agglutination titres (1/16 to 1/32). All five control sera (Del to Dc5) showed no agglutination. All patients' sera were additionally tested in agglutination assays with L. major promastigotes. No agglutination titres above 1/10 were found (data not shown).

Specific IgM and IgG levels measured by immunofluorescence on living T. b. rhodesiense procyclics corroborated the agglutination results in some, but not all, sera. Sera fron 37 patients gave high levels of specific IgM (>10-2 dilution), one serum (No. D458) showed low immunofluorescence titres (1/10) and one serum (D376) was negative. Immunofluorescence reactions demonstrated high specific IgG levels (>10*2 dilution) in 22 of the 39 patients' sera examined; 7 sera showed low IgG titres and 10 sera gave no immunfluorescence reaction. Low titres of specific IgM (1/10) were observed in 9 49 patients' sera using L. major promastigotes, 2 sera gave high levels (1/100) of specific IgG and 3 sera gave low IgG titres(l/10).

PCF of three subspecies of Trypanosoma brucei and promastigotes of Leishmania major were used as test organisms in a separate agglutination test. Positive agglutination of PCF of all three T. brucei subspecies was observed with sera from trypanosome-infected patients (Fig. 3) and several of the post-treatment sera from vervet monkeys and mice (data not shown), whereas no agglutination was seen with the L. major promastigotes (data not shown). The possible non-specific agglutination of trypanosomes resulting from the increased Ig levels in the sera of trypanosome-infected patients was tested. The addition of 25 mg/ml of purified human IgG or IgM to normal control sera did not cause agglutination of PCF of any of the three subspecies used. 50 Fig. 3

Maximum serum dilutions causing agglutination of trypanosomes in the Procyclic Agglutination Trypanosomiasis Test. Agglutination reactions with T. b. rhodesiense (open), T. b. brucei (solid) and T. b. gambiense (cross-hatch) are shown separately. Controls include sera from four uninfected North Americans (cl- c4), serum from an uninfected mouse (c5) and serum from a T. b. rhodesiense-irdcctsd mouse (c6). Agglutination results of test sera from six T. b. gambiense-infected human sleeping sickness patients are also presented. 1 2 8

3 2

1 8

1 2

0 Iw c1 c2 c3 c4 c5

control sera T ab le 1

Measurement of anti-procyclic surface antibodies in vervet monkey sera before and during infection with T. b. rhodesiense and at various times after treatment with trypanocidal drugs

Serum Monkey No. of No. of days Remarks Ig levels Immunofluor­ Agglut­ N o. N o. days since since drug (mg/ml) escence titerb ination (V) infection3 treatment titrec (drug in brackets) IgM IgG IgM IgG

1 34 940 859 (Mel. B.) 12.0 51.5

2 35 978 933 (Suramin/ 3.3 16.0 - - - M.K.436) 3 47-1 0 0 Preinfection N.D. N.D. N.D.N.D. 4 47-2 7 0 Pretreatment N.D. N.D. N.D. N.D. 5 47-3 14 0 Pretreatment N.D. N.D. N.D. N.D. 1/80 6 47-4 28 0 Pre treatment N.D. N.D. N.D.N.D. 1/160 7 47-5 56 0 Pretreatment N.D. N.D. N.D. N.D. >1/320 8 49-1 0 0 Preinfection N.D. N.D. N.D. N.D. 9 49-2 7 0 Pretreatment N.D. N.D. N.D. N.D. 1/20 10 49-3 14 0 Pretreatment N.D. N.D. N.D. N.D. 1/160 11 49-4 28 0 Pretreatment N.D. N.D. N.D. N.D. 1/80 12 49-5 56 0 Pretreatment N.D. N.D. N.D.N.D. 1/80

13 61 875 793 (Berenil) 528 days since last pos. 1.9 8.0 - - - (CSF) 14 85-1 392 130 (Suramin/ 29 days since last pos. N.D. N.D. N.D. N.D. M.K.436) (blood,CSF) (2nd treatment)

15 85-2 605 343 (Suramin/ 213 days since last pos. 5.6 10.7 - -- M.K.436) (blood) (2nd treatment)

cn rv> Table 1 (cont'd)

Serum Monkey No. of No. of days Remarks Ig levels Immunofluor­ Agglut­ N o. N o. days since since drug (mg/ml) escence titer** ination (V) infection3 treatment titrec (drug in brackets) IgM IgG IgM IgG

16 94-1 0 0 Preinfection N.D. N.D. N.D. N.D. 17 94-2 28 0 Pretreatment N.D. N.D. N.D. N.D. >1/320 18 94-3 352 158 (Suramin/ 11 days since last pos. 77.5 30.0 1/100 1/5 1/320 T.S.88) (3rd treatment) 19 96-1 0 0 Preinfection 2.3 15.3 1/10 • 1/20 20 96-2 14 0 Pretreatment N.D. N.D. N.D. N.D. 1/40 21 96-3 28 0 Pre treatment 35.0 30.0 1/100 - >1/320 22 96-4 58 14 (Suramin/ 1st and only treatment N.D. N.D. N.D. N.D. 1/80 T.S.88)

23 96-5 71 27 (Suramin/ 1st and only treatment 11.5 26.5 1/10 - 1/40 T.S.88) 24 96-6 202 158 (Suramin/ 1st and only treatment 1.6 26.5 1/10 - - T.S.88) 25 113 0 0 Preinfection 1.6 14.5 1/10 - 1/20 26 115-1 0 0 Preinfection 2.1 32.5 _ 1/10 1/80 27 115-2 16 0 Pretreatment N.D. N.D. N.D. N.D. 1/40

cn co Table 1 (cont'd)

Serum Monkey No. of No. of days Remarks Ig levels Immunofluor­ Agglut­ N o. N o. days since since drug (mg/ml) escence titer*5 ination (V) infection3 treatment titrec (drug in IgM IgG IgM IgG brackets) 28 115-3 33 0 Pretreatment 21.0 31.0 1/10 1/160 29 115-4 56 12 (TS.t>8) 1st treatment (relapsed N.D. N.D. N.D. N.D. 1/320 later) 30 115-5 70 26 (TS.88) 1st treatment (relapsed N.D. N.D. N.D. N.D. 1/160 later) 31 115-6 98 54(TS.88) 1st treatment (relapsed 24.0 51.5 1/10 1/10 1/320 later)

32 124-1 0 0 Preinfection 2.6 30.0 1/10 1/10 • 33 124-2 16 0 Pretreatment N.D. N.D. N.D. N.D. 1/40 34 124-3 33 0 Pretreatment 19.2 23.5 _ - 1/320

35 124-4 352 105 (TS. 88) 3rd treatment 4.1 32.5 -- - (Suramin/TS.88,1st and 2nd) 36 128-1 0 0 Preinfection 4.8 36.5 1/10 1/10 37 128-2 16 0 Pretreatment N.D. N.D. N.D. N.D. 1/320 38 128-3 33 0 Pretreatment 24.0 31.0 1/100 1/5 1/80 39 128-4 41 0 Pretreatment N.D. N.D. N.D. N.D. 1/80 40 128-5 352 278 (Suramin/ 132 days since last pos. 26.0 21.5 1/10 1/10 1/320 T.S.88) (blood) (2nd treatment)

Ol a Table 1 (cont'd)

Serum Monkey No. of No. of days Remarks Ig levels Immunofluor­ Agglut­ N o. N o. days since since drug (mg/ml) escence titerb ination (V) infection3 treatment titrec (drug in IgM IgG IgM IgG brackets) 41 128-6 366 292 (Suramin/ 146 days since last pos. N.D. N.D. N.D. N.D. - T.S.88) (blood) 42 96-CSF 71 27 (Suramin/ N.D. N.D. N.D. N.D. - T.S.88) 43 115-CSF 33 0 N.D. N.D. N.D. N.D. _

44 neg. 0 0 Human serum (North N.D. N.D. -- - American)

45 neg. 0 0 Human serum (North N.D. N.D. - - - American)

46 neg. 0 0 Mouse serum N.D. N.D. -- - (uninfected)

47 pos. 21 0 Mouse serum (infected N.D. N.D. -- 1/320 with T. h. rhodesiense) a Animals were infected with T. b. rhodesiense KETRI 2537 except No. 61 which was infected with T. b. rhodesiense KETRI 2545. b Immunofluorescence of T. b. rhodesiense ViTat 1.1 PCF. c Agglutination T. b. rhodesiense ViTat 1.1 PCF. N.D. = not done.

oi o i 5 6 Table 2.

Total IgM and IgG levels3 in vervet monkey sera before and during infection with T. b. rhodesiense and after drug treatment

Sera IgM IgG (Mean ± S.D.) (Mean ± S.D.)

Preinfection n = 5 2.7 / 0.4^ 26.1/9.1 Infection n = 4 24.8/7.1 28.9 / 3.6 Drug treated n = 10 16.8/23.1 26.3/15.5 a Determined by rad’al immunodiffusion ** ** mg/ml Table 3. Cellular, biochemical, parasitological and serological measurements on sera from African sleeping sickness patients from Oaloa, Cote d'Ivoire.

Clinic CSFa Parasito!ogye IgM** IgG** Serology file no. Cells** Total Tryp<* MHCT m- Micro­ GGP PATTf IgMS IgGS (D-) Proteinc AECT filaria

1. 359 3 21 . N.D. N.D. N.D. + 6.9 33.5 1/128 1/102 i/io2 2. 367 0 20 - + N.D. N.D. + 10.5 23.0 N.D. 1/103 1/102 3. 370 0 20 - N.D. N.D. N.D. + 5.9 28.8 N.D. 1/102 1/102 4. 373 1 12 - + + -- 10.5 18.9 N.D. 1/102 1/102 5. 376 0 33 - + N.D. N.D. + 5.9 15.0 1/32 1/10 6. 377 0 14 - N.D. N.D. N.D. + 3.5 20.2 1/16 1/IQ3* 1/102 7. 378 1 33 - 4- N.D. N.D. + 10.9 23.0 1/32 1/102 1/102 8. 383 0 22 - N.D. N.D. N.D. + 7.2 20.2 1/2048 1/103* 1/10 9. 389 0 28 - N.D. N.D. N.D. + 9.8 9.2 1/128 l/lO2 - 10. 394 10 32 - + + - + 10.1 18.9 1/256 1/103* - 11. 402 0 20 - - + - - 6.2 16.3 1/512 1/102 1/103 12. 421 0 23 - + ++ + + 11.3 13.8 1/32 1/103* - 13. 433 146 N.D. - + + - + 2.9 11.5 1/32 1/102 - 14. 435 2 22 ---- + 2.9 9.2 1/128 1/102 1/102 15. 436 0 25 -- + -- 6.9 20.2 1/64 1/102 1/102* 16o 437 4 16 -- + -- 5.9 17.6 1/256 1/102* 1/10 17. 438 2 25 + - + - + 8.3 18.9 1/512 1/102 1/102 18. 439 0 30 --- - + 20.1 30.3 1/1024 1/103 1/103 19. 445 10 28 - + + + + 9.4 21.6 N.D. 1/102 1/103

tn ant11

CSFa Parasitologye IgM*1 IgG*1 Serology

Cell Total Trypd MHCT m- Micro­ GGP PATT* IgMS IgGS Protein0 AECT filaria

6 17 + + + 6.9 21.6 1/256 1/102* 154 103 + + + - + 4.6 18.9 1/256 1/10 1/10 472 56 + N.D. N.D. - + 16.1 27.3 1/256 1/102 1/102 4 14 + + + - + 11.3 20.2 1/128 1/103 1/10 326 74 + + + - + 7.6 21.6 1/1024 1/103 1/103 \ \ l 85 + - - - - 21.6 18.9 1/1024 1/103 1/102 128 57 + + + - + 13.0 18.9 1/256 1/103* 1/103 242 42 + + + - + 5.2 18.9 1/512 1/103 1/103* 26 57 + + +++ - + 13.9 20.2 1/512 1/103 1/103* 24 26 - + +++ - - 11.3 20.2 1/512 1/103 1/103* 164 48 + - + -- 10.4 23.0 1/512 1/103 1/103* 119! 89 + + +++ - + 10.5 27.3 1/512 1/103* - 4 18 - + +++ - - 10.5 21.6 1/64 l/l 03* 1/102 6 25 + + +++ -- 2.7 16.3 1/512 1/103 - 132 51 ++ + ++ + + 15.7 30.3 1/512 1/103 1/102 382 58 ++ - + - - 14.8 20.2 1/128 1/103 - 16 38 + + +++ + + 6.9 20.2 1/256 1/102 - 78 32 + + ++ - + 14.3 28.8 1/512 1/103 * 154 65 ++ + + -- 14.8 16.3 1/128 1/103 1/10 214 80 + + +++ - - 3.5 21.6 1/64 1/10 2 1/10

Ul oo Table 3 (cont'd)

Clinic CSFa ParasitoIogye IgM*1 IgG Serology h file no. Cells** Total Tryp<* MHCT m- M icro­ GGP PATTf IgMg IgGl (D-) Protein0 AECT filaria

DC1 HC N.D. N.D. N.D. N.D. N.D. N.D. N.D. 0.9 13.7 DC2 TWP N D. N.D. N.D. N.D. N.D. N.D. N.D. 0.8 10.1 - - - DC3 Toxo. N.D. N.D. N.D.N.D.N.D. N.D. N.D. N.D. N.D. - N.D. N.D. DC4 Toxo. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. - N.D. N.D. DCS Leish. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. - N.D. N.D. a Cerebrospinal fluid obtained by lumbar puncture. I* Lymphocytes/mm^. c mg protein/dl. <* Presence of trypanosomes by double centrifugation in capillary tubes. e Parasitology MHCT: microhematocrit centrifugation technique. m-AECT: mini-Anion-Exchange centrifugation technique. Microfilaria: presence of blood microfilaria. GGP: results of microscopic examination of lymph node biopsy material; + presence of parasites; - absence of parasites; N.D., biopsy not done since no adenopathies seen. IPATT: Procyclic Agglutination Trypanosomiasis Test; agglutination assay using T. b. rhodesiense ViTat 1.1 PCF. g Antibody titres measured by immunofluoresence with procyclic trypanosomes. mg immunoglobulins/ml blood. N.D., not done. * positive reactions observed with live Leishmania major promastigotes in immunofluorescence.

cn t o 6 0 Discussion

Parasitological diagnosis of African sleeping sickness, particularly of T. b. gambiense infections, is often equivocal due to the small number of parasites that may be present in infected individuals, and to fluctuation in parasite numbers during the course of an infection (Croft, 1985). The present results showed that parasite detection was improved by both the microhaematocrit centrifugation technique and mini-anion exchange centrifugation technique. However, trypanosomes were still not identified in blood samples from some T. b. gambiense -infected patients (patients Nos. D435, D439 and D490). The positive infection status of these patients was subsequently confirmed by microscopic examination of patients' CSF or lymph node biopsy materials (Table 3).

Because procyclic forms of African trypanosomes share common antigens with bloodstream forms (reviewed by Shapiro and Pearson, 1986), it is not surprising that I detected antibodies to procyclic surface antigens in sera from mice, vervet monkeys and humans infected with bloodstream stages of the parasites. The data showed that it is possible to detect anti-procyclic antibodies in both T. b. gambiense and T. b. rhodesiense infections using PCF of T. b. rhodesiense in a simple agglutination assay and by immunofluorescence (Tables 1 and 3). The agglutination assay appears to be more sensitive than immunofluorescence in detecting anti-procyclic antibodies in sera from T. b. gambiense -infected patients. For instance, positive agglutination was obtained with all sera from trypanosome infected patients, while in the immunofluorescence assay, serum from patient No. D376 and sera from an additional ten patients (D389, D394, D421, D433, D446, D499, D501, D502, D503, D504 and D505) showed no fluorescence results. Control sera from uninfected North Americans (Del and Dc2) or sera from Toxoplasma- or Leishmania-infected patients (Dc3, Dc4 and Dc5, respectively) showed no agglutination assay with PCF (Table 3).

Anti-procyclic antibodies were detected by the agglutination assay in vervet monkey sera as early as 7 days after infection with T. b. rhodesiense. Indeed, all sera from monkeys with active, untreated infections (except one 7 day infection serum - no. V4) gave good agglutination titres as did sera taken soon after treatment of infected monkeys with trypanocidal drugs. Thus 14 and 27 days after drug treatment, animal no. 96 (sera nos. V22 and V23) showed titres of 1/80 and 1/40, respectively. Animal 94 (serum 18) was exceptional in that it produced a high agglutination titre (1/320) 158 days after drug 61 treatment (3 injections of suramin/TS 88). This serum sample was taken 11 days after parasites were last seen in the blood (147 days after the 3rd drug treatment) indicating that a relapse had occurred. Sera from animal 115 (sera nos. V29, V30 and V31) taken 12,26 and 54 days after drug treatment had high agglutination titres. This animal was treated only once with drug and also had a subsequent relapse.

In all cases but two (serum V18 - see above, and serum V40), sera taken from vervet monkeys long after successful drug treatment showed no agglutination. As early as 14 day" after treatment with trypanocidal drugs and certainly by 146 days agglutination titres dropped to zero. The only false positive agglutination reaction was with preinfection serum no. V26 which showed a titre of 1/80. All other preinfection or negative control sera showed titres of 0,1/10 or 1/20 (Table 1).

It is clear therefore, that simple agglutination assays using PCF can provide useful information regarding the infection status of vervet monkeys infected with T. b. rhodesiense. Indeed, positive agglutination reactions were found with 6/6 of animals with active trypanosome infections and in animal 96 (sera V22 and V230) taken early after successful drag cure. Negative agglutination reactions were seen with all sera taken long after drag cure except those discussed above which subsequently relapsed (Table 1). In addition, this assay is also useful in detecting T. b. gatnbiense infections of humans as demonstrated by the positive agglutination results observed with all sera obtained from infected patients (Table 3).

The presence of shared antigenic determinants between the three subspecies of Trypanosoma brucei PCF is indicated by the positive agglutination results obtained with all three subspecies using sera from infected vervet monkeys (data not shown), mice and humans (Fig. 3). In addition to its diagnostic potential for human trypanosomiasis, the procyclic agglutination trypanosomiasis test (PATT) may also be of value for the T. b. brucei -infected cattle. The simplicity of the PATT and its ability to detect both Garnbian and Rhodesian forms of sleeping sickness confer the PATT to be a viable alternative to the CATT in the diagnosis of human sleeping sickness. Further testing of well-documented sera from T. b. gambiense - and T. b. rhodesiense - infected patients and control sera from endemic areas, however, is necessary to fully assess the diagnostic value of the PATT for human African sleeping sickness. 6 2 Chapter 2. Detection of circulating trypanosomal antigens by double antibody sandwich ELISA using antibodies to procyclic trypanosomes.

Introduction

One of the major obstacles to the development of reliable diagnostic tests for African trypanosomiasis is the complex nature of the major surface antigen of bloodstream trypanosomes (Vervoort et al., 1983). The antigens of bloodstream trypanosomes consist of both non-variable and variable antigens, the most notable of the latter being the variant surface glycoproteins (VSGs). Because of their surface disposition and strong immunogenicity, VSG antigens are typically employed as reagents for serological tests. However, the complex nature of VSG expression means that the degree of specificity attained by these tests in largely dependent on the selection of the appropriate VSGs (Vervoort et al., 1983). This is a serious problem, particularly with T. b. rhodesiense infections (WHO, 1981). Utilizing non-variable antigens of trypanosomes may provide an alternative approach for the immunodiagnosis of this disease (Sailyo et al., 1980).

Chapter one presented data indicating that mice:, vervet monkeys and humans infected with either T. b. rhodesiense or T. b. gambiense can produce antibodies to proc) ic culture form (PCF) trypanosomes in their sera and that simple agglutination assays using these organisms can be used to diagnose infections. However, the persistence of anti-trypanosomal antibodies even after the elimination of parasites by trypanocidal drug therapy complic es the accurate assessment of the infection status of individuals (Luckins et al., 1978). Diagnostic tests designed to detect circulating parasite antigens may therefore provide a more accurate means of identifying active infections. Indeed, antigen detection has been shown to be useful for detection of animal trypanosomiasis (Rae and Luckins, 1984).

I report here on the development of an antigen detection method which allows immunodiagnosis of infections with both T. b. rhodesiense and T. b. gambiense, the causative agents of human sleeping sickness. The strategy was to detect antigens using monoclonal antibodies specific for non-VSG molecules of trypanosomes which might be released into the blood of infected animals after lysis of parasites by host effector 6 3 mechanisms, both induced and otherwise. Because procyclic and bloodstream trypanosomes share antigens (Shapiro and Pearson, 1986) and some of these are procyclic membrane antigens that are recognized by the antibodies produced by trypanosome-infected hosts, it seemed rational to attempt to use antibodies raised against surface molecules of procyclic culture forms of T. brucei spp. for antigen detection in sera horn infected animals. A panel of 10 MAbs which were shown to be specific for non-VSG surface molecules of T. brucei spp. (Richardson et al., 1986) and rabbit antisera to membranes of T. b. rhodesiense procyclic culture forms were used in this study. Mice infected with T. b. rhodesiense bloodstream forms were bled at different time intervals and their sera were tested in order to explore the potential of this assay as an immunodiagnostic test for African human sleeping sickness.

A solid-phase double antibody sandwich ELISA (Engvall and Perlmann, 1972; Yolken, 1980; Yolken etal., 1982; Butler etal., 1987) was used as an antigen-trapping assay. Advantages of this technique include its high sensitivity, requiring only minute quantities of test reagents and serum. It is also an objective test, is relatively simple and readily adapted to large-scale screening (Butler et al., 1987). This assay is performed by immobilizing antibodies onto a solid matrix, such as wells in plastic microtiter plates, beads or membrane filters. These antibodies bind specific antigens from test sera. Other unbound antigens are removed by washing and a second enzyme-conjugated antibody specific for a separate epitope of the target antigen is added. The antigen is now 'sandwiched' between the two different antibodies and the overall antigen-antibody reactions can then be measured by an enzyme-substrate reaction (Engvall and Perlmann, 1972; Wisdom, 1976; Voller et al., 1980). Since a single molecule of enzyme can catalyze the conversion of a large number of molecules of substrate, enzyme systems are employed to magnify the antigen-antibody reactions (Yolken, 1982). The choice of enzyme system is thus important in determining the overall sensitivity of the assay. A variety of enzymes have been used to conjugate antibodies for ELISA. Alkaline phosphatase, horseradish peroxidase (HRP) and B-D-galactosidase are the most widely used enzymes. Recently, urease has been employed as a label for enzyme immunoassays and has resulted in an improvement of titration end point visualization which make it useful for field studies where spectrophotometers are not available (Chandler et al., 1982). A biotin-streptavidin system has also been developed for enhancing signal detection in immunoassays (Chaiet and Wolf, 1964; Guesdon et al., 1979; Buckland, 1986). This chapter is concerned with the stepwise development of a sensitive antigen trapping double antibody sandwich ELISA 6 4 for detection of trypanosomal antigens. The relative sensitivity of three enzyme systems: alkaline phosphatase, urease and biotin-streptavidin HRP were compared in an indirect ELISA. In addition, the relative sensitivity obtained using plastic microtiter plates and nitrocellulose papers (dot assay) as the solid-phase matrices for sandwich ELISA were also determined. 6 5 Materials and Methods

Experimental animals

Female 8 to 16-week-old BALB/c mice were purchased from Charles River Breeding Laboratories, St Constant, Quebec. Outbred Swiss mice were produced in the animal care unit (University of Victoria) from parental stock purchased from the Charles River Breeding Laboratories. Long-Evan rats were bred at the University of Victoria from parental stock obtained from Canadian Breeding Farms and Laboratories Ltd, St Constant, Quebec. A New Zealand white rabbit was obtained from R and R Rabbitry, Bellingham, Wa., USA.

Parasites

Trypanosoma brucei rhodesiense ViTat 1.1 (bloodstream form) was cloned in our laboratory from an uncloned population EATRO 1895 (Hill et al., 1978) obtained from Dr. George Hill, Nashville, Tennessee, USA. Bloodstream trypanosomes were grown in cyclophosphamide-suppressed (Smith et al., 1982) Long-Evan rats and purified from heparinized blood by chromatography on diethylaminoethyl cellulose (Lanham and Godfrey, 1970). The corresponding PCF were established from cloned bloodstream populations (Richardson et a l, 1986) and maintained in culture at 26°C in SDM-79 medium (Brun and Schonenberger, 1979) supplemented with 10% (v/v) foetal bovine serum (FBS).

Leishmania major promastigotes were doubly cloned in our laboratory from uncloned promastigotes of L. major NIH (Seidman strain) obtained from Dr. N. Reiner, Vancouver, B.C. The resulting population was designated L. major 1.1. The promastigotes were maintained at 26°C in SDM-79 medium containing 10% FBS.

Preparation of parasite lysates and membrane fractions

PCF trypanosomes or Leishmania promastigotes were harvested and washed once by centrifugation (800 g, 10 mins) in phosphate-buffered saline (PBS) containing glucose 6 6 (PSG) (0.01 M sodium phosphate, C.15 M NaCl, 1% glucose). Water lysates were then prepared as described previously (Richardson et al., 1986). Parasite membrane fractions were prepared according to Mancini et al., (1982), the protein concentration of lysates was determined by the Lowry method (Lowry et al., 1951) and the protein concentrations in membrane preparations were established using a modified Lowry procedure as described by Peterson (1983).

In separate experiments, trypanosomes were lysed in different solubilization buffers at 5 x lO^/ml for procyclic culture forms and 5 x 10^/ml for bloodstream forms of T. b. rhodesiense ViTat 1.1 (TBRP1) for 20 mins on ice. Soluble materials was assayed at 1/4, 1/8,1/16 and 1/32 dilutions in an indirect ELISA and sandwich ELISA using microtiter plates (see below). This was used to compare the relative quantities of trypanosomal antigens recognized by the three anti-TBRPl MAbs (MAbs # 247,346 and 477) in the bloodstream and PCF trypanosomes. Controls included the different solubilization buffers: 1%, 2% and 3% butanol/ 50mM Tris-saline pH7.4; 0.1% trifluoroacetic acid (TFA)/ 50mM Tris-saline pH7.4; 1% NP-40; 0.1% SDS; 0.1% SDS/ 0.5% NP-40/ 50mM Tris-saline pH 7.4; distilled water, PBS pH 7.4.

Antisera and monoclonal antibodies

An antiserum against crude membranes of T. b. rhodesiense ViTat 1.1 PCF was produced in a New Zealand white rabbit. Membrane protein (8.5 mg) was emulsified with Freund’s Complete Adjuvant (FCA) (Gibco, Vancouver, B. C., Canada) and injected subcutaneously and intramuscularly. Three weeks and six weeks 'ater, booster injections of 1.0 mg PCF membrane protein in Freund's Incomplete Adjuvant (FLA) (Gibco, Vancouver, B. C., Canada) were given both subcutaneously and intramuscularly. The antibody titre was determined 10 days later by ELISA (see below) and the rabbit was bled out 20 days after testing. Serum was prepared by standard methods.

Ten hybridomas producing specific monoclonal antibodies (MAbs) against T. b. rhodesiense PCF were produced in our laboratory by Jennifer Richardson who kindly made them available for this study. The MAbs bind surface-exposed T. brucei spp.- specific epitopes (Richardson etal., 1986) on the membrane glycoprotein, procyclin 6 7 (Richardson et al., 1988). Monoclonal antibodies were produced in the ascites fluid of pristane-primed BALB/c mice using standard techniques (Hoogenraad et al., 1983).

The IgG fractions of rabbit anti-PCF sera and ascites bearing the anti-PCF MAbs were concentrated by precipitation with a final concentration of 50% saturated ammonium sulphate. These fractions were the' used for radio-labelling and competitive solid-phase radioimmunometric assay (see below,,. Protein concentrations of the isolated IgG fractions were determined by the Lowry method (Lowry et al., 1951).

For the biotinylation of antibodies, the IgG fractions of rabbit anti-trypanosome PCF antisera and MAbs were isolated by ammonium sulphate precipitation and Protein A chromatography using Wood's (1984) methodology. The IgG fractions were labelled with biotin using a modified version of Focus' (1985) method. Biotin succinimide ester (CAB- NHS) (Bethesda Research Laboratories, Burlington, Ontario) was dissolved at a concentration of 30 mg/ml in N, N-dimethylformamide (dried over a molecular sieve, 4A, 4-8 mesh; Aldrich, Milwaukee, Wis. USA). A 50 pi aliquot of this solution was added to 4 mg of the IgG fraction in 2 ml of PBS (preadjusted to pH 9.0 using a 0.5 M Na 2C0 3 solution). After incubating the mixture for 2.5 h at room temperature the reaction was stopped by the addition of 1 M NH 4CI2 (Fisher Scientific, Vancouver, B.C.) to a final concentration of 0.1 M. The mixture was then dialysed overnight at 4°C against PBS containing 0.01% thimerosal (sodium ethyl mercurithiosalicylate) (Sigma, St Louis, Mo. USA) and stored at 4°C until used. The biological activities of the biotin-labelled antibodies were examined in indirect ELISA tests (see below) using streptavidin- biotinylated horseradish peroxidase complex (Amersham, Oakville, Ontario) to detect bound first antibody.

Indirect ELISA

Solid-phase ELISA was used to determine the binding specificity and the titre of specific antibodies in the rabbit antiserum or ascites fluids using the methodology of Parish et al. (1985). ELISA plate wells (Costar, Cambridge, MA, USA) were coated with 100 pi of either trypanosome PCF lysates or Leishmania promastigote lysates diluted in PBS to a concentration of 50 pg/ml. Antigen was dried onto the wells by incubating the microplates overnight at 37°C. Serial dilutions of rabbit antisera or ascites fluid were tested. Alkaline 6 8 phosphatase-labelled goat anti-rabbit Ig or goat anti-mouse Ig (TAGO Inc., Burlingame, CA, USA) were used as detecting antibodies at a dilution of 1/2000.

The relative sensitivities of three different enzyme systems (alkaline phosphatase, urease and biotin-streptavidin HRP) were compared in indirect ELISA. ELISA wells were coated with serial dilutions of parasite lysates (10? cells/well to 10 cells/well) and the assays were carried out with ammonium sulphate precipitated anti-TBRPl MAbs (#247, 346 and 477) at a predetermined dilution of a 1/4000 as described above. Alkaline phosphatase-labelled goat anti-mouse Ig or urease-conjugated rabbit anti-mouse Ig (ALLELIX Inc., Mississauga, Ontario) were used as detecting antibodies at a dilution of 1/2000. Alkaline phosphatase substrate solution was prepared by dissolving 100 mg of p- nitrophenol-phosphate (Sigma, St. Louis, Mo. USA) in 100 ml 0.1 M diethanolamine buffer, pH 9.8, containing 0.5 mM MgCl2. Urease substrate solution was prepared as described by Chandler et a!. (1982) by dissolving 8 mg bromocresol purple powder (Sigma, St. Louis, Mo. USA) 1.48 ml 0.01 M NaCH to a final volume of 100 ml with distilled water with the addition of 100 mg urea and EDTA (0.2M final concentration). The pH of this substrate was adjusted to 4.8 and the solution was stored at 4°C until used. Micotitre plates were read 60 mins after the addition of substrate solutions. Alkaline phosphatase remits were read at 405 nm while the urease results were determined both visually (yellow change to purple for a positive reaction) and spectrometrically at 405 nm. In addition, biotinylated MAbs at 1/350 dilution and streptavidin-horseradish peroxidase (HRP) complex at 1/1250 dilution were tested in ELISA. HRP substrate solution was developed (see below) and colour development was measured at 405 nm after 60 mins at room temperature. The optimum dilutions of both the unlabelled antibodies and the enzyme-conjugated antibodies in all three enzyme systems were determined by checkerboard titration of both types of antibodies using trypanosome PCF lysates (25 pg/ml) to ensure high readings.

Competitive solid-phase radioimmunometric assays

Eight MAbs were labelled individually with ^ 5 j (i.o mCi, New England Nuclear, Thunder Bay, Gntarir ) using the Chloramine-T method (Greenwood et al., 1963). Twenty ug of each antibody were iodinaied and competitive solid-phase radioimmunometric assays used to determine the number of epitopes recognized. Assays were performed as described 6 9 by Hall and Esser (1984). The optimum antigen coating concentration (20 pg/ml) and amount of radiolabelled MAbs (5 x 10^ c.p.m./well or 10^ c.p.m./ml) needed were previously determined by checkerboard titration in a direct solid-phase radioimmunoassay. Wells of 96-well, U-bottomed polyvinyl chloride microtest flexible assay plates (Falcon, Oxnard, Ca., USA) were coated with 100 pi of trypanosome lysates (25 pg/ml in water) and dried as described previously. The wells were then washed 3 times with PBS/0.05% Tween 20 (PBS-Tween) and unoccupied sites on the plastic were blocked by incubation with 200 pl/well of PBS containing 3% (w/v) fish gelatin (Hipure, Norland Products Inc., New Brunswick, N J., USA) at 37'C for 1 h. Fifty pi of unlabelled rabbit antiserum of MAb dilutions (10, 1,0.1,0.01,0.001 and 0.0001 pg/ml in PBS, pH 7.4) and 50 pi of radioisotope labelled MAb (adjusted to 10^ c.p.m./100 pi with PBS-Tween, pH 7.4) were added simultaneously to each antigen-coated well. The antibody mixtures were incubated for 2 h at room temperature and the wells were washed thrice with PBS-Tween to remove unbound radiolabelled MAbs. The plates were then cut with a hot wire and individual wells were counted for 1 min in an LKB/282 Compugamma Universal gamma counter (Waiiach Oy, Turku, Finland).

Double antibody sandwich ELISA

For this procedure, 3 MAbs (anti-TBRPi 247,346 and 477) Sj Lie for Jstinct epitope (determined by competitive solid-phase radioimmunometric assays) were selected. Landwkh ELISAs were performed as follows: 'capture' antibodies were coated on microplate wells by incubating 100 pi of unlabelled ammonium sulphate-precipitated rabbit anti-procyclic membrane antibodies or MAb mixture in PBS overnight at 37"C. Alter the plate was washed 3 times with PBS-Tween, unbound sites were blocked with 200 pi of PBS containing 3% fish gelatin at 37°C for 1 h. The wells were washed thrice with PBS- Tween and then 100 pi of antigen solution or test sera (diluted in PBS-T ween containing 1% fish gelatin and 0.01% thimerosal) were added to the wells. Similarly, buffer or normal serum from uninfected animals was added to control wells. Following an incubation of 16 h at 4°C, the plate v/as washed thrice as described above and 100 pi of biotinylated anti-TBRPi rabbit antibodies or biotinylated MAb mixture, diluted in the buffer, were delivered to each well. After incubation at 37°C for 2 h, the plate was washed and 100 pi of streptavidin-biotinylated horseradish peroxidase complex (Amersham, Oakville, Ontario, Canada) diluted to 1/1250 in buffer was added to each well. Following incubation for 60 mins at 37°C, the plate was again washed and 100 pi ABTS (2,2’-azino- bis {3-ethylbenzthiazoline sulphonic acid)) (Sigma, St. Louis, Mo. USA) substrate solution (50 ml 0.04 M citrate buffer, pH 4.0, 250 pi 0.04 M ABTS and 200 pi 30% H2O2) were added to each well. The plate was incubated for 60 mins at room temperature and colour development was measured at 405 nm using an EIA autoreader (Bio-Tek Instruments Inc., Burlington, WA., USA).

In order to determine whether or not individual MAbs bound to the same molecule, a double MAb sandwich EiJSA was performed as described above except that a single MAb was used as the capture antibody arid the biotin-labelled detector antibody. The capture MAb was assayed at dilutions between 1/250 and 1/16000, whereas the labelled MAb was assayed at a dilution that was pre-determined by indirect ELISA to give high O.D. readings. Trypanosome PCF water lysate (25 py/ml) were used in these assays.

In order to optimise the 1 jnditions for double antibody sandwich ELISA, checkerboard titrati *ns of unlabelled capture antibody against biotinylated detecting antibody were carried out, with the addition of 2.5 pg of trypanosome PCF membrane per well. The optimum dilutions of unlabeiied capture antibody and the biotinylated detecting antibody were determined ro be 1/4000 and 1/350, respectively. The sensitivity of this was defined by adding know n concentrations of trypanosome PCF lysates (ranging from 10^ to 2 x 107 trypanosome equivalents/ml) and trypanosome PCF membrane proteins (5-0.001 pg/ml). Lysates and membrane proteins of Leishmania major were also tested at the same dilutions.

Sera from infected or uninfected mice were tested using the double antibody sandwich ELISA. Dilutions of 1/4,1/8,1/16. 1/32 and 1/64 of sera in buffer were added to wells coated with either unlabelled anti-TBRPi MAb mixture or rabbit anti-TBRPi antibodies at a dilution of 1/4000. Biotinylated anti-TBRPi rabbit antibodies or MAb mixture at a dilution of 1/350, as stated previously, were used as the detecting antibodies. In addition, dilutions of known concentrations of trypanosome PCF water 1' sates or membrane fractions were used as positive controls. Samples were considered positive (i.e. presence of antigen) if wells showed absorbance values of a* least twice that of the average of the controls. 7 1 Nitrocellulose papers (Biorad, Richmond, CA., USA) were used in a separate sandwich ELISA* experiment to compare the relative sensitivities obtained with the nitrocellulose membrane and the standard plastic microtitre plates as the solid phase matrix. Double sandv/ich ELISA were carried out using either nitrocellulose papers (dot assay) or microtiter wells as described .ibove. Ammonium sulphate purified anti-TBRPi membrane rabbit antisera or normal rabbit sera at 1/4000 dilutions was used as the capture antibodies and the corresponding biotinylated rabbit antisera at 1/350 dilutions were used in the sandwich ELISA in microti .er wells. The optimum dilutions for these two types of antibodies used in the dot assay were 1/2000 and 1/350, respectively, determined by the checkerboard titrations as described above using trypanosome PCF antigens (25 pg/ml). Ten microliters of unlabelled rabbit antisera or normal rabbit antisera were spotted onto the nitrocellulose membrane. Spots were then dried and sandwich ELISA were carried out similar to that described for the microtiter plates. Horseradish-conjugated streptavidin at 1/700 was used in dot assay and the positive reactions (black spots) were visualized by immersing the nitrocellulose strips in 5 ml alpha-chloronaphthol (0.6 mg/ml in PBS containing 0.01% hydrogen peroxide). Color was allowed to develop for 30 min and then washed with tap water before drying. The sensitivity of both dot assay and micotiter ELISA was define/' by adding known concentrations of trypanosome PCF lysates (ranging from 103 to 2 x iO? trypanosome equivalents/ml). Lysates of Leishmania major were also tested at the same dilutions.

The stability of the immobilizing 'capture' antibodies on both the microtiter plates and nitrocellulose papers were compared by storing the antibody-coated microtiter plates or antibody spotted nitrocellulose papers at 4°C or room temperature for various time intervals (0,10, 20, 30, 40, 50, 60, 70, 80, 90, 100 days). Sandwich ELISA were then carried out using the stored nitrocellulose papers and microtiter plates with four different concentrations of trypanosome lysates (10^, 10^ , 10^ and 10^ trypanosome equivalents/ml) and 1/350 dilutions of biotinylated rabbit antibodies to determine the relative stabilities of these two assay systems.

Mouse sera

Two sets of sera from infected mice were examined in an attempt to detect uypanosomal antigens using the double antibody sandwich ELISA. In the fu st experiment, 15 outbred 7 2 Swiss white mice were injected intraperitoneally (i.p.) with 2 x 10^ bloodstream form trypanosomes ( T. b. rhodesiense ViTat 1.1) in 0.2 ml of sterile physiological saline (0.15 M NaCl). Six control mice received 0.2 ml of sterile physiological saline. Two or three mice were bled out at 3,6,9,13,17 and 21 days post-infection. Three control mice were bled out at 0,3 and 21 days post-infection. In the second experiment, 33 mice were infected by injecting (i.p.) 2 x 10^ bloodstream form trypanosomes and were bled out at 2, 4,6,10,12,14,16,18,20,22 and 26 days post-infection. Three control mice were bled out at every second time point. Blood was collected from the chest cavity of each mouse and was allowed to clot overnight at 4'C. After centrifugation at 500 g for 10 mins at 4°C, serum from each mouse was collected and stored separately at -20°C. Control mice were processed similarly. Blood from each mouse was examined microscopically for the presence of parasites using undiluted, 1/10 and 1/20 dilutions of blood in PSG containing heparin (25 units/ml). 7 3 Results

Grouping of monoclonal antibodies based on epitope recognition

Based on competition for binding to epitopes, the 10 MAbs could be classified into 3 different groups (Table 4). Group 1 was comprised of MAbs TBRP1 16,20,137,160 and 247; MAb 346 formed Group 2 and MAbs 324,415,418 and 477 were assigned to Group 3. The degree of inhibition of radiolabelled MAbs by unlabelled MAbs varied within groups, with individual MAbs varying from 10-30% to 80-90% inhibition within Group 1. Unlabelled MAbs 137 and 247 gave better inhibition than MAbs 16,20,160 (>80% versus 50% inhibition respectively). In Group 3, unlabelled MAbs 415,418 and 477 showed greater inhibition (>80%) than MAb 324 (10-30%) against radiolabelled 415, 418 and 477. None of the unlabelled MAbs, with the exception of its unlabelled homologue, exhibited any inhibition of the binding of radiolabelled MAb 346. Unlabelled anti-TBRPi rabbit antibodies inhibited the binding of all radiolabelled MAbs to varying degrees. Anti-gonococcal R-core lipopolysaccharide monoclonal antibody (T. W. Pearson, unpublished data) was used as a negative control and did not cause any inhibition.

Binding of biotin-labelled anti-TBRPi MAbs to trypanosomal PCF antigens trapped by homologous unlabelled MAbs or by heterologous MAbs was examined using a sandwich ELISA. As shown in Table 5, binding was observed in all cases except when the anti-gonococcal MAb (23.17.6, negative control) was used.

Selection of monoclonal antibodies for development of a sandwich ELISA

Three MAbs, one from each inhibition group (247, 346 and 477), were chosen based on their strong binding in the radioimmunometric assays. These MAbs and anti-TBRPi rabbit antibodies were purified on Protein A columns. Purity was confirmed by the appearance of only the immunoglobulin heavy and light chains on SDS-PAGE gels (data not shown). The purified antibodies were biotinylated and their activities were measured by indirect ELISA (Fig. 4). The binding curves of all three biotinylated MAbs and the rabbit polyclonal antibodies - ere similar. The curves for MAb 247 and 477 and rabbit antibodies were omitted from the graph for reasons of clarity. A mixture of these 3 MAbs was then 7 4 used in subsequent double antibody sandwich ELISA (see below) to detect the presence of trypanosomal antigens in sera from infected mice.

Optimization of sandwich ELISA

The double antibody sandwich ELISA technique utilizing the biotinylated anti-TBRPi MAb mixture and rabbit anti-membrane antibodies was shown to be specific for trypanosomal antigens. Fig. 5 shows a linear positive response using a sandwich ELISA with the addition of known concentrations of trypanosomal PCF antigens but not with Leishmania antigens. There were no differences observed in the sensitivity or specificity of tests that utilized either anti-TBRPi rabbit antibodies of the MAb mixture as the capture antibodies. Similar results were also obtained when either the biotinylated rabbit antibodies or the biotinylated MAb mixture was used to determine the minimum detection levels of these assays. However, the biotinylated MAb mixture did show a better linear response than the labelled rabbit antibodies (Fig. 5 B, C). The minimum detection for the double antibody sandwich ELISA was 1-5 ng trypanosome membrane protein/well (i.e. 10-50 ng/ml or the equivalent of extract from 5000 trypanosomes/ml).

The relative minimum sensitivity obtained by three different enzyme systems, alkaline phosphatase, urease and biotin-streptavidin HRP, ultilizing anti-TBRPi MAbs mixture is shown in Table 6. Biotin-streptavidin-HRP appeared to be the most sensitive system with a minimum detectable level of 5 x 10^ trypanosomes/ml. This was followed by alkaline phosphatase and urease detecting 10^ trypanosomes/ml and 5x10^ trypanosomes/ml, respectively. However, a rapid and vivid colour change (yellow to deep purple) was observed in the positive reaction of the urease system, while a more gradual colour changes over a period of time were observed with the alkaline phosphatase (colourless to yellow) and the biotin-streptavidin-HRP (colourless to purple).

The minimum detectable level obtained by the anti-TBRPi rabbit antisera using nitrocellulose papers was 10^ trypanosomes/ml while the minimum detectable level lor the microtiter plate was 5 x 10^ trypanosomes/ml in a sandwich ELISA . The sensitivity of antibody-spotted nitrocellulose paper remained unchanged after storing at either 4°C or room temperature for 100 days (Table 7). The minimum detectable level for antibody- coated microtiter wells was unchanged for 20-30 days at 4°C or room temperature. 7 5 Thereafter, a rapid decline in detection levels was observed with these stored ELISA plates. Plates that were stored at 4°C showed a slower decline in sensitivity than those maintained at room temperature (Table 7). No binding was seen with control normal rabbit sera as capture antibody in both microtiter plate and nitrocellulose membrane assays (data not shown).

A quantitative difference was observed when different solubilization buffers were used to extract antigens recognized by the anti-TBRPi MAb mixture (MAb # 247,346 and 477), from both bloodstream and procyclic form trypanosomes (Fig. 6). Procyclics yielded an order of magnitude higher O.D. readings than their bloodstream counterparts (Table 8). Differences in the solubility profiles of these antigens were also noted between the bloodstream and procyclic stages. While trypanosome antigens identified by these MAbs were extracted from the procyclics using different butar >\ concentrations (1%, 2% and 3%), 1% TFA and 0.1% SDS, these antigens were almost unobtainable from the bloodstream trypanosomes (Fig. 6). The solubilization buffer composed of 0.1% SDS/ 0.5% NP-40/ 50mM Tris-saline pH7.4 gave the hig.i st O.D. readings for both bloodstream and procyclic stages (Fig. 6).

Antigen detection in sera from infected mice

Sera collected from two sets of T. b. rhodesiense-infected mice were tested using the double antibody sandwich ELISA. The same 4 antibody combinations used in the artificial system were re-employed to test the mouse sera. Best results were achieved when anti- TBRPi rabbit polyclonal antibodies were used as the capture antibodies and the biotinylated MAb cocktail was used as the detecting reagent. Data are presented only for this assay combination (Figs. 7 and 8). Parasitemia data have also been incorporated into Fig. 7 and 8 to provide a direct comparison of these two diagnostic methods. Each data point represents the mean value obtained with each serum dilution in two repeated experiments.

Fig. 7 shows the ELISA results for the first set of mouse sera. Trypanosome antigens were first detected in sera six days post-infection. Antigen levels rose dramatically by the 9th day of infection and thereafter remained at an elevated level of 0.08-3.2 |ig protein/ml of serum. Antigen concentrations were determined using reference 7 6 concentrations of PCF membrane antigens. Microscopic detection of parasites in the blood was first achieved at 6 days post-infection. The parasite number remained thereafter at > 10^ cells/ml of blood.

The second set of sera from infected mice showed high antigen levels 2 days post­ infection (Fig. 8). Mean antigen levels then remained generally elevated with minimum concentrations occurring 6 and 20 days post-infection. Antigen concentrations ranged between 0.02 and 0.16 (ig/ml of serum. Trypanosomes were microscopically detected on the 6th day post-infection and remained at levels of >10^ cells/ml in later samples.

Both experiments showed similar profiles in the levels of parasitemia and antigen concentration with time (Figs. 7 and 8), but a temporal difference in the onset of detectable clinical symptoms was apparent. Within each experiment, a peak in parasitemia levels was typically associated with a drop in antigen concentration, and vice versa. Trypanosome antigens were not detected in any of the sera from uninfected mice. 77 Fig. 4

The effect of adding different biotinylated anti-T. b. rhodesiense procyclic monoclonal antibodies on binding to trypanosome procyclic water lysates in indirect ELISA. The following biotinylated monoclonal antibodies were assayed at a dilution of 1/2000 as described in the Materials and Methods section. MAb 346 (open circles), MAbs 346 and 247 (solid squares) and MAbs 346,247 and 477 (solid triangles). 7 8

1.0

: 0.8

0.6

0.4

0.2

6 5 4 3 2 NO. OF TRYPS ( 5 X 1 0Y / W ELL J 79 Fig. 5

Optimization of double antibody sandwich ELISA for detection of antigens in trypanosome water lysates or membranes. In A and B, a 1/4000 dilution of ammonium sulphate precipitated MAb mixture (MAbs 247, 346 and 477) was used as 'capture' reagent. In C and D, a 1/2000 dilution of ammonium sulphate precipitated rabbit anti-TBRPi membrane antibodies was used as 'capture' reagent The open symbols (squares, triangles) represent results obtained using a 1/350 dilution of bioH > lated rabbit antibody to PCF membrane as detecting reagent. The closed symbols (squaics, triangles) represent results obtained using a 1/350 dilution of biotinylated anti-PCF MAb mixture. Known concentrations of T. b. rhodesiense PCF lysates (A, C) or membrane (B, D) were used as reference antigens (solid squares, open squares). Control antigens used were lysates (A, C) or membranes (B, D) of Leishmania major 1.1 promastigotes (solid triangles, open triangles). ABSORBANCE (4 05nm) aSORBANCE (405 nm)

1 0

(i> a

z «

rO

m P S «

/ o 4 r o 8 1 Fig. 6

Detection of different solubilized extracts from Trypanosoma brucei rhodesiense bloodstream forms (open columns) and procyclic culture forms (crossed columns) using rabbit anti-PCF membrane polyclonal antibodies as 'capture' antibodies and biotinylated anti-TBRPi MAbs (247,346 and 477) as detector antibodies in a double antibody sandwich ELISA. Different trypanosome extracts were: 1 (1% butanol), 2 (2% butanol), 3 (3% butanol), 4 (0.1% TFA), 5 (1.0% NP-40), 6 (0.1% SDS), 7 (0.5% NP-40/ 0.1% SDS/50 mM Tris-saline pH 7.4), 8 (water) and 9 (PBS). Negative controls were buffers used to solubilize parasite lysates (black columns). 1 2 3 4 5 6 7

Solubilisation buffers 8 3 Fig. 7

Double antibody sandwich ELISA results (solid triangles) and parasitemia data (open squares) for sera from T. b. rhodesiense infected mice. A 1/4000 dilution of rabbit anti- PCF membrane antibody was used as the capture antibody and a 1/350 dilution of aotinyiated anti-TBRPi MAb mixture was used as the detecting reagent. Each data point represents the mean value ± standard deviations of either the maximum detectable amount of trypanosomal antigens in the dilutions of mouse sera (solid triangles) or the parasite number (open squares) found in the blood of infected mice. ABSORBANCE C 405nm ) 0.2 0.4 5 0 15 10 5 0 AS F POST-INFECTION OF DAYS oo -t*. 8 5 Fig. 8

Double antibody sandwich ELISA results (solid triangles) and parasitemia data (open squares) for the second set of sera from T. b rhodesiense -infected mice. The results are shown as outlined in the legend to Fig. 7. ABSORBANCE (405nm ) 0.4 0.6 0-0 1 2 30 20 10 0 / DAYS OF POST-INFECTI ON POST-INFECTI OF DAYS O) oo Table 4 Competitive solid-phase radioimmunometric assay of anti-trypanosome monoclonal antibodies

Unlabelled antibodies*

125j. 16 20 137 160 247 324 346 415 418 477 Negative Anti-TBRPl labelled controlt rabbit seruml MAbs* 16 4+ 2+ 4+ - 4+ ------4+ 137 1+ 1+ 3+ - 4+ ------3+ 160 3+ 2+ 4+ 3+ 4+' 3+ - 2+ 3+ 3+ - 4+ 247 2+ 1+ 3+ - 4 + ------3+ 346 ------4+ --- - 2+ 415 -- - ' - 2+ - 4 + 3+ 4 + - 2+ 418 -- - - - 2+ - 4+ 4+ 3+ - 2+ 477 - - - -- 1+ - 4+ 3+ 4+ - 2+

4+, over 80% inhibition; 3+, 50-80% inhibitor ; 2+, 30-50% inhibition; 1+, 10-30% inhibition; -, no inhibition.) * Monoclonal antibodies specific for T. brucei spp. procyclin (TBRP1 series), t Anti-gonococcal R-core lipopolysaccharide monoclonal antibody 23.17.6. 0 Made against crude membranes of T. b. rhodesiense procyclic culture forms. Table 5. Binding of biotin- or enzyme-labelled MAbs to trypanosomal antigens trapped by homologous or heterologous MAbs in double antibody sandwich ELISA

Detector antibodies *

Trapping 137 247 346 415 477 Rabbit serum@ antibodies*

16 3+ 3+ 3+ 3+ 3+ 3+ 20 3+ 3+ 3+ 3+ 3+ 3+ 137 3+ 3+ 3+ 3+ 3+ 2+ 160 3+ 3+ 3+ 3+ 3+ 3+ 247 3+ 2+ 3+ 3+ 3+ 3+ 324 1+ 2+ 2+ 1+ 2+ 2+ 346 3+ 1+ 2+ 3+ 2+ 2+ 415 3+ 1+ 1+ 1 + 1+ 1+ 418 2+ 1+ 1+ 1+ 1+ 1+ 477 2+ 1+ 1+ 1+ 1+ 1+

Rabbit 2+ 2+ 2+ 2+ 2+ 1+ serum

Negative controlf

Buffer -- - - -

3+, Absorbance > 0.5; 2+, absorbance > 0,3; 1+, absorbance > 0.08; -, absorbance < 0.04. * Monoclonal antibodies made against T. brucei spp. procy :lir' (TBRP1 series), t Anti-gonococcal R-core lipopoiysaccharide monoclonal antibody 23.17.6. (& Made against crude membranes of T. b. rhodesiense PCF. 8 9 Table 6 Comparision of three different enzyme assay systems in indirect ELISA

Enzyme system Detection Levels* (trypanosomes/ml)

Alkaline phosphatase® 104 Urease® 5x 104 Biotin-Streptavidin-HRpt 5x 103

* The minimum detectable level was determined by an indirect ELISA using anti-TBRPl MAbs mixture (#247,346 and 477) as described in Materials and Methods.

® Enzyme-conjugated antibodies were used as second antibodies at a 1/2000 dilution. t Biotinylated anti-TBRPl MAbs were used and Streptavidin-horseradish peroxidase was used as signal generating system. 90 Table 7

Stability of antibody coated microtiter plates and nitrocellulose paper in ELISA and dot-blot assays after storage at different time intervals and temperatures

Nitrocellulose papert Microtiter Platest

Days of 4’ C Room 4tTT Room Storage* Temperature Temperature

0 105 105 10^ 104 10 105 105 104 104 20 10- 105 104 104 30 105 105 104 105 40 105 105 105 105 50 105 105 105 106 60 105 105 106 107 70 105 105 106 107

80 105 105 107 -

90 105 105 --

100 105 105 - - t The minimum detectable levels of antigens of T. b. rhodesiense ViTat 1.1 PCF (trypanosome equivalents/ml) in sandwich ELISA using rabbit anti-TBRPl membrane antibodies and their corresponding biotinylated antibodies. * Antibody spotted nitrocellulose papers and microtiter plates were stored covered in the dark at either 4°C or room temperature. 9 1 Table 8 Detection of antigens or epitopes in bloodstream form and procyclic culture form trypanosomes by ELISA8

Water extracts1* Detergent extracts0 (A) Direct PCF 24.20d 29.45 Bloodstream 14.34 21.02

(B) Antigen trapping PCF 3.59 14.15 Bloodstream 2.56 15.20 a Direct (A) and antigen trapping (B) ELISAs were performed with the biotin-streptavidin system as described in Materials and Methods. The MAbs used in the cocktail were biotinylated TBRP1/247, 346 and 477. b Trypanosomes were lysed in distilled water at 5 x 10^ per ml for PCF and 5 x 10^ per ml for bloodstream forms and soluble material was used in the two assays. c Trypanosomes were lysed in 50 mM Tris-saiine pH 7.4/0.5% NP-40/0.1% SDS at 5 x 107 per ml for PCF and 5 x 10^ per ml for bloodstream forms and soluble material was used in the two assays. d Calculated as the ratio:of the absorbance obtained with trypanosome extracts absorbance obtained with antigen controls (distilled water or buffer). 9 2 Discussion

A stepvdse approach was taken in the work reported in this chapter to develop a sensitive antigen-trapping assay using MAbs and rabbit antibodies raised against the surface of T. b. rhodesiense PCF trypanosomes. A double antibody enzyme-linked immunoassay (ELISA) was chosen as the antigen-detection method because of its simplicity and sensitivity (Voller et al., 1978) and a variety of different enzyme-systems for maximum signal generation was explored. The biotin-streptavidin-HRP complex system was shown to be more sensitive than the alkaline phosphatase or the urease systems (Table 6) and was thus employed as the signal generating system. In addition, microtiter plates were used as the solid-phase support because they gave a higher detection level than that obtained with the nitocellulose supports (5 x 1()3 Vs 10^ trypanosomes/ml, respectively). However, the sensitivity of antibody-coated nitrocellulose paper was not affected by the long-term storage (3 months), even at room temperature (Table 7) and may therefore be useful for mass surveys in which temperature stable reagents are required.

Results from competitive solid-phase radioimmunometric assays allowed the designation of 3 groups of MAbs based on binding to different epitopes. One MAb from each group was chosen on the basis of high O. D. readings in indirect ELISA. A double MAb sandwich ELISA (Adtim et al., 1985) using homologous or heterologous MAbs a? capture or detecting antibodies was performed to determine whether multiple copies of epitopes were present and whether the epitopes were distributed on the same or different molecules. The results indicated that the epitopes recognized by each of the 3 different monoclonal antibodies are present on the same molecule and that there are multiple copier, of these epitopes on the antigen. It is therefore not surprising that an increased signal was achieved by mixing the different biotinylated MAbs to form a detecting reagent (Fig. 4). The presence of multiple epitopes on the trypanosomal antigen also explains the observation that the polyclonal rabbit anti-membrane antisera and biotinylated purified rabbit anti-membrane antibodies performed equally well as capture and detecting reagents respectively and worked in conjunction with the 3 MAbs.

The combination of rabbit anti-membrane antisera as antigen capture reagent and biotinylated MAb mixture as detecting reagent allowed the optimization of the sandwich ELISA. In an artificial system where known quantities of antigen were added to normal sera the assay was capable of detecting as little as 1-5 ng/well, equivalent to 500 93 ttypanosomes/well (i.e 10-50 ng/ml or 5000 trypanosomes/ml). The sensitivity of this ELISA is significantly higher than microscopic detection of trypanosomes (5 x 10^ cells/ml; Paris et al., 1982) and previously reported sandwich ELISA (1.5 (ig/ml; Rae and Luckins, 1984).

The MAbs used in the assay are specific to T. brucei spp. (Richardson et al., 1986) and in this study they showed no reactivity to Leishmania lysates. This implies that the assay is potentially useful in the diagnosis of both T. b. rhodesiense and T. b. gambiense infections of humans and T. b. brucei infections of cattle.

Circulating parasite antigens were detected in the sera of mice as early as 2 days post-infection. Trypanosomes were not seen microscopically in these sera until the 6th day post-infection. Antigen levels remained elevated in subsequent samples, although levels fluctuated with time. Differences in the onset of detectable antigens levels and in antigen profiles with time were observed between the two main sets of mouse sera. In the first experiment, detection of the parasite antigens on the 6th day post-infection was followed by a sharp increase in antigen titre by the 9th day, and levels remained relatively high in subsequent samples. In contrast, antigen levels in the second experiment rose dramatically on the second day post-infection, followed by a rapid decline to relatively low concentrations. Because sera were not collected from the same animals throughout each experiment, variation of the immune responses of different mice at various sampling times may have contributed to the diverging results obtained in the two experiments. The sharp initial increase in circulating antigens observed in the second experiment may indicate that the trypanosomes were destroyed more quickly in these mice. It is possible that parasit; antigens present in later infection stages in these experiments were partially in the form of immune complexes which would have lowered the detectable antigen titre.

In both experiments, an interesting temporal relationship was observed between parasitemia and antigen concentration levels. Peaks in parasitemia levels were temporally associated with falls in antigen concentrations and vice versa. This is indirect evidence that the trypanosome antigens detected in this assay were not exoantigens as observed in Babesia (Smith et al., 1981) and Plasmodium (James et al., 1985) infections. Trypanosome antigens observed in our assay might, however, occur in membrane fragments or internal antigens released by cell lysis. 9 * Previous immunofluorescence assays, solid-phase ELISA (Richardson et al., 1986) and ultrastructural immunolocation studies (Paul Webster and Terry Pearson, unpublished results) did not detect the procyclic trypanosome antigens identified by the 10 anti-TBRPl MAbs in bloodstream form trypanosomes. The present data, however, clearly detect antigens in sera of mice infected with bloodstream trypanosomes (Fig. 7 and 8). These observations were not at first easily reconcilable. However, it is apparent that bloodstream forms contain much lower amounts of this antigen (Table 8). It is now known that all ten anti-TBRPl MAbs bind to a single glycoprotein called 'procyclin' (Richardson et al., 1988). Although Roditi et al. (1987) could not detect any mRNA encoding procyclin' i the bloodstream trypanosomes, Mowatt and Clayton (1987) had occasionally found trace amounts of procyclin rnRNA in the bloodstream trypanosomes using both Northern blots and in situ hybridization to bloodstream trypanosomes. The latter authors proposed that the intermittent signal was due to the variation in trypanosome treatment before RNA preparation w which low levels of unstable message might be lost during trypanosome purification. Recently, Trypanosoma equiperdum, which is not cyclically transmitted and thus should not develop procyclic forms, has been found to express low levels of procyclin mRNA (Roditi and Pearson, 1990). In addition, coexpression of both procyclin and VSG on the surface of the bloodstream trypanosomes was observed using two-color flow cytometry as early as 10 hrs after the initiation of in vitro transformation of slender bloodstream trypanosomes to procyclics. It is possible that such transformation occurs wrhin the mammlian host. Indeed, it has been shown that bloodstream forms, incubated in vitro at 37#C in the presence of citrate and cis-aconitate, are capable of transforming to procyclic-Uke forms, but are urible to grow (Overath et al., 1986). The fine details of procyclin expression in bloodstream trypanosomes during normal infections is still obscure.

These data show that the double antibody sandwich ELISA employed is potentially a useful diagnostic tool for the detection of active T. brucei spp. infections. Indeed, studies on a number of sera obtained from vervet monkeys and humans infected with T. b. rhodesiense and T. b. gambiense have shown the presence of trypanosomal antigens in these sera using the present double antibody sandwich ELISA. The simplicity of the sandwich ELISA and its ability in detecting circulating trypanosomal antigens in patient's serum provides appropriate specificity and sensitivity that could possibly eliminate the need for parasitological confirmation of infection. However, testing of large numbers of well- documented sera from infected vervet monkeys, trypanosome-infected patients and control patients from endemic area; and the me ureroent of antigen levels persisting after trypanocidal drug therapy is needed to fully evaluate the utility of this technique for diagnosis of active infections. 9 6 Chapter 3. Serodiagnosis of human African sleeping sickness by detection of anti-procyclic antibodies and trypanosome antigens.

Introduction

Microscopic detection of trypanosomes in blood samples from human African sleeping sickness patients is often unreliable and many attempts have been made to do rise serological methods as diagnostic aids (Van Meirvennc et a l 1985). Jne of these, the Card Agglutination Trypanosomiasis Test (CATT) (Magnus et a l 1978) is the most widely used test for detection of T. b. gambiense infections which occur mainly in western and central Africa. The CATT does not detect infections caused by T. b. rhodesiense which occurs primarily in southern and eastern Africa. In order to improve upon existing diagnostic tests for both forms of African sleeping sickness, the Procyclic Agglutination Trypanosomiasis Test (PATT) was developed and it has proved to be effective in detecting anti-trypanosomal antibodies in the sera of T. b. rhodesiense-infected Vervet monkeys and in T. b. gambiense- infected humans (Chapter one). Serological tests based on the detection of anti-trypanosomal antibodies do not distinguish between past and currently active infections. An antigen detection assay for African sleeping sickness using monoclonal antibodies and rabbit antibodies to the surface antigens of procyclic trypanosomes was therefore developed as an unambiguous test for diagnosing currently active infections (Chapter two). However, when the test was used with a large number of sera from trypanosome-infected Vervet monkeys and humans, it gave disappointing results since only 70% of confirmed positive sera were positive in the assay (data not shown). Because the monoclonal antibodies employed in this assay recognize a single trypanosomal antigen, "procyclin", and the rabbit antiserum was made against procyclic membrane antigens, this test could be theoretically improved by using reagents which bound to a larger repertoire of trypanosomal antigens.

This chapter describes the use of rabbit polyclonal antibodies, made against whole lysates of T. b. rhodesiense procyclic culture forms (PCF), in a double antibody sandwich enzyme-linked immunoabsorbent assay (ELISA) for detection of circulating trypanosomal antigens in the sera of trypanosome-infected vervet monkeys, and in the sera of patients from Cote d'Ivoire, Kenya and Sudan. Purified unlabelled rabbit antibodies to whole lysates of T. b. rhodesiense procyclic culture forms were used as the antigen capture 9 7 reagent and biotinylated purified antibodies were employed as the detection reagent in a sensitive biotin-streptavidin-horseradish peroxidase system for the detection of antigens in sera. A comparison of the results with those obtained using the PATT and CA1T for detection of anti-trypanosome antibodies in patients’ sera and a discussion of the ultility of the sandwich ELISA for diagnosis of sleeping sickness are presented. 9 8 Materials and Methods

Parasites

PCF trypanosomes were established from cloned bloodstream populations of T. b. rhodesiense ViTat 1.1 (Richardson etal., 1986), T. b. gambiense TREU 1285 (U2) (Gray, 1972) and T. b. brucei LUMP 1026 (Bienen et al., 1983) using the methods of Brun and Schonenberger (1979). T. congolense 45/1 PCF were obtained from Dr. Reto Brun, Swiss Tropical Institute, Basel, Switzerland. Promastigotes of Leishmania major (NIH Seidman strain) were obtained from Dr. Neil Reiner, Vancouver, and promastigotes of L. donovani IS2D were from a cloned population obtained through Dr. R. W. Olafson from Dr. Dennis Dwyer, NIH, Bethesda, MD, U.S.A. The parasites were grown at 26°C in SM medium (Cunningham, 1973) containing 10% foetal bovine serum (FBS) and 50 |ig/ml gentamycin. The bloodstream forms of T. b. rhodesiense KETRI2537 or KETRI2545, derived from EATRO 1989 (Fink and Schmidt, 1979), were used to infect vervet monkeys as described in Chapter 1.

Preparation of Lysates

Trypanosome PCF were harvested and washed twice by centrifugation at 800 x g for 10 mins at room temperature in phosphate buffered saline pH 7.4 containing 1% glucose (PSG). Trypanosomes were then resuspended in phosphate buffered saline (PBS) to give a final concentration of 10^ trypanosomes per ml. Whole lysates were prepared by 6 cycles of 30 sec sonication bursts at 45 watts (setting 5 in a sonifier cell disrupter, model wl85E, Heat System-Ultrasonics, Inc., Plainview, New York, USA) on ice. Protein concentrations of the lysate were determined by the Lowry method (Lowry et al., 1951).

Water lysates or membrane fractions of trypanosome PCF and Leishmania promastigotes were prepared as described in Chapter 2.

Rabbit antibodies

Whole lysate of T. b. rhodesiense ViTat 1.1 PCF (1 mg) was emulsified with Freund's complete adjuvant and injected subcutaneously and intramuscularly into a male New 9 9 Zealand white rabbit. Three weeks later a boost of 0.5 mg lysate in Freund's incomplete adjuvant was given. Sera were tested against lysate in indirect ELISA (Chapter 2) 10 days after the boost. The rabbit was then bled out, serum prepared by standard methods and the IgG fractions were purified by ammonium sulphate precipitation followed by protein A chromatography (Wood, 1984). The protein concentration of the isolated IgG fractions was determined by the Lowry method (Lowry et al., 1951). Four milligrams of the purified IgG fractions were labelled with 1 mg of biotin succinimide ester (CAB NHS; Bethesda Research Laboratories, Burlington, Ontario) using the method described in Focus (1985). The activity of the biotin-labelled antibodies was determined by indirect ELISA (Chapter 2).

Monkeys, Infection and Treatment

Vervet monkeys ( Cercopithecus aethiops) were infected by Dr. Paul Sayer (KETRI, Muguga, Kenya) with T. b. rhodesiense and were treated with trypanocidal drugs at 28-42 days post infection as previously described (Chapter 1). Blood from these monkeys was collected prior to infection, at different time intervals during the infection and after trypanocidal drug treatment. Sera were tested with the PATT and with a double antibody sandwich ELISA (see below).

Patients from Lambwe Valley, Western Kenya.

Twenty-four patients infected with Trypanosoma brucei rhodesiense durng z natural outbreak of sleeping sickness (1978-1984) in the Lambwe Valley, South Nyanaza, Kenya (Wellde et al., 1989b) were bled before and at intervals during and after drug treatment at Homa Bay Hospital, Western Kenya. The infection status of these patients was confirmed microscopically by thick blood smears (Baker, 1970) and sometimes by subinoculation into rats (Wellde etal., 1989d). Trypanosome isolates and sera were collected and preserved by members of both the Division of Disease Control, Ministry of Health, Nairobi and the Walter Reed Project, Veterinary Research laboratory, Kebete, Kenya. Using the patient data collected from 1980 to 1984 as a guide (Wellde et al., 1989c), patients' status were classified as primary sleeping sickness infection and relapsed sleeping sickness infection (Wellde et al., 1989e). Patients with primary infections were parasitologically diagnosed 100 individuals with no previous history of sleeping sickness and they were subsequently treated with trypanocidal drugs: Suramin or Melarsoprol (Mel B). Patients with relapsed infections were diagnosed either parasitologically or clinically within three years of suramin therapy of a primary sleeping sickness infection (Wellde et al., 1989e). Relapsed infections described in this study were subsequently treated with Mel B (Wellde et al., 1989c). A number of patients required drug treatment a second or third time following relapse. The progress of all patients was examined clinically: blood was collected for serology and haematology, electrocardiography was performed and CSF was collected.

All procedures were performed in Kenya by Dr. Brace Wellde and his research team except the PATT and the double antibody sandwich ELISA which were performed in Victoria after frozen samples of the patients' sera were shipped from Kenya.

Patients from Cote d'Ivoire

Blood and cerebrospinal fluid (CSF) from 39 T. b. gambiense -infected sleeping sickness patients hospitalized at the clinic of the Project de Recherches sur la Tiypanosomiase (PRTC) in Daloa, Cote d'Ivoire, were examined by cellular, biochemical and parasitological parameters described in Chapter I. All patients showed positive parasitemias by at least one technique and thus all were confirmed sleeping sickness cases. The PATT and the double antibody sandwich ELISA were performed in Victoria on the patients' lyophilized sera shipped from Daloa.

Patients from Sudan

A mass survey of 1300 people for Gambian form sleeping sickness was carried out by Drs. Tony Vervoort and Simon Van Nieuwenhove in South Sudan using the CATT. Venous blood and CSF samples were subsequently obtained from 104 serologically CATT positive persons. Samples were examined for various cellular, biochemical, parasitological and serological parameters (see below). With the exception of some serological tests, all the analyses were performed at local clinics in Sudan immediately after the samples were collected. 101 Blood samples from Sudanese patients were examined microscopically for the presence of trypanosomes using the wet blood film preparation and the Giemsa-stained thick blood film method (Baker, 1970). The mini-anion exchange centrifugation technique (m-AECT)(Lumsden et al., 1979) was performed on some samples to assist microscopic detection. Lymph samples were taken from patients with enlarged cervical nodes and were examined microscopically for the presence of trypanosomes.

The total number of white blood cells in CSF of patients were determined microscopically using a haemocytometer. Protein concentrations were determined using a commercial kit (Sopar-biochem, Brussels) based on the reaction between red Pyrogallol- Molybdate complexes and proteins.

The sera were kept for 7-10 days at -15°C in Sudan and then shipped to Belgium on ice. The CATT was repeated on the sera samples at the Institute for Tropical Medicine in Antwerp, Belgium. Lyophilized sera were mailed to the University of Victoria, Victoria, Canada, where the PATT and the double antibody sandwich ELISA were carried out. With the exception of PATT and sandwich ELISA, all the cellular, biochemical, parasitological and CATT tests were performed by Drs. Tony Vervoort and Simon Van Nieuwenhove.

Other patients

Sera from 17 patients who were either parasitologically or serologically diagnosed as having other parasitic diseases were collected prior to drug treatment by Dr. Judy Issac Renton, Vancouver General Hospital, Vancouver. Frozen samples were sent to Victoria for testing in the PATT and the sandwich ELISA.

Sandwich ELISA

Sera from vervet monkeys and sera from patients in Daloa, Kenya and Sudan were examined in an attempt to detect the presence of trypanosomal antigens using sandwich ELISA. Assays were performed as previously described (Chapter 2). Briefly, the ELISA steps were as follows: a 1/4000 dilution of unlabelled, ammonium sulphate precipitated rabbit antibodies to whole lysate of T. b. rhodesiense PCF and a 1/400 dilution of the 102 corresponding biotin-labelled antibodies were used as the capture and the detecting reagents, respectively. Bovine serum albumin (BSA; Sigma, St. Louis. Mo., USA) was used, instead of fish gelatin, in the blocking solution. In addition, 1% BSA was added to the buffer that was used to dilute the test sera and the biotinylated rabbit antibodies. All sera were used at dilutions of 1/4,1/8,1/16 and 1/32. Control wells contained PBS, lysates or membrane fractions of trypanosome PCF or sera from uninfected North Americans. In addition, sera from 106 individuals who lived or worked in the Lambwe Valley, western Kenya, and sera from 17 North Americans infected with other parasitic diseases were tested with the sandwich ELISA. Absorbance ratios were calculated by dividing the mean O.D. readings for test sera with the O. L). readings of North American egative controls. Samples were considered positive (i.e. presence of antigens) if sera showed absorbance ratios above two.

The sensitivity of the sandwich ELISA was determined by adding known concentrations of trypanosome PCF lysates (ranging from 10^ to 2 x 10^ trypanosomes/ml) and trypanosome PCF membrane proteins (5-0.001 pg/ml). Lysates and membrane proteins of Leishmania major were also tested at the same concentrations.

Agglutination test

The titres of antibodies specific for the surface of T. b. rhodesiense ViTat 1.1 PCF were measured by the PATT as described in Chapter 1. Both trypanosome PCF and Leishmania promastigotes were used as test organisms. Sera from vervet monkeys, and from human patients from Cote d'Ivoire, Kenya and Sudan were tested at doubling dilutions of 1/10 - 1/320 in PSG. Controls included PSG/10% FBS and sera from healthy, uninfected North Americans. In addition, sera from patients infected with other parasites, such as Toxoplasma and Leishmania, and sera from uninfected humans in endemic areas were tested.

The presence of anti-trypanosome variant surface glycoprotein antibodies in sera from trypanosome infected Daloan and Sudanese patients was determined by the CATT (Magnus et al., 1978) using a commercially available kit (Testryp CATT, Smith Kline-RTT, s.a., Rixensart, Belgium) according to the procedure specified by the manufacturer. Blood was collected from patients and uninfected Sudanese using the finger prick method (WHO, 103 1983) and was placed into a heparinized microhaematocrit tube. One drop of reconstituted CATT antigen and one drop of the collected blood were placed onto a reaction card. The reagents were mixed with a clean glass rod and they were agitated on an electric rotator for 5 min before the reaction was assessed (WHO, 1983). Sera from Sudanese patients were also tested in CATT at dilutions of 1/2,1/4,1/8,1/16 and 1/32. Controls included PBS and sera from uninfected humans. 1 0 4 Results

The double antibody sandwich ELISA using rabbit anti-T. b. rhodesiense PCF whole lysate specifically trapped trypanosomal antigens but not Leishmania major antigens when parasite lysates or membrane proteins were tested (Fig. 9). This assay was shown to detect antigen from as few as 100 ± 16.25 trypanosomes /well or 5 ± 2.73 ng trypanosomal membrane proteins/well (mean ± S.D.) in four different experiments over a three-month period.

Vervet monkeys

Sera from twelve vervet monkeys infected with T. b. rhodesiense were examined using the double antibody sandwich ELISA and the PATT. All sera were numbered consecutively from VI to V47 as described in Chapter 1. Results for both tests are shown in Table 9. The data are further summarized with respect to the status of infection (Fig. 10).

The sandwich ELISA results correlated with the infection status of vervet monkeys. No detectable antigen was observed in sera collected prior to experimental infection (sera V3, V8, V16, V19, V26, V32 and V36) or in sera taken either shortly after drug treatment (12 - 54 days, sera V14, V18, V23, V30, V31 and V40) or long after drug treatment (105 - 933 days, sera VI, V13, V15, V35, V40 and V41) (Table 9). One serum (no. V2) (Table 9), which was taken long after drug cure, was positive in ELISA but negative in the agglutination assay. During the infection, trypanosomal antigens were detected as early as 7 days post-infection (serum no. V4) (Table 9). Antigen levels reached a maximum at about 14 -16 days post-infection in all infected monkeys (sera V5, V10, V20, V27, V33 and V37) (Table 9). Thereafter, trypanosomal antigen levels decreased at 28 days and increased slightly prior to drug treatment. Control sera from mice infected with T. b. rhodesiense bloodstream forms (V47) gave positive results, as did T. b. rhodesiense water lysates and membrane proteins. Sera from uninfected North American humans (sera V44 and V45) and mice (V46) were negative (Table 9).

Parallel testing of the vervet sera for anti-trypanosome antibodies in the agglutination test (PATT) gave similar results to those described in Chapter 1. This is not surprising since the same sera were used in the previous study. Anti-trypanosome PCF 1 0 5 antibodies were detected only in sera from vervet monkeys with active, untreated infections or sera taken shortly after drug treatment (12 - 54 days post-drug therapy) (Fig. 10). Prior to drug therapy, infected individuals were usually positive in both sandwich ELISA and PATT assays. The results of ELISA and PATT assays diverged in that the antigen levels as detected by ELISA decreased or disappeared in sera taken soon after drug treatment whereas detectable antibody levels persisted (Fig. 10).

The results of both antibody and antigen detection assays in two monkeys (numbers 47 and 49) were plotted graphically in order to show the relationship between antigen and antibodies (Fig. 11). Both antibody and antigen levels increased markedly by 28 days and had increased again at 56 days post-infection. Antibody levels remained high throughout the infection period. A similar oscillation in antigen levels associated with a much more stable level of antibodies was also found in monkeys 96,115,124 and 128 (Table 9).

Patients from Lambwe Valley, western Kenya

A total of 251 sera from 24 trypanosome-infected patients were collected before trypanocidal drug treatment and at various times after treatment Each serum was individually labelled to record the patient and sampling sequence, e.g. serum N-9 is from patient N, bleed no. 9. Th* presence or absence of trypanosomes in the blood of these patients was also determined. All patients had positive trypanosome parasitemias prior to the trypanocidal drug treatment (Weh et al. 1989b). After the initial drug therapy, parasites were observed in the blood of 8 out of the 14 relapsed patients (data not shown). PATT and double antibody sandwich ELISA were used to respectively determine the presence of specific anti-trypanosome antibodies and trypanosomal antigens in each serum sample. Elevated anti-trypanosome antibodies and trypanosomal antigens were observed in patients’ sera collected before trypanocidal drug treatment and at the time of relapse (Table 10). All controls gave negative results in sandwich ELISA. They included 30 Kenyans from the Lambwe Valley who tested negative for trypanosomiasis by inoculation of blood into rodents and by blood smear, 6 Americans working in the Lambwe Valley and 3 healthy Canadians from Victoria. Although sera from Americans and Canadians showed undetectable anti-trypanosome antibody titres by PATT, a low antibody titre (1/35) was observed in the 30 Kenyan controls (Table 10). 1 0 6 Trypanosomal antigens were detected in sera taken from all patients prior to drug treatment except serum # 1 (patient M) (Table 11). After trypanocidal treatment, a rapid decrease in the detectable antigen level was observed in the sera of those 10 patients that showed no relapse for 4-6 years post-drug treatment (Fig. 12A). The antigen levels in these patient's sera became undetectable at 7-41 days post-treatment (Fig. 12A). Four patients relapsed shortly after the initial drug treatment (days 74-250 post-treatment). In these cases, persistently high antigen levels were observed in sera collected before relapse (Fig. 13A). Eight patients who relapsed long after drug treatment (days 291-833), showed a rapid decrease to undetectable antigen levels in sera taken shortly after the initial drug treatment (Fig. 14A). However, the trypanosomal antigen levels soon rose to a clearly detectable level when sera were taken shortly before (30 days) or during relapse. After secondary drug treatment, antigen levels in sera of these relapsed patients decreased rapidly and disappeared (Fig. 14A). Control sera from T. b. gambiense -infected patients (sera # 21 and # 22; Table 11) were positive in the sandwich ELISA, as were the lysates and membrane preparation of T. b. rhodesiense ViTat 1.1 PCF (# 23 and 24, Table 11, respectively). Sera from uninfected North American humans and 30 uninfected Kenyans were negative. However, further testing of 106 individuals who lived or worked in the Lambwe Valley yielded 10 positive reactors. These individuals all had their blood examined and inoculated into mice with negative results. Thus, the ELISA test appeared to give false positive reactions in 9.4% of the "normal" samples from the Lambwe Valley.

Parallel testing of the sera for anti-trypanosome antibodies in the agglutination test (PATT) gave similar results to those observed in the double antibody sandwich ELISA. Anti-trypanosome antibodies were detected in sera taken from patients with active untreated infections or sera taken prior to or during relapse (Table 10). An increase in anti- trypanosome antibody levels, followed by a decrease to low antibody titres (1/10-1/40), was observed in sera taken from drug treated patients (Fig. 12B and 14B). The results of sandwich ELISA and PATT for sera taken from drug cured patients differed in that antigen levels as detected by ELISA decreased or disappeared by 7-41 days post-treatment, whereas the antibody levels persisted over 6 months to a year (Fig. 12B, 13B and 14B). 1 0 7 Patients from Cote d'Ivoire

Ail 39 patients sera from Daloa, Cote d'Ivoire, were tested for anti-trypanosome antibodies using the PATT and CATT. In the PATT, anti-trypanosome antibodies were detected in all sera from patients infected with T. b. gambiense. While high agglutination titres (1/64 to 1/2048) were observed in most sera from trypanosome-infected patients, five of the sera (Nos. D376, D377, D378, D421 and D433) showed no agglutination in the CATT and sera from patients Nos. D376 and D377 gave only a marginally positive agglutination reaction (Table 12). Control sera were negative in both tests.

Using the double antibody sandwich ELISA, trypanosomal antigens were detected in all sera from human sieeping sickness patients but not in sera from controls (Table 12). High absorbance ratios (ratios between 3.3 and 18.7) were observed in all sera from sleeping sickness patients but not in sera from controls (Table 12).

Patients from Sudan

Venous blood and CSF from a total of 104 Sudanese were examined for various cellular, biochemical, parasitological and serological parameters. Results for these tests are presented in Tables 13-16. Patients, of ages between 1 - 60 yrs, were divided into 3 groups according to their clinical manifestations: 29 patients had trypanosomes in their blood or lymph extracts ("P", Parasitological cases), 38 patients had a previous history of human sleeping sickness and had received 3-12 injections of trypanocidal drugs 1 - 4 yrs prior to the sample collection ("O", oid cases) and 26 patients had no previous history of trypanosome infection but were shown to be positive serologically by the strong agglutination reaction in the CATT ("S", Serological cases). Three serological case patients (S 24 - S26; Table 15) showed prevalent clinical signs of human sleeping sickness - including intermittent fever accompanied by headache, joint pain, splenomegaly and lymphadenopathy. Patients were numbered consecutively within each group and were referred to by the group numbers. Samples from 11 presumably uninfected Sudanese were grouped and labelled as N1 - N il ("N", Negative cases); five of these control samples (N1 - N5) were obtained from Laboratory personnel in Sudan while others (N6 - N11) were local Sudanese who were monitored for two years but showed no clinical sign of sleeping sickness. Therefore, they were assumed to be false positives in CATT. The control 108 samples from 4 uninfected North Americans and 2 sleeping sickness patients from West Africa (Cote d'Ivoire) were labelled as NV1 to NV6 (Table 16). Results for all three serological tests were further summarized with respect to the patient's infection status in Table 17.

Elevation in the total number of white blood cells and protein concentrations in the CSF was observed in the majority of the parasitologically proven patients (Parasitological cases) and some trypanocidal drug treated patients (Old cases) (Table 13 and 14). In addition, 65% of the serological case patients also showed elevated protein concentrations (Table 15).

The presence of anti-trypanosome antibodies in sera was determined using the PATT and the CATT. In the PATT, agglutination of T. b. rhodesiense PCF occurred in sera from 27/28 parasitologically proven patients, i.e. Parasitological cases (Table 13). Cases no.serum P5 was the exception. High agglutination titres (1/80 - 1/320) were obtained in 18 sera while weak agglutination titres (1/10 -1/40) were found in 8 sera (Table 13). In the 33 trypanocidal drug treated patients (Old cases), 13 sera gave no agglutination reaction while 20 sera resulted in either a strong or weak agglutination in PATT (Table 14) (a total of 9 vs 11, respectively). Only 19 of the 26 serologically positive sera in CATT (Serological cases) were positive in the PATT. While 6 serological CATT positive sera were negative in PATT, 15 sera gave strong agglutination reaction (1/80 -1/320) and 4 sera showed in weak agglutination reactions (1/10 -1/40) in PATT (Table 15). Four sera from the 11 persumably uninfected Sudanese (Negative cases; Table 16) gave weak agglutination titres (1/10 -1/40) in PATT and only one sera from these Sudanese controls resulted in a high agglutination titre (1/320). In addition, sera from 4 uninfected North Americans produced no agglutination in PATT, whereas sera from 2 trypanosome-infected patients from West Africa showed positive agglutination (Table 16).

The CATT’ results were broadly similar to those observed in the PATT, in that positive agglutinations in the PATT correlated with the positive results in CATT. However, the number of false positives, as indicated in the Negative cases (Table 3 6), obtained by CATT were slightly higher than those using the PATT (6 vs 5 false positives; CATT vs PATT). Furthermore, a decrease in the total number of strongly agglutinating sera was observed using the PATT with sera from trypanocidal drug treated patients (Old case; Table 14) (16 in CA fT vs 9 in PATT) and in sera from parasite positive patients (27 109 in CATT vs 19 in PATT), In general, fewer positives were obtained in CATT when sera rather than whole blood were tested.

Trypanosomal antigens were detected by the double antibody sandwich ELISA in sera of all 28 parasitologically proven patients (Table 13). The relative abundance of parasite antigens in these sera was reflected by the high absorbance ratios (ratios between 4.0 to 15.4), with the exception of sera P6 and P7. In general, sandwich ELISA gave results that were similar to those observed with the PATT and CATT in that high absorbance ratios or strong agglutination titre-s were observed in parasitologically proven patients with a decline of absorbance ratios and anti-trypanosome antibody titres observed in trypanocidal drug treated patients (Table 17). However, fewer trypanocidal drug treated patients and uninfected Sudanese gave positive sandwich ELISA results (Table 14 and Table 16). A higher number of serological case patients were positive in the sandwich ELISA as compared to PATT and CATT when patients’ sera were tested (22 in ELISA vs 19 in PATT and 16 in CATT) (Table 15).

Patients with other parasitic diseases

Sera from North American patients with other parasitic diseases were used to determine the specificities of both the sandwich ELISA and the PATT. Only one of the 17 sera from these patients gave a positive though very weak reaction (absorbance ratio = 2.24) in sandwich ELISA while three resulted in weak agglutinations in PATT (Table 18). Control sera from 2 uninfected North Americans gave negative results and T. b. gambiense-infected patients from the Cote d'Ivoire were positive in both assays. 110 F.g. 9

Detection of trypanosomal antigens in water lysates of parasites by double antibody sandwich ELISA. T. b. rhodesiense ViTat 1 PCF (open triangles); T. b. gambiense U2 PCF (open circles); T. b. brucei 10-26 PCF (open squares); T. congolense 44/1 PCF (solid triangles); L. major Al (solid circles); L. donovani IS2D (solid squares); PBS (open hexangles). ABSORB ANCE (405 nm) i 6 1 0.8 1.2 0 -o T PARASITES WELL PER .mo. T T

BO ■ ___ T 1 112 Fig. 10

Summary of antibody and antigen detection tests with respect to the infection status of vervet monkeys infected with T. b. rhodesiense. Sera were collected before infection (A), at days 7-56 post-infection (B), at days 12-54 post-drug treatment (C) and at days 105-933 post-drug trertment (D). Mean ± S.D. values of anti-trypanosomal antibody titres (open squares) and trypanosomal antigens, as represented by absorbance ratios (solid triangles), in sera are shown respectively. I

ABSORBANCE RATIO

>-

n- ■QH

o - £3—

SERA DILN

SU 1 1 4 Fig. 11

Anti-trypanosome antibodies and trypanosomal antigens in sera of vervet monkeys No. 47 and No. 49 during T. b. rhodesiense infection. Antibody titres are presented as closed symbols (triangles, No. 47; squares, No. 49). Antigen levels are indicated by the absorbance ratios and are shown as open symbols (triangles, No. 47; squares, No. 49).L “64

32v o - l x 16^

"58 _J 2 1 5 ’* «

42 s to 1 "1

- 0

60 DAYS POST-? TECTION

in 1 1 6 Fig. 12

Detection of anti-trypanosome antibodies and trypanosomal antigens in sera from 10 Kenyan patients. Sera were collected before drug treatment (A) and at various time intervals after successful trypanocidal drug treatment, 0 -7 days (B); 8-30 days (C); 31-90 days (D); 90 -180 days (E); 180 - 360 days (F) and > 360 days (G) post-drug treatment. Mean ± S.D. values of anti-trypanosomal antibody titres (open triangles; B) and trypanosomal antigens, are represented by absorbance ratios (solid squares; A), in sera are shown respectively. (1/x

TITRES ) 200 RATIOS 160 40 * otdu Treatment Post-drug B 7 1 1 1 1 8 Fig. 13

Measurement of anti-trypanosome antibodies (A) and circulating trypanosomal antigens (El in sera from patients who relapsed shortly (74 - 250 days) after trypanocidal drug treatment Sera were collected before drug treatment (A); at various time intervals after the first drug treatment: < 1 week (B), between 1 week to 1 month (C) and 1 -2 months (D) post-treatment; at relapse or within a month prior to relapse (E); and at different times after the second drug treatment: < 1 month (F), 1-3 months (G), 3-6 months (H), <1 year (I) and > 1 year post-treatment (J). Mean ± S.D. values of anti-trypanosomal antibody titres (open triangles; B) and trypanosomal antigens, are represented by absorbance ratios (solid squares; A). TI TRES (1 / X ) RATIOS 200 1 4 0 - - 0 4 1 80- • * * • • * » * i • • " 0 2 12 4 4 0 8 i e ore f Infection of CourseTime

- - A B C D E F G H | J ABCDEFGHIJ B 119 120 Fig. 14

Detection of anti-trypanosome antibodies (A1 and circulating trypanosomal antigens (EQ by the PATT and the sandwich ELISA in sera from eight patients who relapsed long (291-833 days) after trypanocidal drug treatment. Sera were collected before drug treatment (A); at various time intervals after the first drug treatment: < 1 month (B), 1-3 months (C), 3 -6 months (D), and 6-12 months (E) post-treatment; at relapse or within a month prior to relapse (F); and at different times after the second drug treatment: < 1 week (G), 1-4 weeks (H), 1-3 months (I), 3-6 months(J), 6-12 months and > 1 year post-treatment (L). Mean ± S.D. values of anti-trypanosomal antibody titres (open triangles; B) and trypanosomal antigens, are represented by absorbance ratios (solid squares; A)- 121

1 2

<0 O H- < CC

B

X ^ 120-

(/> Hi CC 60- 1“

ABCDEFGH I J KL

Post-drug Treatment Table 9. Measurement of anti-procyclic antibodies and circulating trypanosomal antigens in vervet monkey sera before and during infection with T. b. rhodesiense and at various times after treatment with trypanocidal drugs.

Serum Monkey No. of No. of days Remarks Agglutination Sandwich N o. N o. days since since drug titre& ELISA (V) infection8 treatment ratioc (drug in brackets)

T"' 1 4 940 859 (Mel. B.) • 1.00 2 35 978 933 (Suramin/ - 3.18 M. K.436) 3 47-1 0 0 Preinfection . 1.00 4 47-2 7 0 Pretreatment 2.00 5 47-3 14 0 Pretreatment 1/80 6.80 6 47-4 28 0 Pre treatment 1/160 3.63 7 47-5 56 0 Pretreatment >1/320 4.84 8 49-1 0 0 Preinfection 1.16 9 49-2 7 0 Pretreatment 1/20 1.16 10 49-3 14 0 Pretreatment 1/160 3.73 11 49-4 28 0 Pretreatment 1/80 2.02 12 49-5 56 0 Pretreatment 1/80 3.72 13 61 875 793 (Berenil) 528 days since last pos. - 0.99 (CSF) 14 85-1 392 130 (Suramin/ 29 days since last pos. “ 0.79 M.K.436) (blood,CSF) (2nd treatment) 15 85-2 605 343 (Suramin/ 213 days since last pcc 1.06 M.K.436) (Mood) (2nd treatment) 2 2 1 Table 9 (cont'd)

Serum Monkey So. of No. of days Remarks Agglutination Sandwich No. No. lays since since drug titreb ELISA (V) nfection3 treatment ratio® (drug in brackets)

16 94-1 0 0 Premfection 1.10 17 94-2 23 0 Pretreatment >1/320 3.80 94-3 352 158 11 days since last pos. 1/320 0.94 (Suramin/T.S. (3rd treatment) 88) 19 96-1 0 0 Freinfection 1/20 1.21 20 96-2 14 0 Pretreatment 1/40 2.92 21 96-3 28 0 Pretreatment >1/320 0.30 22 96-4 58 14 1st and only treatment 1/80 2.81 (Suramin/T.S. 88) 23 96-5 71 27 1st and only treatment 1/40 1.31 (Suramin/T.S. 88) 24 96-6 202 158 1st and only treatment • N.D. (Suramin/T.S. 88) 25 113 0 0 Preinfection 1/20 N.D. 26 115-1 0 0 Preinfection 1/80 0.81 27 115-2 16 0 Pretreatment 1/40 2.62 Table 9. (cont'd)

Serum Monkey No. of No. of days Remarks Agglutination Sandwich N o. N o. days since since drug titreb ELISA (V) infection3 treatment ratioc (drug in brackets)

28 115-5 33 0 Pre treatment 1/160 2.51 29 115-4 56 12 (TS.88) 1st treatment (relapsed 1/320 1.83 later) 30 115-5 70 26 (TS.88) 1st treatment (relapsed 1/160 1.26 later) 31 115-6 98 54 (TS. 88) 1st treatment (relapased 1/320 1.01 later) 32 124-1 0 0 Preiiifection - 0.47 33 124-2 16 0 Pre treatment 1/40 2.97 34 124-3 33 0 Pretreatment 1/320 0.68 35 124-4 352 105 (1'S. 88) 3rd treatment “ 0.84 (Suramin/TS.88, 1st and 2nd)N.D. 36 128-1 0 0 Preinfection 1/10 1.07 37 123-2 16 0 Pretreatment 1/320 2.72 38 .z8-3 33 0 Pretreatment 1/80 0.06 39 128-4 41 0 Pretreatment 1/80 3.61 40 128-5 352 278 132 days since la^t pos. 1/320 0.75 (Suramin/T.S. (blood) (2nd treatment) 88) 24 1 Table 9. (cont'd)

Serum Monkey No. of No. of days Remarks Agglutination Sandwich N o. N o. days since since drug titreb ELISA (V) infection3 treatment ratio0 (drug in brackets)

4l ll8-d 366 291 146 days since last pos. “ 0.94 (Suramin/T.S. (blood) 88)

42 96-CSF 71 27 •* 0.56 (Suramin/T.S. 88) 43 115-CSF 33 0 • 0.48

44 neg. 0 0 Human serum (North - 1.00 American)

45 neg. 0 0 Human serum (North - 1.00 American)

46 ncg. 0 0 Mouse serum - 0.78 (uninfected) 47 pos. 21 0 Mouse serum (infected 1/320 5.79 with T. b. rhodesiense)

a Animals were infected with T. b. rhodesiense KETRI2537 except No. 61 which was infected with T. b. rhodesiense KETRI2545. b Agglutination T. b. rhodesiense ViTat 1.1 PCF. c Ratio = O. D. for test sera + O. D. for the average of the negative control sera (numbers V44 and V45) N.D. = not done. 5 2 1 Table 10.

Detection of anti-trvpanosome antibodies and trypanosome antigens in sera of trypanosome-infected Kenyans before drug treatment and at the time of relapse.

Pretreatment Relapsed Control * Control ** Controlt Samples# (n=17) Samples^ (n=10) (.1=30) (n=6) (n=3) (mean ± SD) (mean ± SD) (mean ± SD) (mean ± SD) (mean ± SD)

Absorbance 7.14 ± 2.79 4.46 ± 2.89 1.49 ± 0.78 1.36 ± 0.35 1.08 ± 0.19 Ratios® PATT (titre) 1/74.12 ± 96.4 1/88 ± 116.34 1/35 ± 79.78 0 ± 0 0 ± 0

# Sera collected from trypanosome-infected patients from Lambwe Valley, western Kenya, pricr to any drug treatment $ Sera collected from sleeping sickness patients at relapse before the second drug treatment * Kenyans from the Lambwe valley who tested negative for trypanosomiasis by inoculation of blood into rodents and by examination of thick blood smear. ** Americans working in the Lambwe Valley. *** Healthy North Americans from Victoria, Canadi. @ Detection of trypanosomal antigens in patients' sera with sandwich ELISA using rabbit antibodies against whole lysate of T. b. rhodesiense PCF.

Ratios: O. D. readings for the test sera + O. D. readings fen* the mean of the Canadian negative controls. 126 Table 11

Measurement of anti-procyclic surface antibodies and circulating trypanosomal antigens in sera of trypanosome-infected patients before and at various times after trypanocidal drug treatment.

Serology Days post- Parasitology T a TT Sandwich treatment3 ELISA Serum No. Patient (drug used) Remarks Blood smear (titre) (Ratio)b N o. i M-l 0 Relapsed sample + 1/20 1.31 2 M-2 147 (Mel B) 2nd treatment -- 1.24 3 M-3 231 - 1/10 1.15

4 M-4 425 - 1/40 6.25 5 M-5 445 1/160 4.39 455 patient died

6 N-l 0 Pretreatment + 1/40 10.08 7 N-2 6 1st treatment - 1/160 10.89 8 N-3 27 1st treatment - >1/320 7.94 9 N-4 67 - 1/20 8.58

10 N-5 138 - 1/160 6.21

11 N-6 262 - 1/40 4.86

12 N-7 438 - 1/80 10.40 13 N-8 603 (Mel B) Relapsed + >1/320 3.19 14 N-9 633 (Mel B) 2nd treatment - 1/80 10.17 15 N-10 899 )Multiple - 1/160 4.35 16 N -ll 1048 )relapses after - 1/40 3.62 17 N-12 1125 )Mel B therapy 1/80 4.32

' 27 1 Table 11 ( cont'd)

Serology Days post­ Parasitology PATT Sandwich treatment3 ELISA Serum No. Patient (drug used) Remarks Blood smear (titre) (Ratio)** N o. 1ft Neg. 0 Human serum ND - 0.88 (N. American.) 19 Neg. 0 Human serum ND - 1.34 (N. American.) 20 Neg. 0 Human serum ND - 1.02 (N. American.) 21 Pos. 7 Human serum ¥ >1/320 4.42 (Cote d'Ivoire) 22 Pos. 7 Human serum + 1/320 2.52 (Cote d'Ivoire) 23 Pos. 0 T. b. rhodesiense ND ND 5.79 water lysate (2x10^ tryps) 24 Pos. 0 T. b. rhodesiense ND ND 3.9 membrane proteins (1.0 ng) aDays post drug treatment: patients were treated with either suramin or Mel B for 5 weeks. In general, a test dose of 200 mg trypanocidal drug was given on the first day followed by a 1 g dosage. bRatio: O. D. readings for test sera + O. D. readings for negative controls. ND = not done. 8Z 1 8Z Table 12. Serological measurements on sera from African sleeping sickness patients from Daloa, Cote d'Ivoire, using the PATT, the CATT and the double antibody sandwich ELISA.

Clinic CSFa Parasitologye Serology file no. Cells** Total Trypd MHCT m- M icro­ GGP ELISAS CATT PATT (D-) Proteinc AECT filaria

1. 359 3 21 N.D. N.D. N.D. + 5.7 ++ 1/128 2 . 367 0 20 - + N.D. N.D. + N.D. ++ N.D. 3 . 370 0 20 - N.D. N.D. N.D. + N.D. + N.D. 4 . 373 1 12 - + + - - N.D. ++ N.D. 5 . 376 0 33 - + N.D. N.D. + 9.0 ± 1/32 6 . 377 0 14 - N.D. N.D. N.D. + 7.3 ± 1/16 7 . 378 1 33 - + N.D. N.D. + 5.8 + 1/32 8 . 383 0 22 - N.D. N.D. N.D. + 18.7 ++ 1/2048 9 . 389 0 28 - N.D. N.D. N.D. + 8.4 ++ 1/128 10. 394 10 32 - + + - + 13.2 ++ 1/256 11. 402 0 20 -- + - - 6.5 ++ 1/512 12. 421 0 23 - + ++ + + 13.6 ++ 1/32 13. 433 146 N.D. - + + - + 8.6 - 1/32 14. 435 2 22 -- - - + 3.3 + 1/128 15. 436 0 25 - - + -- 4.8 ++ 1/64 16. 437 4 16 - - + -- 6.8 ++ 1/256 17. 438 2 25 + - + - + 10.7 ++ 1/512 18. 439 0 30 - --- + 6.7 ++ 1/1024 19. 445 10 28 “ + + + + 8.7 + N.D. 29 1 Table 12. (cont'd)

Clinic CSFa Parasitologye Serology^ file no. Cells** Total Trypd MHCT m- Micro­ GGP ELISAS CATT PATT (D-) Protein0 AECT filaria

20. 446 6 17 + + + 9.7 + 1/256 21. 458 154 103 + + + - + 13.2 + 1/256 22 . 460 472 56 + N.D.N.D. - + 17.8 + 1/256 23 . 462 4 14 + + + - + 8.1 + 1/128 24 . 467 326 74 + + + - + 11.8 + 1/1024 25 . 490 1124 85 + -- - - 18.4 ++ 1/1024 26. 492 128 57 + + + - + 14.3 ++ 1/256 27. 494 242 42 + + + - + 7.7 ++ 1/512 28. 496 26 57 + + +++ - + 11.1 ++ 1/512 29 . 497 24 26 - + +++ - - 8.7 ++ 1/512 30. 498 164 48 + - + - - 6.5 ++ 1/512 31. 499 1198 89 + + +++ - + 12.0 ++ 1/512 32. 500 4 18 - + +++ -- 12.7 ++ 1/64 3 3 . 501 6 25 + + +++ -- 7.9 ++ 1/512 34. 502 132 51 ++ + ++ + + 6.8 ++ 1/512 35. 503 382 58 ++ - + -- 16.9 ++ 1/128 36. 504 16 38 + + +++ + + 10.8 ++ 1/256 3 7 . 505 78 32 + + ++ - + 10.5 ++ 1/512 38. 508 154 65 ++ + + - - 6.3 ++ 1/128 3 9 . 509 214 80 + + +++ - - 6.3 + 1/64 DC1 HCN.D. N.D. N.D. N.D. N.D. N.D. N.D. 1.2 - - DC2 TWP N.D.N.D.N.D. N.D. N.D. N.D. N.D. 0.8 0 3 1 Table 12. (cont'd)

Clinic CSFa Parasitologye Serology* file no. Cells** Total Trypd MHCT m- Micro- GGP ELISAS CATT PATT (D-) Protein6 AECT filaria

DC3 Toxo. N.D. N.D. N.D. N.D. N.D. 1.1 DC4 Toxo. N.D. N.D. N.D. N.D. N.D. 1.0 b b b Z Z 2 DC5 Leish. Z Z 2 b b b N.D. N.D. N.D. N.D. N.D. 1.2

PBS 0.6 TBRP1 water lystes (2 x 105 trypanosomes) 5.1 TBRP1 membrane 1 pg 3.1 a Cerebrospinal fluid obtained by lumbar puncture. ** Lymphocytes/mm^. c mg protein/dl. d Presence of trypanosomes by double centrifugation in capillary tubes. e Parasitolology MHCT: microhematocrit centrifugation technique. m-AECT: mini-Anion-Exchange centrifugation technique. Microfilaria: presence of blood microfilaria. GGP: results of microscopic examination of lymph node biopsy material; + presence of parasites; - absence of parasites; N.D., biopsy not done since no adenopathies seen, f Serology PATT: Procyclic Agglutination Trypanosomiasis Test. CATT: Card Agglutination Trypanosomiasis Test 8 ELISA: Table shows the absorbance reatios of the sandwich ELISA. Absorbance ratio = O. D. for test sera + O. D. for North American controls. N.D., not done. Table 13. Cellular, biochemical, parasitological and serological measurements on sera from trypanosome-infected Sudanese patients

Seroloevt Parasitoloev** CSF*** CATT PATT Sandwich Blood Lvmph ELISA Group* Protein Serum ABS No. WBF TS m-AETC LJ WBC mg/dl Blood (titre) (dtre) ratio

PI N D tt ND T 500 80.0 0.0 1/64 >1/320 14.77 P2 - -- + 123 52.8 2.0 1/64 >1/320 15.40 P3 --- + 7 20.4 2.0 1/16 1/80 11.57 P4 -- + 1 15.9 2.0 1/32 1/160 11.67 P5 + - ND ND 10 25.7 2.0 -- 3.97 P6 - - - + 2 17.0 2.0 1/8 1/20 3.59 P7 - + - + ND 123.8 2.0 1/16 1/20 3.01 P8 - -- + 54 45.2 2.0 1/32 1/160 8.75 P9 + + + + 3 15.5 2.0 1/8 1/20 4.01 P10 - + + + 4 24.9 2.0 1/16 1/40 6.11 P ll - + -- 16 21.9 2.0 1/32 >1/320 12.88 P12 ND + ND + 162 49.0 2.0 1/8 1/80 8.00 P13 ND + _ + 16 20.8 2.0 1/32 1/20 6.80 P14 + - ND + 22 20.4 ND 1/32 ND ND P15 + + ND + 52 47.2 2.0 1/8 1/20 11.72 P16 -- ND + 2 16.2 2.0 1/16 1/80 8.56 P17 -- ND + 4 24.7 2.0 1/32 1/320 8.83 P18 - + ND + 4 26.0 2.0 1/16 1/160 7.14 P19 + + ND + 75 49.1 2.0 1/16 1/320 8.36 P20 + + ND + 21 29.4 2.0 1/32 1/40 4.86 P21 + + ND + ND 31.3 2.0 1/32 1/160 8.76 P22 + + ND + 34 47.9 2.0 1/16 1/80 9.96

P23 - - ND + 16 49.4 2.0 1/16 1/160 3.91 132 Table 13 (cont’d)

Serologvt Parasitoloev** CSF*** CATT PATT Sandwich Blood LymDh ELISA Group* Protein Serum ABS No. WBF TS m-AETC U WBC mg/dl Blood (titre) (titre) ratio

P24 ND + 3 26.4 2.0 1/16 1/320 14.31 P25 -- ND + 12 30.6 2.0 1/16 1/320 4.41 P26 - - ND + 2 19.3 2.0 1/16 1/160 8.59 P27 -- ND + 154 44.5 2.0 1/8 1/80 5.21 P28 + + ND + 5 27.6 2.0 1/16 1/20 9.02 P29 - ND “ + 10 32.4 2.0 1/16 1/80 6.00

* group: p = parasitological cases ** parasitology: WBF = wet blood film TS = thick smear m-AETC = micro-anion-exchange thin column LJ = lymph juice *** CSF: WBC = number of white blood cells per microlitre protein = total protein in mg per 100 ml t serology: CATT = Card Agglutination Trypanosomiasis Test 0 = negative 2.0 = positive 1.0 = weakly politive PATT = Procyclic Agglutination Trypanosomiasis Test ELISA = Absorbance ratios of the sandwich ELISA Abs. ratio = O.D. for test sera/O.D. for controls t t ND = not done 133 Table 14. Cellular, biochemical, parasitological, and serological measurements on sera from trypanocidal drug treated human sleeping sickness patients from Sudan

Previous trypano­ Seroloevtt somal historv** Parasitology*** CSF+ CATT PATT Sandwich Treated Blood Lvmoh ELISA Group* Parasite since Protein Serum ABS no. blood (weeks) WBF TS m-AETC LJ WBC mg/dl Blood (titre) (tide) ratio

0 1 WBF+ 211 ND 2 21.5 1.0 1/4 1/10 1.30 0 2 GP+ 116 - - ND ND 3 24.2 2.0 ND ND ND 0 3 pos 211 --- ND 3 15.5 0.0 0.0 - 1.46 0 4 TS+ 116 - -- ND 4 52.4 2.0 0.0 - 1.05 0 5 TS+ 116 - - - ND 2 25.7 ND 0.0 - 3.40 0 6 WBF+ 217 - - - ND 2 24.2 2.0 1/16 ND ND 0 7 TS+ 48 --- ND 2 12.5 2.0 1/4 1/80 4.67 0 8 GP+ 116 - - ND ND ND ND 0.0 1/2 - 1.05 0 9 GP+ 116 - - ND - 1 26.8 1.0 1/4 - 7.84 0 10 TS+ 127 -- ND ND ND ND 0.0 0.0 - 9.58 0 11 TS+ 124 -- ND ND 2 19.3 0.0 0.0 - 1.49 0 12 GP+ 125 ND - ND ND 4 23.0 0.0 0.0 1/20 1.83 0 13 GP+ 33 ND - ND ND 9 26.0 0.0 0.0 1/20 6.29 0 14 pos 28 ND - ND ND 2 29.8 0.0 0.0 - 0.91 0 15 GP+ 9 - - ND ND 16 52.5 2.0 1/8 1/10 1.56 0 16 GP+ 92 - - ND ND 4 21.9 2.0 1/4 1/20 2.91 0 17 GP+ 116 - - ND ND ND ND 0.0 0.0 1/40 1.76 0 18 GP+ 116 - - ND ND 4 15.9 1.0 1/4 1/160 1.80 0 19 GP+ 116 -- ND ND 3 20.8 0.0 0.0 - 0.98 0 20 GP+ 116 -- ND ND 2 16.2 0.0 0.0 - 1.09

co Table 14 (cont'd)

Previous trypano­ Seroloevtt somal historv** Parasitoloev*** CSFt CATT PATT Sandwich Treated Blood Lvmph ELISA Group* Parasite since Protein Serum ABS No. blood (weeks) WBF TS m-AETC LJ WBC mg/dl Blood titre titre ratio

0 21 GP+ 116 ND 3 16.6 2.0 1/8 1/160 5.35 0 22 GP+ 84 - - ND ND 10 33.6 0.0 0.0 1/40 1.17 0 23 GP+ 116 - - ND ND 6 65.7 2.0 ND ND ND 0 24 GP+ 217 - - ND ND 8 33.6 2.0 1/4 - 1.41 0 25 GP+ 217 - - ND ND 1 29.8 0.0 0.0 1/20 2.32 0 26 GP+ 116 -- ND ND ND 55.1 2.0 0.0 1/80 5.67 0 27 GP+ 211 -- ND ND 2 24.9 0.0 0.0 1/40 1.57 0 28 GP+ 116 -- ND ND 4 40.2 2.0 0.0 1/20 7.37 0 29 GP+ 211 - - ND ND ND ND 2.0 0.0 >1/320 12.47 0 30 GP+ 116 -- ND ND 1 27.2 0.0 0.0 1/80 4.90 0 31 GP+ 116 -- ND ND 1 39.3 1.0 0.0 1/40 1.42 0 32 GP+ 211 - ND ND ND 3 16.6 2.0 0.0 1/320 3.01 0 33 GP+ 116 - - ND ND 1 22.3 0.0 0.0 - 1.61 0 34 GP+ 116 -- ND - 2 35.5 0.0 0.0 - 1.57 0 35 GP+ 60 -- ND ND 52 38.1 2.0 1/16 1/160 3.51 0 36 pos 116 - - ND ND 32 70.0 2.0 ND ND ND 0 37 pos 116 -- ND ND 1 21.5 2.0 0.0 1/80 1.23 0 38 pos 116 - - ND 2 36.2 0.0 ND ND ND

* group: O = old cases ** previous trypano­ somal hist. WBF+ = wet blood smear positive GP+ = gland puncture positive TS+ = thick smear positive co *** parasitology: WBF wet blood film TS thick smear m-AETC micro-anion-exchange thin column LJ lymph juice t CSF: WBC number of white blood cells per microlitre protein total protein in mg per 100 ml f t serology: CATT Card Agglutination Trypanosomiasis Test 0 negative 2.0 positive 1.0 weakly politive . ^TT Procyclic Agglutination Trypanosomiasis Test ELISA Absorbance ratios of the sandwich ELISA Abs. ratio = O.D. for test sera/O.D. for controls t t t ND not done Table 15. Cellular, biochemical, parasitological and serological measurements on sera from serologically positive (CATT) patients from Sudan

Serologvt Parasitoloev** CSF* ** CATT PATT Sandwich Blood Lvmph ELISA Group* Protein Serum ABS No. WBF TS m-AETC LJ WBC mg/dl Blood (titie) (titre) ratio

S 1 ND 2 47.6 2.0 1/4 1/80 12.71 S 2 - - - ND ND ND 2.0 1/8 - 1.10 S3 -- -- ND ND 2.0 0.0 - 3.21 S 4 - -- ND 3 42.4 2.0 0.0 1/80 3.87 S 5 - - - ND 3 19.1 2.0 1/32 1/160 7.90 S 6 -- - - 1 41.5 1.0 1/8 1/40 3.11 S 7 --- ND 3 62.2 2.0 1/4 1/40 3.20 38 --- ND 5 23.8 2.0 1/16 1/160 6.58 S 9 - - - ND 4 35.9 2.0 1/4 1/80 7.02 S 10 --- ND 3 29.4 2.0 1/4 1/160 3.53 S .ll - -- ND 2 28.7 2.0 1/16 1/40 3.21 S ’7 --- ND 3 24.9 2.0 0.0 - 1.65 S 13 -- ND ND 1 32.1 2.0 1/4 1/20 2.75 S 14 ND - ND ND 1 14.0 2.0 1/4 - 5.08 S 15 ND - ND - 5 19.6 2.0 1/8 - 3.70 S 16 -- ND ND 1 28.3 2.0 ND ND ND S 17 - - ND ND ND ND 2.0 1/8 1/320 6.78 S 18 -- ND ND 4 33.6 2.0 0.0 1/80 3.50 S 19 - - ND - 2 12.8 2.0 0.0 - 3.70 S 20 -- ND ND 2 43.0 2.0 0.0 1/80 1.85 S 21 - ND - 2 17.0 2.0 0.0 1/160 3.01 S 22 -- ND ND 4 23.8 2.0 0.0 1/80 3.21

S 23 - - ND - 4 18.9 2.0 1/32 1/320 6.13 137 Table 15 (cont'd^

Seroloevt Parasitoloev** CSF*** CATT PATT Sandwich Blood Lvmph ELISA Group* Protein Serum ABS No. WBF TS m-AETC LJ WBC mg/dl Blood (titrc) (titie) ratio

S 24 -- - ND 173 61.7 2.0 1/64 1/80 5.98 S 25 ND - ND ND 6 44.5 2.0 1/8 1/320 6.27 S 26 - - ND ND 6 28.3 2.0 0.0 1/160 7.35

* group: s = serological cases ** parasitology: WBF = wet blood film TS = thick smear m-AETC = micro-anion-exchange thin column LJ = lymph juice *** CSF: WBC = number of white blood cells per microlitre protein = total protein in mg per 100 ml t serology: CATT = Card Agglutination Trypanosomiasis Test 0 = negative 2.0 = positive 1.0 = weakly politive PATT s= Procyciic Agglutination Trypanosomiasis Test ELISm = Absorbance ratios of the sandwich ELISA Abs. ratio = O.D. for test sera/O.D. for controls t t ND = not done 138 Table 16. Cellular, biochemical, parasitological and serological measurements on sera from uninfected Sudanese, uninfected North Americans and human sleeping sickness patients from Daloa, Cote d’ Ivoire.

J. Seroloev t Parasitoloev** CSF*** CATT PATT Sandwich Blood Lvmph ELISA Group Protein Serum ABS No. WBF TS m-AETC LJ WBC mg/dl Blood (titre) (titre) ratio

N 1 Microscopy lab ND ND ND 1 0.21 0.0 0.95 N 2 Microscopy lab ND ND ND ND ND ND - 0.0 - 0.94 N 3 Microscopy lab ND ND ND ND ND ND - 0.0 - 1.04 N 4 Microscopy lab ND ND ND ND ND ND - 0.0 - 0.94 N 5 Microscopy lab ND ND ND ND ND ND - 0.0 1/40 2.25 N 6 --- ND 2 36.2 2.0 0.0 1/320 3.53 N 7 --- ND 1 11.3 2.0 0.0 1/20 1.59 N 8 --- ND 1 37.4 2.0 ND - 5.02 N 9 -- ND - 2 17.4 2.0 ND 1/20 1.23 N 10 - ND ND - 4 10.25 2.0 0.0 ND ND N 11 - - ND ND 5 10.7 2.0 0.0 1/10 1.45 V 1 Human serum (NA) ND ND ND ND ND ND ND ND - 0.95 V 2 Human serum (NA) ND ND ND ND ND ND ND ND - 0.97 V 3 Human serum (NA) ND ND ND ND ND ND ND ND - 1.03 V 4 Human serum (NA) ND ND ND ND ND ND ND ND - 1.05 V 5 HSSS (Ivory Coast) ND + + - 26 ND ND 1.0 >1/320 . 11.10 V 6 HSSS (Ivory Coast ND + + - 214 ND ND 0.5 1/80 6.30 PBS 0.73

TBRPI water lysates (2 x 105 trypanosomes) 5.41

TBRP1 membrane (1 pg) 4.06

------Cj3 * group: N negative controls from Sudan V controls from Victoria (Canada) ** parasitology: WBF wet blood film TS thick smear m-AETC micro-anion-exchange thin column U lymph juice *** CSF: WBC number of white blood cells per microlitre protein total protein in mg per 100 ml t serology: CATT Card Agglutination Trypanosomiasis Test 0 negative 2.0 positive 1.0 weakly politive PATT Procyclic Agglutination Trypanosomiasis Test ELISA Absorbance ratios of the sandwich ELISA Abs. ratio = O.D. for test sera/O.D. for controls t t ND not done t t t NA North America

o 141 Table 17. Detection of anti- trypanosome antibodies and circulating trypanosomal antigens in sera of trypanosome-confirmed, trypanocidal drug-treated or serologically (CATT) positive Sudanese.

Trypanosome Drug-treated Serologically Control* Control@ -confirmed positive (Mean±S.D.) (Mean±S.D.) (Mean±S.D.) (Mean±S.D.) (Mean±S.D.)

ELISA 8.18 ±3.15 3.02 ± 2.70 4.65 ± 2.49 1.87 ± 1.30 1 ± 0.04 Ratios^ PATT (titres) 1/145.71 ± 1/50.58 ± 1/104.80 ± 1/41.00 ± 0 ± 0 1/119.00 1/82.00 1/97.20 1/93.8 CATTt 1/20.80 ± 1/2.17 ± 1/8.76 ± 0 ± 0 (1/1.0 ND 1/15.40 1/4.12 1/13.98 ± 1/0.99)**

* Sera from 11 presumably uninfected Sudanese. @ Sera from 4 healthy North Americans. # Ratios: O. D. readings for test sera + O. D. readings obtained with the mean of the North American controls, t Aggultination results using CATT with Sudanese’ sera. ** CATT results using Sudanese' whole blood. The reaction was scored as "2" (strong agglutination), "1" (weak agglutination) and "0" (no agglutination). 1 42 Table 18

Measurement of antibodies and antigens in sera from patients with different parasitic diseases.

Patient's file # Parasitic PATT ELISA (ratios**) Disease*

J ' i ' ' Schistosomiasis - 1.03 J - 2 Amoebiasis / - 1.08 Schistosomiasis J - 3 Malaria 1/20 1.60 J - 4 Amoebiasis - 1.27 J * 5 Toxoplasmosis - 1.24 J - 6 Echinococcosis / * 1.52 Amoebiasis J - 7 Amoebiasis - 0.90 J - 8 Echinococcosis / - 1.13 Amoebiasis J - 9 Amoebiasis - 1.29 J - 10 Trichinosis - 0.94 J - 11 Echinococcosis - 1.39 J - 12 Amoebiasis - 1.19 J - 13 Filariasis - 0.98 J - 14 Amoebiasis - 0.84 J - 15 Filariasis / - 0.94 Cysticercosis J - 16 Toxoplasmosis 1/40 2.24 J - 17 Chagas' disease 1/20 1.35

NHS@ 1 - 1.02 NHS@ 2 - 0.96 W. Africa patient Gambian human 1/160 7.45 1 sleeping sickness W. Africa patient Gambian human 1/320 10.94 2 sleeping sickness

PBS _ 0.93 TBRP1 water lysate (5xl(H tryps) - 4.05 TBRP1 membrane proteins (0.05 3.76 l*g)

@ Normal human sera from healthy, uninfected North Americans. * Patients were diagnosed either parasitologically or serologically using appropriate ELISA. ** Ratio = O. D. reading for test sera + O. D. reading for the average of the uninfected North Americans. 143 Discussion

The potential uldlity of non-variant antigens, shared among bloodstream form and procyclic form trypanosomcs, in the immunodiagnosis of African trypanosomiasis has been demonstrated in several studies (Sailyo et al. 1980, Rae and Luckins 1984; Chapter 1 and Chapter 2). Results presented in Chapter 3 show that both PATT and sandwich ELISA tests were effective in revealing the infection status of T. b. rhodesiense -infected vervet monkeys and patients from Lamb we Valley, western Kenya. Trypanesomal antigens were detected in sera from T. b. rhodesiense -infected vervet monkeys as early as 7 days post­ infection. Thereafter, antigens remained at detectable levels in all sera from monkeys with active, untreated infections with the exception of sera nos. V21, V34 and V38 at days 28, 33 and 56 post-infection, respectively (Table 9). Interestingly, both sera nos. V21 and V34 showed a high anti-trypanosome antibody titre (Table 9). It is possible that parasite antigens in these sera were in the form of immune complexes, which are often found in the blood and CSF of patients with human African trypanosomiasis (Lambert et al., 1981), and which could hinder antigen detection by sandwich ELISA. After 33 days of trypanosome infection, detectable antigen increased moderately in all sera until drug treatment (Table 9).

Similarly, sera of all trypanosome-infected Lambwe Valley patients taken prior to drug treatment (with the exception of serum # 1; Table 10) gave positive results in sandwich ELISA while sera from uninfected controls did not (Table 10). The high absorbance ratios observed in most of these patient's sera suggest the presence of a large amount of trypanosomal antigen (Table 10). This is not surprising because fill of these patients had trypanosomes in their blood prior to trypanocidal drug treatment (W elide et al. 1989b).

In part through the action of trypanocidal drugs, trypanosomes are quickly eliminated from the bloodstream of the infected host, although residual trypanosomes may remain in several organs of the host and may mitiate bloodstream infections from such sites (Poltera, 1985). A rapid decrease in the amount of circulating antigens is expected as the disease begins to regress. Such a decrease in the detectable antigen level (as measured by the sandwich ELISA) was in fact observed in sera taken from trypanosome-infected vervet monkeys shortly after trypanocidal drug treatment (14 days post-treatment; Table 9). Antigens were not detected in sera as early as 27 days post-drug treatment (sera nos. V23, V24, V30 and V31; Table 9). With the exception of serum no. V2, which was taken 933 1 4 4 days after drug treatment, none of the sera taken long after successful drug treatment showed any detectable trypanosomal antigens. Since this serum was taken from monkey No. 35 which gave no indication of a relapse, the ELISA result must be interpreted as a false positive (Table 9).

A decrease in detectable trypanosomal antigen levels in the sera of drug treated patients from Lambwe Valley was also observed (Fig. 12A). Trypanosomal antigen levels, as detected by the sandwich ELISA, disappeared by 7-41 days post-drug treatment in sera from patients that were subsequently confirmed drug-cured (Fig. 12A). Conversely, a persisting level of detectable trypanosomal antigens was observed in sera of patients who subsequently relapsed, particularly those that relapsed shortly after drug treatment (relapsed by days 74-250 post-treatment, Fig. 13A). With patients who relapsed long after the drug treatment (between 291-833 days post-drug treatment), a fluctuation antigen profile was observed in that a drop in antigen to a very low or undetectable level was followed by an incre ase in antigen to a clearly detectable level in sera taken shortly before relapse (Fig.l4A). It is possible that the initial drop in antigen levels resulted from the near-elimination of the parasites through the action of trypanocidal drugs and that the relapse and subsequent rise in antigen level resulted from the proliferation of a few surviving trypanosomes. Alternatively, these patients could have been cured by the first drug treatment and were reinfected later. One patient (patient N, Table 11) showed multiple relapses after the second drug treatment Persisting detectable trypanosomal antigen levels were observed in this patient even long after the second drug treatment (days 492 post- second drug treatment).

Only one relapsed Lambwe Valley patient (patient M) showed undetectable trypanosomal antigen in sera taken during relapse (serum # 1, Table 11) and the same negative results were found in the two serum samples (sera # 2 and 3; Table 11) taken from this patient shortly after secondary drug treatment. Because sera # 2 and 3 were taken long after secondary drug treatment (days 147 and 231 post-drug treatment, respectively, trypanosomes may have been eradicated from the host during the sampling period. This interpretation is supported by the presence of low anti-trypanosome antibody levels (1/10), as detected by the PATT, in these sera. A sudden increase in both trypanosomal antigen levels (ELISA ratios from 1.15 to 6.52) and anti-trypanosome antibody levels were observed 10-30 days prior to the patient's death from trypanosomiasis (Table 11). These results underline the accuracy of the two tests. 1 4 5

All preinfection sera collected from vervet monkeys or mice (Table 9) or control sera from uninfected individuals (Table 9 and 10) were negative while trypanosome- infected monkeys and parasite-confirmed patients were positive. Results from the sandwich ELISA thus correlated with the infection status of vervet monkeys or humans infected with T. b. rhodesiense.

The rabbit antibodies used in this sandwich ELISA were specific to T. brucei spp. and T. congolense and, in this study, they showed no reactivity to Leishmania lysates (Fig. 9). This implies that the assay is potentially useful in the diagnosis of human sleeping sickness caused by T. b gambiense and T. b. rhodesiense and also of Nagana in cattle caused by T. b. brucei and T. congolense. Further testing of well-documented sera from T. b. gambiense-infected patients from Cote d'Ivoire and from Sudan have clearly demonstrated the diagnostic potential of this ELISA assay for T. b. gambiense infections. Trypanosomal antigens were detected in all sera from T. b. gambiense sleeping sickness patients from Cote d'Ivoire (Daloa) (Table 12) and parasitologically-confirmed Sudanese cases (Table 13), but not in control sera from uninfected North Americans or sera from Toxoplasma and Leishmania-infected patients (Table 12). The high absorbance ratios (ratios between 4.0 and 18.7) observed in most of these patients seems to reflect the abundance of parasite antigens. This is confirmed by the presence of trypanosomes in the blood or in the CSF of these patients (Tables 12 and 13).

Our results confirm the value of the CATT in diagnosis of T. b. gambiense infections (Van Nieuwenhove, 1983; WHO, 1986; Zillman etal., 1986). Anti- trypanosome antibodies were detected by the CATT in sera from parasitologically confirmed, T. b. gambiense-infected Daloan (Table 12) and Sudanese patients (Table 13) and in sera from patients from Sudan with prevalent clinical signs of human sleeping sickness (S24 - S26; Table 15). However, persistently detectable antibody levels were also observed in 20 out of the 38 trypanocidal drug treated patients (Old cases, Table 2) and 6 of the 11 presumably uninfected Sudanese gave positive agglutination reactions in CATT (Table 4). The number of false positives decreased when sera, rather than whole blood, were used in CATT (11/38 and 0/11 positives in CATT for drug treated patients and uninfected Sudanese, respectively (Table 2 and 4). 1 4 6 The use of procyclic trypanosomes and antibodies made against them in PATT and sandwich ELISA, respectively, for detecting anti-trypanosome antibodies and trypanosomal antigens, gave results that were broadly similar to those observed in the CATT. However, both the PATT and sandwich ELISA appeared to be more sensitive in that positive results were obtained in all 39 trypanosome-infected Daloan patients using these two assays, while serum from patient No. D433 showed no agglutination and sera from patients Nos. D376, D377 resulted only in a marginal agglutination reaction in CATT (Table 12). Sera from all 28 trypanosome-confirmed Sudanese patients (Table 13) also showed detectable antigen levels using the sandwich ELISA. Only sera P6 and P7 of Sudanese patients (Table 13) gave low detectable trypanosomal antigen levels (ratios of 3.59 and 3.01; respectively) in sandwich ELISA. The low antibody ti'cres (1/20) detected by the PATT in these 2 sera suggest that they were collected soon after onset of the infection in these patients. All parasitologically proven Sudanese patients, with the exception of one (P5; Table 13), gave positive agglutination titres in PaTT. The same patient (P5) also gave a negative result in CATT when sera were used. It is possible that this serum sample was collected at early stage of the infection when the anti-trypanosome antibody response was not yet induced or at a low level.

The PATT and sandwich ELISA gave results which correlated with the CATT in sera of Sudanese patients who showed some clinical signs of trypanosome infection but lacked trypanosomes in their blood and CSF samples and had no history of trypanosomiasis (Serological cases; Table 15). Of the 17 serological case patients who had elevated protein levels (> 25 mg/dl) (some of which showed an increased number of white blood cells in their CSF samples), 14 were positive in sandwich ELISA while 15 and 17 patients were positive in PATT and CATT, respectively. The presence of antibodies and antigens in sera of these patients, detected by the CATT, PATT and the sandwich ELISA, strongly indicated a positive infection status for these patients. However, three of these CATT positive samples (S2, S15 and S19) showed no detectable levels of anti- trypanosome antibodies with the PATT. Because variant surface glycoproteins, in general, are more immunodominant than the common antigens in the bloodstream trypanosomes (Turner, 1985), this discrepancy may result from temporal differences in the elicitation of humoral responses.

Results obtained with the CATT, PATT and sandwich ELISA differed in sera from trypanocidal drug treated patients (Table 14) and sera from uninfected Sudanese (Table 16). 1 4 7 Trypanosomal antigen levels, as detected by the sandwich ELISA, were less persistent than the anti-trypanosome antibodies levels, as detected by the CATT and the PATT, in drug treated patients (14/33 in ELISA vs 20/35 in CATT and 20/33 in PATT). Among these serologically positive drug treated patients, only 9 sera had high levels of trypanosomal antigens (Absorbance ratios > 4.0) while 16 sera in the CATT showed strong agglutination reactions. The persistent antigen levels detected by the sandwich ELISA in some of these drug treated patients' sera could reflect the presence of chronic Gambian form infections in these patients. This was confirmed by the persistently high antibody titres, detected by both the PATT and the CATT, in sera of these Sudanese patients (07,09,021,026,028, 029,032 and 035; Table 14). However, the presence of detectable antigen levels, in the absence of detectable antibodies, in two patients long after drug treatment (05 and 010; Table 14) is indeed puzzling. It is possible that the sandwich ELISA is detecting an early reinfection or anti-idiotypic antibodies produced by these patients during the course of the trypanosome infection. Only 3 of 11 uninfected Sudanese showed detectable antigen levels (N5, N6 and N8; Table 16) while 4 of these 11 individuals gave low antibody titres (1/10 -i/40) and 1 of these 11 Sudanese gave high antibody titres (1/320) in PATT. On the contrary, 6 out of the 11 uninfected Sudanese gave positive agglutination reactions in two repeated experiments with CATT testing of blood samples. However, none of these CATT positive samples gave any agglutination when tested with sera samples in CATT. The cause of this discrepancy is unknown. One of the antigen positive sera (N5; Table 16) also gave positive results in indirect immunofluorescence using fixed bloodstream trypanosomes (data not shown) while other two antigen positive sera had elevated protein levels in their CSF samples and they also showed high agglutination titres in either PATT (N6; Table 16) or CATT (N8; Table 16). The "uninfected" status of these 3 Sudanese patients is thus questionable. Together these results suggest the sandwich ELISA gave a better correlation to the infection status of individuals than the PATT or the CATT.

The present data clearly demonstrate the diagnostic value of the PATT and the double antibody sandwich ELISA using rabbit antibodies against whole lysates of procyclic trypanosomes in both T. b. gambiense and T. b. rhodesiense infections. In addition, these tests are specific for African sleeping sickness in that very few false positives were observed in sera from patients with other parasitic diseases (Table 18) and sera from control individuals in endemic areas (Tables 10,17). The sandwich ELISA appears to be a more precise method than the PATT or the CATT for pinpointing an active infection since there is a shorter persistence of antigen than there is of anti-trypanosome 1 4 8 antibodies in sera taken after drug treatment (Tables 9,14) and fewer numbers of false positives were observed (Tables 10,17 and 18). However, the simpler reaction format of the PATT might render it more suitable for development of a modified assay useful in the wilds of Africa. Identification and biochemical characterization of the relevant antigens are needed for further refinement and adaptation of these two existing assays to a simpler format for field application. 1 4 9 Chapter 4. Identification of procyclic trypanosomal antigens that have serodiagnostic potential for human sleeping sickness.

Introduction

Although it has long been recognized that trypanosome-infected mammalian hosts produce antibodies against both VSG and non-variant, common parasite antigens (Gray, 1960; De Raadt, 1974a), the identity of most of these common antigens has not been determined. Data presented in chapters 1-3 have demonstrated the diagnostic potential of procyclic trypanosomes in agglutination tests for detection of antibodies and specific anti-procyclic antibodies in antigen trapping assays for detection of trypanosomal antigens. Both antibody detection and antigen detection assays could possibly be improved by making monoclonal antibodies to well characterized antigens that have diagnostic potential. Identification of diagnostically useful antigens would expand our knowledge of trypanosomal common antigens and might also facilitate the adaptation of diagnostic assays to a simpler format for field application.

This chapter describes the identification and isolation of procyclic antigens with diagnostic potential. Trypanosomal antigens were identified and isolated by high performance liquid chromatography (HPLC), immunoaffinity purification and immunoblotting of T. b. rhodesiense ViTat 1.1 PCF lysates using pooled sera from trypanosome-infected patients (human sleeping sickness sera; HSSS) as a probe. The use of HSSS was based on the assumption that during the course of disease infected hosts will produce antibodies to those non-variant trypanosomal antigens that are circulating in the host system and that are immunogenic. Diagnostically important antigens have been identified in other parasitic diseases using antibodies produced by infected hosts (Kemp et al., 1983; Bowtell et al., 1984; Lanar et al., 1985). Monoclonal antibodies (MAbs) were then made against selected trypanosomal antigens and the diagnostic potential of the monoclonal antibodies was tested in an experimental murine model. 1 5 0 Materials and Methods

Parasites and parasite lysates

Procyclic culture forms (PCF) of trypanosomes were established from cloned bloodstream populations of T. b. rhodesiense ViTat 1.1 (Richardson etal., 1986), T. b. rhodesiense WraTat 3 (Hall and Esser, 1984), T. b. rhodesiense 1799 (from The International Laboratory for Research on Animal Diseases; ILRAD, Nairobi, Kenya) T. b. gambiense TREU 1285 (U2) (Gray, 1972), T. b. gambiense Th-1 (ILRAD), T. b. gambiense Th-17 (ILRAD),T. b. brucei LUMP 1026 (Bienen et al., 1983), T. b. brucei 427-01 (Roditi et al., 1987), T. b. brucei AmTat 1.1 (ILRAD), T. b. brucei MiTat 1.2 (ILRAD) using the methods of Brun and Schonenberger (1979). T. congolense 45/1 PCF were obtained from Dr. Reto Brun, Swiss Tropical Institute, Basel, Switzerland. Promastigotes of Leishmania major (NIH Seidman strain) were obtained from Dr. Neil Reiner, Vancouver, and promastigotes of L. clonovani IS2D were from a cloned population obtained through Dr. R. W. Olafson from Dr. Dennis Dwyer, NIH, Bethesda, MD, U.S.A. The parasites were grown at 26°C in SM medium (Cunningham, 1973) containing 10% foetal bovine serum (FBS) and 50 pg/ml gentamycin. The bloodstream forms of T. b. rhodesiense Vitat 1.1 (EATRO 1895) (Richardson et al., 1986) were grown in Cyclophosphamide- suppressed (Smith et al., 1982) Long-Evans rats (Charles River Breeding Laboratories, St Constant, Quebec.) and were purified from heparinized blood using Percoll (Grab and Bwayo, 1982) followed by chromatography on diethylaminoethyl cellulose (Lanham and Godfrey, 1970).

Whole lysates of trypanosome PCF were prepared by sonication as described in Chapter 3 at a final concentration of 10^ trypanosomes/ml of PBS (pH 7.4) containing protease inhibitors (1 mM TLCK (N-tosyl-lysine chloromethyl ketone; Sigma, St. Louis, Mo., USA), lmM TPCK (N-tosyl-phenyl-alanine chloromethyl ketone; Sigma, St. Louis, Mo., USA), lmM PMSF (phenylmethylsulfonyl fluoride; Sigma, St. Louis, Mo., USA), lmM Leupeptin ( Acetyl-L-leucyl-L-leucyl-L-arginal; Sigma, St. Louis, Mo., USA), lmM E64 (L-trans-epoxysuccinyl leucyl-amidol-4-guanidino butane; Sigma, St. Louis, Mo., USA) and 5 mM EDTA(Ethylenediamine-tetraacetic acid; Sigma, St. Louis, Mo., USA). Water lysates or membrane fractions of trypanosome PCF and Leishmania promastigotes were prepared as described in Chapter 2. Protein concentrations of the lysate were 1 5 1 determined by the Lowry method (Lowry et al., 1951) and the protein concentrations in membrane preparations were established using a modified lowry method (Peterson, 1983).

In separate experiments, trypanosomes were lysed in different solubilization buffers at 5 x 107/ml for PCF and 5 x lOfyml for bloodstream forms of T. b. rhodesiense ViTat 1.1 (TBRP1) for 20 mins on ice. Soluble antigens were assayed at 1/4,1/8,1/16 and 1/32 dilutions in an indirect ELISA and sandwicn ELISA (see below). This was done to compare the relative quantities of trypanosomal antigens recognized by the six anti-PCF MAbs (# 20,65,91,148,236 and 401) in the bloodstream and PCF trypanosomes. Controls included the different solubilization buffers: 1%, 2% and 3% butanol/ 50mM Tris- salinepH7.4; 1% CHAPS (3-{[3-chlolamidopropyl] dimethyl-aminio} 1-propane sulfonate; Sigma, St Louis, Mo., USA)/ 50mM Tris-saline pH7.4; 0.5?' Nonidet P-40 (NP-40; Sigma) /50mM Tris-saline pH7.4; 0.1% SDS/50mM Tris-saline pH7.4; 0.5% Triton X- 100 (TX100; Sigma)/ 50mM Tris-saline pH7.4); 1% Triton X-l 14 (TX114; Sigma)/50mM Tris-saline pH7.4; distilled water; PBS pH 7.4.

TX114 phase separation of TBRP1 antigens was preformed as described by Bordier (1981). For phase separation of TBRP1 proteins, a 2% TX114 solution was prepared from 10% stock by dilution in 20 mM Tris-saline, pH 7.5. Water lysates of T. b. rhodesiense ViTat 1.1 PCF (lOfyml) or bloodstream forms (lO^/ml) were diluted 1:1 with the 2% TX114 solution. The phase separation was performed by cooling samples on ice for 20 min, then warming them in a 37 °C water bath for 20 min, followed by centrifugation in a 37°C microfuge. Aqueous and detergent phases were separated accordingly.

The distribution of antigens recognized by the selected MAbs was determined by differential centrifugation of trypanosome lysates. Approximately 5ml of T. b. rhodesiense PCF or bloodstream forms (lOfyml in Tris-saline with 0.5% NP-40) were centrifuged at 1000 g, 10,000 g and 100,000 g for 60 mins. The supernatants were removed and the pellets were resuspended in 0.5% NP-40/Tris-saline with the same original volumes. Different supernatant and ^suspended pellet suspensions were assayed in indirect ELISA at 1/10 -1/ 12,800 using doubling dilutions. 1 5 2 Antisera and monoclonal antibodies

Sera were pooled from a total of 90 trypanosome-infected patients from the Lambwe Valley, western Kenya. These sera showed positive results in both Procyclic agglutination Trypanosomiasis Test (PATT) and sandwich ELISA (Chapter 3). Normal human sera were obtained from 4 healthy volunteers in Victoria. In addition, human sera were pooled from 37 Gambian patients from Daloa, Cote d'Ivoire (Chapter 1). Rabbit polyclonal antibodies were made against whole lysates of T. b. rhodesiense PCF as described in Chapter 3.

Six selected monoclonal antibodies (anti-PCF MAbs # 20,65,91,148,236,401) (see below) made against gel fractions of T. b. rhodesiense lysates were produced in the ascites fluid of pristane-primed BALB/c mice using standard techniques (Hoogenraad et al., 1983). Anti-urokinase or anti-human transferrin monoclonal antibodies produced in our laboratory ( R. Beecroft and T. Pearson) were used as negative controls.

The immunoglobulin (Ig) fractions of rabbit anti-PCF sera and ascites bearing the anti-PCF MAbs were concentrated by precipitation with a final concentration of 50% saturated ammonium sulphate. These fractions were then used for radio-labelling in competitive solid-phase radioimmunometric assays (see below) . Protein concentrations of the isolated Ig fractions were determined by the Lowry method (Lowry et al., 1951).

For antibody biotinylation, the IgG fractions of rabbit anti-trypanosome PCF antisera and 3 selected anti-PCF MAbs (# 20,148 and 236 )were isolated by ammonium sulphate precipitation and Protein A chromatography using Wood's (1984) methodology. Three anti-PCF MAbs (MAb # 65,91 and 401) of IgM isotype were semi-purified from ascites fluids using a Sephacryl S-200 gel permeation column (GPC) (bed volume = 425 ml). An isocratio elution with 0.01 M PBS (pH 7.4) was used and 4 ml fractions were collected in a model 2111 Multirac fraction collector (LKB, Bromma, Sweden) at a flow- rate of 1.0 ml/min. Eluant was monitored at 280 nm with a Beckman model 2000 spectrometer. Ig containing fractions were identified by ELISA using horseradish peroxidase coupled goat anti-n ouse IgM antibodies. Positive peaks were pooled and concentrated on a speed-Vac concentrator (Savant, Hicksville, NY, USA) to 1/3 of the' original pooled volume. The purified Ig fractions were labelled with biotin using a modified version of Focus' (1985) method. Biotin succinimide ester (CAB-NHS) 153 (Bethesda Research Laboratories, Burlington, Ontario) was dissolved at a concentration of 30 mg/ml in N, N-dimethylformamide (dried over a molecular sieve, 4A, 4-8 mesh; Aldrich, Milwaukee, Wis. USA). A 50 |il aliquot of this solution was added to 4 mg of the IgG fraction in 2 ml of PBS (preadjusted to pH 9.0 using a 0,5 M Na 2CX>3 solution). After incubating the mixture for 2.5 h at room temperature the reaction was stopped by the addition of 1 M MH 4CI2 (Fisher Scientific, Vancouver, B.C.) to a final concentration of 0.1 M. The mixture was then dialyzed overnight at 4°C against PBS containing 0.01% thimerosal (sodium ethyl mercurithiosalicylate) (Sigma, St Louis, Mo. USA) and stored at 4*C until used. The biological activities of the biotin-labelled antibodies were examined in indirect ELISA tests (see below) using stieptavidin-horseradish peroxidase complex (Amersham, Oakville, Ontario) to detect bound first antibody.

Pooled HSSS from the Kenyan patients and anti-PCF MAbs # 20,148 and 236 made against the PCF gel fractions were purified using ammonium sulphate and protein A columns as described above. These antibodies wete used for the preparation of the immunosorbent columns. Antibody purity was assessed by silver staining the SDS-PAGE gels.

Preparation of immunosorbent

Approximately 8 -16 mg of purified from immunoglobulin pooled human sera or anti-PCF MAbs (# 20 and 148) were coupled independently to 1 g of Tresyl-activated Sepharose 4B (Pharmacia, Uppsala, Sweden) according to the manufacturer's instructions.

Chromatographic separation of trypanosome lysates

Sonicated whole lysates of T. b. rhodesiense ViTat 1.1 PCF (5 x 10^ trypanosomes/ 5ml) were separated by GPC on a Sephacryl S-200 superfine column (47 cm x 2.5 cm inside diameters; Pharmacia Fine Chemicals, Uppsala, Sweden). An isocratic elution with 0.1 M sodium phosphate, pH 7.4, and 0.1 M potassium chloride was used and 4 ml fractions were collected in a model 2111 Multirac fraction collector (LKB, Bromma, Sweden) at a 1 5 4 flow-rate of 0.5 ml/min. Eluant was monitored at 230 nm and 280 nm with a Beckman model 2000 spectrometer. Immunoreactivities of these fractions were determined by indirect ELISA. Positive peaks were pooled, dialyzed against water, and lyophilized.

Lyophilized pooled fractions from GPC were dissolved in 10 mM Tris-HCl, pH 8.0, and separated using HPLC with a Mono Q HR 5/5 anion exchange column (50mm x 5 mm inside diameters; Pharmacia Fine Chemicals, Uppsala, Sweden). The HPLC system consisted of two Altex model I00A solvent-delivery systems, a Beckman 421 controller, a Beckman 164 Variable wavelength detector and a model BD 400 chart recorder (Kipp and Zonen, Delft, Holland). Gradients were generated with starting buffer A, 20 mM Tris-HCl pH 8.0, and buffer B, 20 mM Tris-HCl pH 8.0 containing 1.0 M NaCl, for the separation. The optimized gradient programmes used are given in Fig. 16. Eluants were monitored at 280 nm and fractions were tested by indirect ELISA. Immunoieactive peaks were then lyophilized.

Rev^rj e-phase HPLC (RP-HPLC) was performed using an Altex Ultrapore Analytical C-3 column (350 mm x 6.5 mm inside diameter). Immuno-positive peaks from the HPLC ion-exchange column were loaded on to the column in aqueous 0.1 % TFA (Trifluoroacetic acid) (HPLC grade; Pierce Chemical Company, Rockford, Illinois, USA). A stepwise gradient of increasing acetonitrile (HPLC grade; Fisher Scientific, Vancouver, B.C., Canada) concentration was used to elute proteins. The absorbance of eluant at 230 nm was recorded. Collected peaks were lyophilized and their immunoreactivities were determined by indirect ELISA.

Fractions collected after various separation procedures were analysed by SDS- PAGE with the addition of 0.02% mercaptoethanol. Silver and Coomassie blue staining methods were used to visualize the separate bands. Low molecular weight standards (Pharmacia, Uppsala, Sweden) were used for reference.

Antigen preparation

T. b. rhodesiense ViTat 1.1 PCF whole lysate was separated according to Laemmli (1970) using 10% SDS-PAGE reducing gels. Fractions # 1 ,2 and 3, respectively, corresponding to molecular weights > 62KDa, 41-62 KDa and 28-41 KDa, were excised from gels using 1 f 5 a scalpel. Prestained high molecular weight standards (BRL, Gaithersburg, MD) were used as inferences. Proteins from each fraction were eluted by incubating the minced gel slices (3x5 mm) with 10 volumes of distilled water overnight at 4 °C. Samples were concentrated to their original volumes by dialysis and lyophilization. The identity of extracted samples was confirmed by Coomassie blue staining and silver staining gels. Protein concentrations were determined by the method of Bradford (1976).

Immunization protocol

Two female BALB/c mice (Charles River Laboratories, St. Constant, Quebec, Canada) were used for each gel fraction. Mice were injected three times at monthly intervals with approximately 30 |xg of protein from each extracted gel fraction. Each mouse received a primary intraperitoneal inoculation (0.1 ml extracted proteins in 0.01 M PBS, pH 7.4, emulsified in 0.25 ml of Freund’s complete adjuvant) and two subsequent boosts in Freund's incomplete adjuvant Mice were tail-bled after the first and second boost and the anti-trypanosome antibody titres in sera were determined by indirect ELISA with the corresponding gel fractions and TBRP1 whole lysates. Three days prior to fusion with Sp2/0 myelomas, the tail veins of immunized mice were injected with 10 pg of antigen preparation in 200 pi of 0.01 M PBS (pH 7.4). All six mice injected with gel fraction # 3 died approximately 14 days after the primary inoculation, or the first boost, despite modifications to the immunization procedures, i.e. intramuscular injection, lower doses or injection without Freund's adjuvant. As a result, cell fusion was possible only with splenocytes from mice injected with gel fractions # 1 and 2.

Cell fusion and selection of hybridomas

Splenocytes (10®) from immunized mice were fused with Sp2/0 myelomas (10?) in the presence of 30% polyethlene glycol (PEG, MW 1450) (Sigma, St. Louis, Mo. USA) according to standard procedures (Galfrg and Milstein, 1981). Fused cells were incubated overnight at 37*C with 5% C 02,98% humidity, in 100 ml of fusion medium (RPMI1640, 20% heat inactivated bovine calf serum, IX Hypoxanthine/ Aminopterin/ Thymidine 1 5 6 [HAT], lmM Sodium Pyruvate, 5 x 10~^M 2-Mercaptoethanol, 50 jig/ml Gentamycin, 50 units/ml Streptomycin, Pencillin, 24 units/ml Mycostatin and 10?/ml syngeneic thymocytes). Ceils were aliquotted as 200 pi volumes into 96 well plates and were maintained in fusion medium (IX HAT) at 37°C, 5% CC2 and 98% humidity, until they were ready for screening (10-14 days). The remaining 25 ml of cells were then plated in petri dishes containing 2.2% methylcellulose (Terry Fox Laboratories , Vancouver, B.C.) in fusion medium according to the procedure of Davis et al. (1982). When hybridoma colonies had grown to 0.5 mm in diameter (10-14 days after plating), they were picked and transferred individually to 96 well plates in 200 pi of HT medium (fusion medium without Aminopterin). In general, hybridomas were grown in fusion medium for 2 weeks, followed by 2 weeks incubation in HT medium and then were maintained indefinitely in growth medium (fusion medium minus HAT, thymocytes and Mycostatin).

Hybridomas were screened initially using ELISA with detergent solubilized lysates and water lysates of both bloodstream and procyclic T. b. rhodesiense ViTat 1.1 as antigens. Positive clones were transferred to 24 well plates and tested for antibody binding to solubilized lysates and water lysates of other species of Trypanosoma (bloodstream form of T. b. rhodesiense, PCF of T. b. rhodesiense, T. b. gambiense, T. b. brucei and T. congolense) and Leishmania (promastigotes of L. major and L. donovani). Human transferrin (5 pg/well) was used in indirect ELISA to identify "sticky" antibodies. Dilution cloning was performed with hybridomas that secreted antibodies binding specifically to Trypanosoma spp., or strongly to both Trypanoscma and Leishmania spp. (i.e. Kinetoplastida specific). Stable clones were established immediately from their respective hybridoma cells after three dilution cloning steps and testing on ELISA using various parasite lysates as described above during each interval. Samples of these clones were frozen and kept in liquid nitrogen at 10? cells/ml of freezing medium (90% FBS, 10% DMSO).

Indirect ELISA

Pooled HSSS from Kenya were used in a solid phase indirect ELISA to probe for antigens of diagnostic utility. Indirect ELISA were performed as described in Chapter 2 with minor modifications. Fractions from GPC and ion-exchange columns were diluted in PBS to dilutions of 1/4,1/8,1/16 and 1/32. One hundred microlitres of these dilutions were used 1 5 7 to coat wells of an ELISA plate (Costar, Cambridge, MA, USA). RP-HPLC fractions were lyophilized, redissolved in 200 pi PBS and tested in ELISA . Samples were dried onto the wells by incubating the microplates overnight at 37*C. Rabbit polyclonal antiserum to the TBRP1 whole lysates was used as a positive control while normal human serum (NHS) was used as a negative control. Predetermined dilutions of human sera from trypanosome-infected patients (1/2000) and of rabbit antisera to the TBRP1 whole lysate (1/4000) were used as the first antibodies. Alkaline phosphatase labelled goat anti-rabbit or goat anti-human Ig (Tago Inc., Burlingame, CA, USA) were used as detecting antibodies at a dilution of 1/2000. After the addition of alkaline phosphatase substrate solution (Chapter 2), the plate was incubated for 30 mins at room temperature and color development was measured at 403 nm using an EIA auto-reader (Bio-Tek Instruments Inc., Burlington, WA, USA). Samples were considered positive if wells yielded absorbance values of at least thrice that of the negative controls.

Indirect ELISA was used to determine the binding specificities of monoclonal antibodies to different aqueous and detergent lysates. Trypanosoma or Leishmania parasites (10$) were pelleted, washed once in PSG and then lysed in 1 ml of distilled water or detergent solubilized buffer (Chapter 2). One hundred microlitres of 1/20 dilutions of lysates in PBS (approximately 10 pg/ml) were coated onto ELISA plates. Undiluted hybridoma tissue culture supernatant or a 1/800 dilution of ascites fluid were tested. Horseradish peroxidase-labelled goat anti-rabbit IgG/IgM (Caltag, South San Francisco, CA, USA) were used as detecting antibodies at a dilution of i/2000.

Trypanosome lysates treated with heat, proteases and chemicals (see below; deglycosylation of antigens) were assayed at doubling dilutions from 1/25 -1/25600 in indirect ELISA. Ascites fluid of anti-PCF MAbs # 20,65,91,148,236 and 401 were used as the first antibodies (1/800 dilution) and horseradish peroxidase-labelled goat anti­ mouse IgG/IgM (Caltag, San Francisco, CA, USA) were used as detecting antibodies at a dilution of 1/2000. 1 5 8 Isotyping

The isotype of monoclonal antibodies derived against TBRP1 gel fractions # 1 and 2 were determined by an antigen-capture-ELISA isotyping kit (American Qualex International, La Mirada, CA, USA). The procedure was carried out as described by the manufacturer. Undiluted hybridoma tissue culture supernatants were used for isotyping.

Immunofluorescence

Indirect immunofluorescence was performed on suspensions of living, 4% formaldehyde- fixed, 0.25% glutaraldehyde-fixed and acetone-permeabilized parasites. The procedure for living and 0.25% glutaraldehyde-fixed parasites were carried out as described previously (Pearson et al., 1981). Formaldehyde-fixed parasites were prepared by resuspending a pellet of 2 x 10^ cells with 10 ml of freshly made 4% formaldehyde solution an,' were incubated at room temperature for 1 hr. The fixative was then removed by centrifugation and the pellet was washed once with 10 ml PBS (pH 7.4). Fixed parasites (2 x 10^ cells/ ml of PBS containing 5% FBS) were processed for immunofluorescence as described previously (Pearson et al., 1981). Acetone-permeabilized parasites were produced by air- drying live parasites onto clean glass slides and then fixing them for 30 min in pre-chilled acetone at -20CC, prior to the rnmunofluorescence procedures. Neat tissue culture supernatants from hybridomas or ascites fluid (1/500 dilution) were used and the second antibody was 20 pi of 1/50 dilution of affinity purified goat anti-mouse Ig-FTTC (Gibco, Burlington, Ontario). Immunofluorescence was observed using a Zeiss binocular microscope fitted with an epifluorescence attachment and a Zeiss Neofluor 100/1.25 oil immersion objective. All photomicrographic steps were standardized to allow direct comparsions of the relative immunofluorescence intensities obtained with different MAbs. Photomicrographs were taken on TMAX 400 film (Kodak, Rochester, NY, USA) at xlOOO magnification using 60 second exposures. The same protocol was used to develop all negatives (20*C, 13.5 mins in D-76 developer 1:1) and to make positive prints (multigrade paper with filter 3 at 15 second exposures). 1 5 9 Immunoblotting

Electrophoretic transfer of proteins from 10% reducing SDS-PAGE gels onto Immobilon transfer membranes (Millipoie Corporation, Bedford, MA, USA) and subsequent antigen detection were performed by the procedure of Towbin et al. (1979), with modifications to reduce background (Birk and Koepsell, 1987; Bestagno et al., 1987). Undiluted hybridoma tissue culture supernatant or ascites fluids (1/800 dilution) were used as first antibody and either 1 ^5 labelled F(ab')2. portions of goat anti-mouse XgG (Fab')2 (made by R. W. McMaster and T. W. Pearson) or 125l-labelled gOEt anti-mouse IgM (BRL, Gaithersburg, MD, USA) was used as the detecting antibody. Immunoblots were exposed at *0°C (for 3-10 days) to pre-flashed Fuji RX X-ray film in a cassette containing intensifying screens. Radiographs were processed using an automated X-ray processor. Prestained high molecular weight standards (BRT Gaithersburg, MD, USA) were used as ref rence markers in immunoblots.

Heat treatment of parasite lysates

T. b. rhodesiense ViTat 1.1 PCF lysates (10^/ml lOmM Tris-HCl, pH 7.5) were heated at 100°C for 20 min “ollowed by quick cooling on ice for 20 mins. After microfuging for 5 mins, supernatants were assayed for antigen activity by indirect ELISA and immunoblciang.

Protease treatment of parasite lysates

Water lysate of T. b. rhodesiense PCF (2xl0^/ml) was mixed 1:1 with 20mM Tris-HCl (pH 7.5) containing 1.0% SDb to give a final concentration of 10^ trypanosomes / ml. one milligram of Proteinase A (Sigma, St Louis, Mo, USA) was added to 1 ml parasite lysate. The digestion was performed at 37’C for 60 min and the Proteinase K was subsequently heat inactivated at 100°C for 20 min. Pronase (Sigma) digestion was performed by adding 1 mg of the enzyme to 1 ml of premixed parasite lysate (10^/ml lOmM Tris-HCl, pH 7.5) at 37°C for 1 hr. The enzyme was then heat inactivated as described above. The effects of proteases on antigen activity were assessed by indirect ELISA and immunoblotting. 1 6 0

Deglycosylation of antigens

T. b. rhodesiense PCF lysate (10^ parasites/ml equivalent) was dried by lyophilization at room temperature. Anisole (1ml; Fisher Scientific, Vancouver, B.C.) was mixed with trifluoromethanesulfonic acid (2 ml) (TFMS; Pierce, Rockford, Illinois, USA) and the mixture was cooled in a diy ice-ethanol bath for 1 hr. The reaction was then carried out as described by Sojar and Bahl (1987). Prechilled Anisole/TFMS mixture was added to dried parasite pellet and transferred to a reactavial (Pierce, Rockford, Qlinosis, USA), purged with nitrogen and sealed. After stirring at room temperature for 1 hr, the mixture was precipitated with a 50 x volume excess of diethyl ether containing 10% hexane (v/v) for 1 hr in a dry ice-methanol bath. Two drops of pyridine were added to the mixture after 20 mins to help precipitation. The mixture was centrifuged at 2000 g for 5 mins and the pellet was washed once with prechilled (-20*C) ether and 95% ethanol. The pellet was vaccum dried and dissolved in 6 M guanidine-HQ. The reaction mixture was desalted by molecular sieving on a Sephadex G-25 (superfine) column (1 x 10 cm) (Pharmacia PD-10, Uppsala, Sweden), preequilibrated with 0.01 M PBS (pH 7.4). The fractions containing the deglycosylated protein were pooled and lyophilized. Antigenicity of the lyophilized TFMS treated samples was assessed by indirect ELISA and immunoblotting.

Concanavalin A binding study

A Concanavalin-A (Con A) agarose (Vector laboratories, Inc., Burlingame, CA, USA) column (5 ml) was equilibrated with 0.01 M Tris-saline, pH 7.5. T. b. rhodesiense PCF water lysate was mixed 1:1 with 2x Tris-saline to give a final concentration of 10^ trypanosomes/ml and was microfuged at 4*C for 5 mins. The supernatant (1 ml) was loaded on the equilibrated Con A column and the unbound material was washed off with 10 column volumes of 0.01 M Tris-saline, pH 7.5. Bound material was eluted with 0.2 M alpha-methylmannoside in Tris-saline. Fractions containing antigens recognized by the six anti-PCF MAbs were identified by indirect ELISA. 1 6 1

Trypanosome antigen purification

Trypanosome antigens recognized by the pooled human sleeping sickness sera, anti-PCF MAbs # 20 and 148 were purified using their respective antibody coupled affinity columns. Approximately 1 x 10*0 T. b. rhodesiense ViTat 1.1 PCF were washed once with PSG (pH 7.4) and lysed at a concentration of 2 x 10^ trypanosome s/ml in ice cold HPLC-grade water containing ImM PMSF/lmM TLCK. After freezing and thawing, the lysate was mixed 1:1 with 2 x PBS containing 1% Tween 80 detergent (Sigma, St. Louis, Mo, USA) (pH 7.4) and microfuged for 5 mins at 4*C. The supernatant was then applied at 4*C to irr tunoaffinity columns (flow rate 5ml/hr) which had been precycled with 5 volumes of the elution buffers and equilibrated with PBS pH 7.4 prior to processing of the parasite lysates. The breakthrough materials (10 ml) were then recycled once through the column prior to washing with 10 column volumes of PBS/0.5% Tween 80, pH 7.4, and 10 column volumes of PBS pH 7.4. Bound antigens were eluted as 1 ml fractions from MAb # 20 and MAb # 148 immunoaffinity columns using 20mM ammonium bicarbonate/1.0 M NaCl (pH 9.0) and 0.1M sodium citrate/ 0.5 M NaCl (pH 3.0), respectively. The immunoaffinity column made with Ig from pooled human sera was further washed with 5 column volumes of low pH buffer (0.1 M acetate/ 0.5M NaCl, pH 4.0) and 5 column volumes of high pH buffer (20mM Tris-HCl / 0.5 M NaCl, pH 8.0). Bound antigens were eluted from this column with 3M sodium thiocyanate/ 0.15 M NaCl (pH 7.0). Antigen- containing fractions were identified by indirect ELISA using their respective biotinylated MAbs. Positive fractions were pooled, desalted by dialysis against HPLC-grade water and concentrated using a speed Vac concentrator. The purity of immunoaffinity isolated materials were assessed by silver staining of 10% SDS-PAGE gels.

SDS-PAGE for sequence analysis

The immunoaffinity purified, desalted samples (1/3 of the total isolate) from MAb # 20 and MAb # 148 columns were loaded onto minigels containing a 7 -15% gradient of polyacrylamide and electrophoresed according to Laemmli (1970) at 20 Watts constant 1 6 2 power for 1 hr. After electrophoresis, the minigels (10 xlO cm, 1mm thick) were processed as described by Matsudaira (1987). Briefly, gels were soaked in transfer buffer (10 mM 3-[cyclohexylamino]-l-propanesulfonic acid, 10% methanol, pH 11.0) for 5 min and sandwiched between a sheet of Immobilon transfer membrane (Millipore Corporation, Bedford, MA, USA) and two sheets of Whatman 3 mm paper. They were then assembled into a blotting apparatus (Biorad, Richmond, CA, USA) and electroeluted for 30 min at 100 Volts in transfer buffer. The PVDF membranes were washed in deionized water for 5 min, stained with 0.1% Coomassie Blue R-250 in 50% methanol for 5 min, and then destained in 50% methanol, 10% acetic acid for 10 min at room temperature. The membranes were finally rinsed in deionized water for 10 min, air-dried, and stored at -20°C until used.

Amino acid microanalysis

The antigen containing Coomassie blue stained bands were cut from the Immobilon membranes and acid hydrolysed in 6N HC1 / 10% phenol at 165*C for 24 hrs. Amino acid microanalysis was performed using the Applied Biosystems Model 428 derivatizer-analyzer (Applied Biosystems, Foster City, CA, USA). The analysis itself was performed by Sandy Kielland in the University of Victoria Microchemistry center.

Sequence analysis

The antigen (on the Immobilon membrane) was sequenced by Sandy Kielland in the University of Victoria's Microchemistry Center using a gas-pnase sequencer (Model 470A, Applied Biosystems Foster City, CA, USA). 1 6 3 Competitive solid-phase radioimmunometric assays

Six MAbs were labelled individually with (i.o mCi, New England Nuclear, Thunder Bay, Ontario) using the Chloramine-T method (Greenwood et al, 1963). Twenty pg of each antibody were iodinated and competitive solid-phase radioimmunometric assays were used to determine the number of epitopes recognized Assays were performed as described by Hall and Esser (1984). The optimum antigen coating concentration (20 pg/ml) and amount of radiolabelled MAbs (5 x 10^ c.p.m./well or 10^ c.p.m./ml) needed were previously determined by checkerboard titration in a direct solid-phase radioimmunoassay. Wells of 96-well, U-bottomed polyvinyl chloride microtest flexible assay plates (Falcon, Oxnard, Ca., USA) were coated with 100 pi of trypanosome lysates (25 pg/ml in water) and dried as described previously. The wells were then washed 3 times with PBS/0.05% Tween 20 (PBS-Tween) and unoccupied sites on the plastic were blocked by incubation with 200 ill/well of PBS containing 3% (w/v) BSA (bovine serum albumin; Sigma, St. Louis, Mo., USA) at 37*C for 1 h. Fifty pi of unlabelled rabbit antiserum of MAb dilutions (10,1, 0.1,0.01, 0.001 and 0.0001 pg/ml in PBS, pH 7.4) and 50 pi of radioisotope labelled MAb (adjusted to 10^ c.p.m./100 pi with PBS-Tween, pH 7.4) were added simultaneously to each antigen-coated well. The antibody mixtures were incubated for 2 h at room temperature and the wells were washed thrice with PBS-Tween to remove unbound radiolabelled MAbs. The plates were then cut with a hot wire and individual wells were counted for 1 min in an LKB/282 Compugamma Universal gamma counter (Wallach Oy, Turku, Finland).

Double antibody sandwich ELISA

For this procedure, 6 MAbs (anti-PCF MAbs # 20,65,91,148,236 and 401) specific for distinct epitopes (determined by competitive solid-phase radioimmunometric assays) were selected. Sandwich ELISAs were performed as described before (Chapter 2). Unlabelled ammonium sulphate-precipitated rabbit anti-procyclic whole lysate antibodies (Chapter 3) or MAb mixture at predetermined dilutions of 1/4000 and 1/2000, respectively, were used as 'capture' antibodies while biotinylated anti-TBRPl rabbit antibodies or biotinylated MAb mixture at a respective dilutions of 1/400 and 1/2000 were used as 'detecting' antibodies. Mouse sera or mouse urine (diluted in PBS-Tween containing 1% BSA and 0.01% 164 thimerosal) were tested for antigen. Control wells contained b'iffer, normal serum or normal urine from uninfected animals. The presence of antigen-antibody complexes was determined by addition of streptavidin-biotinylated horseradish peroxidase complex (Amersham, Oakville, Ontario, Canada) and ABTS (2,2'-azino-bis {3-ethylbenzthiazoline sulphonic acid}) (Sigma, St. Louis, Mo. USA) substrate solution (50 ml 0.04 M citrate buffer, pH 4.0,250 pi 0.04 M ABTS and 200 p.130 , ’ H 2O2). The plate was incubated for 60 mins at room temperature and colour development was measured at 405 nm using an EIA autoreader (Bio-Tek Instruments Inc., Burlington, WA., USA).

In order to determine whether or not individual MAbs bound to the same molecule, a double MAb sandwich ELISA was performed as described above except that a single MAb was used as the capture antibody and the biotin-labelled detector antibody. The capture MAb was used at dilutions between 1/250 and 1/16000, whereas the labelled MAb was assayed at a dilution that was pre-determined by indirect ELISA to give high O.D. readings. Trypanosome PCF water lysates (25 |ig/ml) were used in these assays.

In order to optimize the conditions for double antibody sandwich ELISA, checkerboard titrations of unlabelled capture antibody and biotinylated detecting antibody were carried out, with the addition of 2.5 (Xg/well of trypanosome PCF whole lysate. The optimum dilutions of unlabelled capture antibody and the biotinylated detecting antibody were determined to be 1/800 and 1/500, respectively. Test sensitivity was defined by adding known concentrations of trypanosome PCF lysates (ranging from 10 to 10? trypanosomes/ml). Lysates of Leishmania major were also selected at the same dilutions.

Sera and urine from infected or uninfected mice were tested using the double antibody sandwich ELISA. Dilutions of 1/4,1/8,1/16.1/32 and 1/64 of sera in buffer were added to wells coated with either unlabelled anti-TBRPl MAb mixture or rabbit anti- TBRP1 whole lysate antibodies at a dilution of 1/2000 and 1/4000, respectively. Biotinylated MAb mixture or anti-TBRPl rabbit antibodies at respective dilutions of 1/2000 and 1/400, as stated previously, were used as the detecting antibodies. In addition, dilutions of known concentrations of trypanosome PCF water lysates were used as positive controls. Samples were considered positive (i.e. presence of antigen) if wells showed absorbance values of at least twice that of control averages. 1 6 5

Mouse sera and urine

Sera and urine from infected mice were examined in an attempt to detect trypanosomal antigens using the double antibody sandwich ELISA. Seventy BALB/c mice were injected subcutaneously with 1 x 1(P bloodstream form trypanosomes (7\ b. rhodesiense ViTat 1.1) in 0.2 ml of sterile physiological saline (0.15 M NaCl). Fifteen control mice received 0.2 ml of sterile physiological saline. Five mice were bled out at 3, 5, 8,10,12,15,17, 20,22 and 24 days post-infection. Five control mice were bled out at 0,12 and 24 days post-infection. Blood was collected from the chest cavity of each mouse and was allowed to clc: overnight at 4*C. After centrifugation at 500 g for 10 mins at 4°C, serum from each mouse was collected and stored separately at -20’C. Control mice were processed similarly Blood from each mouse was examined microscopically for the presence of parasites using undilutui, 1/10 and 1/20 dilutions of blood in PSG containing heparin (25 units/ml) and using thick dried blood smears. Urine was collected from u-ypanosome - infected mice using metabolic cages. Five mice were kept overnight, immediately prior to their blood collection, in a metabolic cage where urine were funnelled collectively into a bowl. Pooled urine was lyophilized, resuspended with water to a final volumes of 2 ml and kept separately at -20*C. Control mice were processed similarly. Ten pypanosome- infected mice were kept separately and had their urine monitored throughout the experiment Urine from five of these mice was collected as described above at different time intervals. 1 6 6 Results

Identification of procyclic trypanosome antigens recognized by sera from trypanosome-infected humans

PCF lysates of T. b. rhodesiense ViTat 1.1 were separated by GPC into five major peaks (Fig. 15). The molecular weights of these separated components were estimated to range from 1.5x10^ - 2.3 xlO^ using a calibrated curve constructed based on the eluted volume and standards with known molecular weights (data not shown). ELISA results indicated that the high molecular weight peak contained a majority of the antigens of interest (Fig. 15). Further separation of pooled fractions #s 40-50 was achieved using a DEAE /HPLC- column. The majority of these proteins was eluted between 25 -50 % 1.0 M NaCl, while a few were eluted between 85 - 100 % 1.0M NaCl (Fig. 16). Various elution conditions were attempted by either decreasing the salt gradient slope or using a slower flow rate (data not shown). The separation was optimized by changing both parameters and the resulting solvent programme was used in the lar ge scale separation of the pooled peak (Fig. 16). A total of 20 fractions were collected from the ion-exchange column and assayed for antigenicity by ELISA using pooled HSSS from Kenyan patients (Fig. 16). Results showed that peaks which eluted at 85% of 1.0 M NaCl (i.e. fractions #11-14) contained the antigen of interest (Fig. 16). Silver stained SDS-PAGE gels revealed that L . antigens have molecular weights ranging between 28 - 92 KD (Fig. 17).

A reverse-phase C3 HPLC column was used to further separate pooled fractions A - C from the ion-exchange column. A stepwise gradient of increasing acetonitrile was used to elute proteins. The antigen of interest (indicated by the indirect ELISA result) was eluted between 82% - 90% acetonitrile, peak D (Fig. 18). Silver-stained SDS-PAGE of peak D from a reverse-phase column showed smears throughout tb<* loaded lane (data not shown). This suggests the sample was either degraded or it was variably glycosylated. Nevertheless, no reliable information on the range of molecular weights or the purity of peak D was obtained.

Immunoblotting of trypanosome and Leishntania lysates using pooled HSSS from Kenyan patients or patients from Daloa, Cote d'Ivoire, gave similar results (Figs. 19 and 20). Most of the immunoreactive trypanosomal components had apparent molecular 1 6 7 weights > 43 KD in a 10% SDS-PAGE gel under reducing conditions. Results obtained with Daloan patients' sera (Fig. 19) differed from those obtained with Kenyan patients (Fig. 20) in that some cross reactivity with the Leishmania lysates (Lane F) was observed in components above 68KD and in a band at 31 KD. In addition, three distinct bands with molecular weights between 25 - 43 KD in irypanosome lysates were recognized by the HSSS from Daloan patients (Fig. 19) but not with sera from Kenyan patients (Fig. 20). No reaction was observed with uninfected human sera in immunoblotting with parasite lysates (data not shown).

SDS-PAGE and silver staining were used to identify the PCF trypanosomal antigens isolated by the immunoaffinity column made with pooled HSSS from Kenyan patients. The eluted fractious from the affinity column were assayed in indirect ELISA using biotinylated Ig from HSSS. The immunoreactive bands are shown in Fig. 21. Multiple bands were observed with v.iolecular weights above 45 KD and distinct bands of molecular weights of approximately I v KD, 31 KD and 24 KD (Fig. 21).

Derivation and characteristics of monoclonal antibodies

Gel fractions o IT. b. rhodesiense ViTat 1.1 PCF (corresponding to molecular weights of >68 KD, 43-68 KD end 25-43 KD) were excised from gels using prestained high molecular weight standards as references. Silver stained gels were used to examine the eluted gel fractions prior to immunization (Fig. 22). Gel fractions #s 1, 2 and 3 were found to correspond to molecular weights of >48 KD, 48-37 KD and 22-37 KD, respectively (Fig. 22). A total of 26 MAbs were produced againstr . b. rhodesiense ViTat 1.1 PCF gel fraction # 1, of which 10 were established by cloning using the methyicellulose method, the remainder being cloned using standard limited dilution methodology. Ten MAbs were raised against gel fraction # 2.

Indirect ELISA v/as employed to determine the binding specificities of these MAbs to water and detergent lysates of Trypanosoma and Leishmania species. Only 14 MAbs showed binding to Trypanosoma lysates, six of which were made against gel fraction # 1 and the remaining eight against gel fraction # 2 (data not shown). The 8 MAbs made against gel fraction # 2 bound to a (glutamic acid-proline)8 peptide repeat in indirect ELISA 1 6 8 and recognize the previously identified "Procyclin" molecule of T. brucei spp. Immunoblotting of these 8 MAbs to both Trypanosoma and Leishmania lysates also revealed a T. brucei species-specific broad band of 30-40 KD that had the same shape and position as "Procyclin" (data not shown). The binding of the remaining MAbs (6) is summarized in Table 19. All six MAbs bound only to the Trypanosoma water or detergent lysates and not to Leishmania water or detergent lysates (Table 19). Five of the anti-PCF MAbs (# 65,91,148,236 and 401) were T. brucei species-specific while MAb # 20 bound to procyclic lysates of both T. bri rei spp. and T. Congolense (Table 19). All six MAbs bound to bloodstream lysates of T. b. rhodeslsnse and were therefore chosen for the development of antigen-trapping sandwich ELISA for serodiagncsis of human sleeping sickness. The isotypes of these MAbs are shown in Table 19. Three IdAbs (# 20,148 and 236) had IgG isotypes while the remaining antibodies (# 65,91 and 401) had IgM isotypes (Table 19).

Immunofluorescence was used to determine the surface disposition of TBRP1 antigens recognized by the six MAbs. While the positive control anti-procyclic surface MAb (# 137) showed positive binding to live and all fixed trypanosomes, no binding was observed with all 6 test MAbs with either live or glutaraldehyde-fixed trypanosomes or Leishmania (data not shown). Very weak binding was observed with MAb # 20 in formaldehyde-fixed T. congolense PCF (Fig. 23C) but not with formaldehyde-fixed bloodstream forms or procyclics of T. brurei spp.(Fig. 23B) or of L. donovani IS2D promastigot s (Fig. 23A). Fluorescence binding was observed with MAb # 20 in acetone- permeabilized PCF of T. brucei spp.(Fig. 23D), T. congolense (Fig. 23E), bloodstream forms of T. b. rhodesiense (Fig. 23F) but not with L. donovani IS2D (data not shown). Both MAbs # 236 and 65 showed positive binding to acetone-permeabilized PCF of T. brucei spp. and very weak reactivities to acetone-permeabilized T. b. rhodesiense bloodstream forms and L. donovani promastigotes ( Fig. 24 and Fig. 25). While the binding of MAb # 236 was localized at the anterior region of trypanosomes (Fig. 24), MAb # 65 showed a weak, uniform fluorescence pattern on formaldehyde-fixed and acetone- permeabilized trypanosomes (Fig. 25). No fluorescence was observed with MAb #91, 148 and 401 with formaldehyde-fixed or acetone permeabilized parasites (Fig. 26). 1 6 9 Characterization of TBRP1 antigens recognized by selected MAbs

Immunoblotting was used to identify the trypanosome antigens recognized by the six selected MAbs. MAb # 20 bound to a Trypanosoma specific antigen with a molecular weight of 76 KD for PCF of T. brucei species. A band corresponding to a molecular weight of 69 KD was detected in the lysates of bloodstream forms of T. b. rhodesiense and in the PCF of T. congolense (Fig. 27). MAb #148 recognized two T. brucei species- specific bands (65 KD and 75 KD) in the reducing SDS-PAGE gels of T. b. rhodesiense bloodstream and procyclic stage lysates, while two closely overlapping bands (~75 KD) were observed in gels of the PCF of T. b. gambiense and one band (75 KD) for T. b. brucei (Fig. 28). A weak binding to three high molecular weight bands (123 KD, 228 KD and 800 KD) for the T. brucei spp. was observed with MAb # 236 (Fig. 29). MAbs # 65, 91 and 401 showed similar immunoblotting patterns for T. brucei spp. (2 bands at 110 KD region). MAb # 65 recognized bands at 102. KD and 1^9 KD (Fig. 30), while MAb # 91 bound to 2 bands at 101 KD and 110 KD (Fig. 31). MAb # 401 bound to bands at 116 KD and 128 KD (Fig. 32).

Exposure of PCF trypanosome lysates to heat, proteases and chemicals was used to determine if the six selected MAbs recognized protein or carbohydrate epitopes. Heat treatment of trypanosome lysates showed very little effect on the binding of MAbs # 20, 65,91 and 148 in ELISA. However, the binding of MAbs # 236 and 401 to heat treated trypanosome lysates was significandy reduced (Fig. 33). No binding was observed for any of the six MAbs to trypanosome lysates treated with Proteinase K while some binding was observed with MAbs # 20,65 and 148 to pronase treated parasite lysates (Fig. 33). TFMS treatment of T. b. rhodesiense lysates showed very litde effect on antibody binding, with the exception of MAb # 236 whose binding to the TFMS treated trypanosome lysate was 37% reduced. Immunoblotting of the treated trypanosome lysates gave the same results as those obtained with indirect ELISA (data not shown).

A Con-A column was used to concentrate glycoproteins containing alpha-D- mannose and alpha-D-glucose in T. b. rhodesiense lysates. Trypanosomal antigens recognized by MAb # 20 showed no binding to Con-A column (Fig. 34). The antigen recognized by MAb # 20 was found only in the Con-A breakthrough fractions but not in 1 7 0 the alpha-D-mannoside eluted fractions (Fig. 34). Results similar to those observed with MAb # 20 were also obtained with MAbs #91,148,236 and 401 (data not shown). However, MAb # 65 was exceptional in that it showed positive binding to the Con-A eluted fractions (Fig. 35).

Differences in the solubility profiles of antigens recognized by the six selected anti- PCF MAbs (# 20,65,92,148,236 and 401) were observed in both the bloodstream and procyclic stages of trypanosomes (Fig. 36). In genera!, detergent extracted parasite lysates yielded higher O. D. readings than butanol extracted lysates. Trypanosomal antigens recognized by these MAbs were distributed mostly in the aqueous phase of Triton X-l 14 and not the detergent phase, except MAb # 65 which showed nearly equal distribution between the two phases (Fig. 36D). Of the 10 different lysate treatments employed only water and Triton X-100 treatments yielded quantitative antigen differences between procyclic and bloodstream trypanosomes (Fig. 36).

Differential centrifugation of the bloodstream and procyclic T. b. rhodesiense lysates at forces of 1000,10000 and 100000 g was performed. The resulting pellets arsd supernatants were assayed by indirect ELISA(Table 20). Most antigens identified by the selected MAbs were found in the supernatant. Tb* distribution profiles of these antigens were similar in both bloodstream and procyclic lysates with the exception of MAb # 65 and the positive control (anti-'procyclin' MAb # 137). MAb # 65 showed higher O. D. reading in the >10000x g trypanosome bloodstream pellets than in the supernatants. Lower O. D. readings were found with MAb # 65 in the corresponding procyclic pellets than those observed with the supernatants. The opposite trends were observed with the positive control (anti-'procyclin’ MAb # 137) in which higher binding was observed with the centrifuged PCF pellets and the bloodstream supernatants (Table 20).

Purification of TBRP1 antigens by immunoaffinity chromatography

Immunosorbents prepared by coupling v.’ith either MAb # 20 or MAb #148 were used to purify antigens recognized by these MAbs. The immunoreactivities of fractions eluted from the MAb # 20 affinity column are shown in Fig. 37. Fractions 4-7 werfe pooledand 171 silver stained SDS-PAGE gels were used to confirm the identity of the dialyzed, concentrated pooled fractions (Fig. 38). Seven distinct silver-stained bands could be seen in the SDS-PAGE reducing gel (Fig. 38). Immunoblotting of these isolated fractions with MAb # 20 identified the antigen containing band with the molecular weight at 76 KD as previously described (Fig. 27). Amino acid analysis was performed on a protein band excised from the corresponding position on the Immobilon and its amino acid composition is shown in Table 21. Gas-phase microsequencing of the immunoaffinity purified antigen revealed an 11 amino acid sequence with an N-terminal proline (Table 22).

Trypanosomal antigens recognized by MAb # 148 were eluded from the immunoaffinity column (Fig. 39). The positive peak fractions (fractions # 4 and 5) were pooled and their identities were shown by silver staining a SDS-PAGE gel (Fig, 40). Some darker staining bands were observed at a molecular weight of 65 KD and additional minor bands (12) were distributed between molecular weights of 31 KD - 65 KD (Fig. 40). Immunoblotting identified the darker staining bands (2) with the same shape and molecular distribution as shown previously (Fig. 28), to be the antigens recognized by MAb #148 (data not shown). Amino acid composition of these two antigens showed similar and yet distinct compositions (Table 24). The 75 KD band contained a higher proportion of Glu/Gln and Ser whereas the 65 KD band contained higher proportions of Glu/Gln, Set and Gly (Table 23). A sequence of 20 amino acids was obtained for the 65 KD protein (Table 24). Uncertainties about several of the amino acids suggests that some impurities were present (Table 24).

Optimization of sandwich ELISA

All six selected MAbs (# 20,65,91,148,236 and 401) inhibited only their homologous radiolabelled MAbs in a competitive radioimmunoassay (Table 25). This suggests that distinct epitopes are recognized by the individual MAbs. An additive effect on the binding of the MAbs to trypanosome lysates was observed when different antibodies were added sequentially to ELISA wells (Fig. 41).

Binding of ihese biotin-labelied MAbs to trypanosomal PCF antigens trapped by homologous unlabelled MAbs or by heterologous MAbs was examined using a sandwich 1 7 2 ELISA. As shown in Table 26, positive binding was observed only with the homologous antibodies. No binding was observed with heterologous antibodies and the anti-urokinase MAb (negative control).

The double antibody sandwich ELISA technique utilitizing the mixture of six biotinylated MAbs and the unlabelled MAb mixture was found to be specific for trypanosomal antigens. Fig. 42 shows a linear positive response using a sandwich ELISA with the addition of known concentrations of trypanosomal PCF antigens but not with Leishmania antigens. Some binding to the T. congolense lysates was also observed with the trapping ELISA (Fig. 42). The minimum detection for the double antibody sandwich ELISA was 1- 3 ng trypanosomal proteins/well (i.e. 10-30 ng/ml of the equivalent of extract from 3000 trypanosomes/ml).

Antigen detection in sera or urine from trypanosome-inltcted mice

Sera and urine collected from T. b. rkodesiense-infected mice were tested using the double antibody sandwich ELISA. The same antibody combination used in the artificial system was re-employed to test the mouse sera and urine and these data are presented in Fig. 43. Each data point represents the mean value obtained with each serum dilution. Trypanosome antigens were first detected in mouse sera as early as 3 days post-infection and thereafter remained at low but detectable levels. An increase in the detectable antigen level was observed on day 17 post-infection and after this time the detectable antigen level was maintained (Fig. 43A). Microscopic detection of parasites in the blood was first achieved at 10 days post-infection. Parasites remained detectable thereafter.

Trypanosomal antigens were detected in mouse urine at 10 days post-infection using the MAb mixture in a sandwich ELISA (Fig. 43B). Mean antigen levels then remained elevated for several days then decreased at 17 days post-infection. An increase in detectable antigen level was observed in urine at 20 days post-infection.

Rabbit antibodies made against whole lysate of T. b. rhodesiense PCF were used in a sandwich ELISA (Chapter 3) for comparing the relative sensitivities of these assay systems. Fig. 44 shows the ELISA results for the mouse sera and urine using the rabbit 1 7 3 anti-TBRPl whole lysate. Parasite antigens were clearly detected at 5 days post-infection and a dramatically increased antigen level was observed by 10 days post-infection. A cyclic fluctuation in the detectable antigen level in mouse sera was observed (Fig. 44 A). Trypanosomal antigens in mouse urine were detected earlier using the MAbs (10 days post­ infection) than with the rabbit polyclonal antibodies (22 days post-infection). Parasite antigens remained at detectable levels thereafter using both antibody assays.

Sequential nine samples were collected from an additional 5 trypanosome-infected mice without sacrificing these individuals. Similar parasite antigen profiles to those described above were found using either the MAb mixture or rabbit polyclonal antibodies (data not shown). Trypanosomal antigens were not detected in any of the sera from uninfected mice using either of the assays. 174 Fig. 15

Separation of T. b. rhodesiense ViTat 1.1 PCF crude lysate by gel permeation chromatography using a Sephacryl S-200 column. The absorbance of collected fractions was determined at the wavelengths of 28Qnm (solid lines) and 230nm (dotted lines) as shown in the top diagram (A). Immunoreactivities of these collected fractions were determined by indirect ELISA and results are shown in the bottom diagram (B). 1 7 5

1.6

u , 12- o z < m o 0.8

0 . 4

120 2 4 0 3 6 0 4 8 0 VOLUME ml )

2 . 4 E

O 1-6-

° 0.8- O

2 4 0 6 0 120 1 8 0 FRACTI0 N N U F R 176 Fig. 16

Ion-exchange-HPLC profiles of pooled fractions # 40-50 from GPC. (A) Absorbance profile of eluate; (B) ELISA profiles of the collected fractions. 177

-100

E c o 00

6030 90 Time ( min.)

■ 1.0 n m 05 4 D,0.

-0.5

30 60 Time ( * t. ) 178 Fig. 17 reparation of fractions # A-D from the ion-exchange HPLC column by a reducing 5-15 % gradient SDS-PAGE gel. Lane 1, fraction A from the ion-exchange column; Lane 2, fraction B; Lane 3, fraction C; Lane 4, fraction D. 179 Fig. 18

Reverse phase HPLC profiles of pooled fractions (A, B and C) from the ion-exchange HPLC column. (A) Absorbance profile of eluate; (B) ELISA profile of the collected fractions. Absorbance 280nm 0 — 1 ■ i - 1 "■ f— B 0 3 0 6 0 3 ie (min.) Time ie (min.) Time 0 9 0 6 0 9 120 120 1-50 -10 0 -I.0 0.5

b o 3 O 3

Antigen profiles detected by pooled HSSS from Daloan patients (Cote d'Ivoire) in various parasite lysates using immunoblotting. Lane A, T. congolense 45/1 PCF; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. rhodesiense WraTat 3 PCF; Lane D, T. b. gambiense U2; Lane E, T. b. brucei 1026; Lane F, L. donovani IS2D promastigotes; Lane G, T. b. rhodesiense ViTat 1.1 bloodstream forms.

A 8 C D E F G 1 8 2 Fig. 20

Antigen profiles detected by pooled HSSS from Kenyan patients in various parasite lysates using immunoblotting. Lane A, T. congolense 45/1 PCF; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. rhodesiense WraTat 3 PCF; Lane D, T. b. gambiense U2; Lane E, T. b. brucei 1026; Lane F, L. donovani IS2D promastigotes; Lane G, T. b. rhodesiense ViTat 1.1 bloodstream forms.

A B C D E F G

- 2 0 0 KDa - 9 7 -6 8

-2 5

-1 8 -14 1 8 3 Fig. 21

SDS-PAGE silver stain profiles of T. b. rhodesiense ViTat 1.1 PCF eluted from an immunoaffinity column coupled to pooled immunoglobulins from HSSS from Kenyan patients. Lane A, T. b. rhodesiense ViTat 1.1 PCF whole lysate; Lane B - E, fractions of ELISA positive peak eluted from the HSSS affinity column. Fraction C was the most positive fraction as determined by indirect ELISA.; Lane F, buffer used for eluting fractions from the affinity column; Lane G, sample buffer for SDS-PAGE gel.

G F E D C B A 184 Fig. 22

Silver stained SDS-PAGE gel patterns of extracted gel fractions used for immunization of BALB/c mice. Lane A, fraction # 1; Lane B, fraction # 2; Lane C, fraction # 3; Lane D,T. b. rhodesiense ViTat 1.1 PCF whole lysate.

KDa A B C 200- 97- 68- I 43- f

25-

1 8 - 14- 1 8 5 Fig. 23

Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmania species using anti-PCF MAb # 20. (A) formaldehyde-fixed L. donovani IS2D; (B) formaldehyde-fixed T. b. rhodesiense ViTat 1.1 PCF; (C) T. congolense 45/1 PCF; (D) acetone-permeabilized T. b. rhodesiense ViTat 1.1 PCF; (E) acetone-permeabilized T. congolense 45/1 PCF; (F) acetone-permeabilized T. b. rhodesiense ViTat 1.1 bloodstream for ms; (G) phase-contrast micrograph of bloodstream trypanosomes in panel F. 1 8 6 1 8 7 Fig. 24

Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmania species using anti-PCF MAb # 236. (A) formaldehyde- fixed L. donovani IS2D; (B) formaldehyde-fixed T. b. rhodesiense ViTat 1.1 PCF; (C) acetone-permeabilized L. donovani IS2D; (D) phase contrast view of L. donovani IS2D in panel C; (E) acetone-permeabilized T. b. rhodesiense ViTat 1.1 PCF; (F) phase-contrast micrograph of PCF trypanosomes in panel E; (G) acetone-permeabilized T. b. rhodesiense ViTat 1.1 bloodstream forms.

1 8 9 Fig. 25

Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmania species using anti-PCF MAb # 65. (A) formaldehyde-fixed L. donovani IS2D; (B) phase-contrast view of L. donovani IS2D in panel A; (C) formaldehyde-fixed T. b. rhodesiense ViTat 1.1 PCF; (D) phase-contrast view of PCF trypanosomes in panel C; (E) acetone-permeabilized L. donovani IS2D; (F) phase-contrast view of L. donovani IS2D in panel E; (G) acetone-permeabilized T. b. rhodesiense ViTat 1.1 PCF; (H) phase-contrast view of PCF trypanosomes in panel G; (I) acetone- permeabilized T. b. rhodesiense ViTat 1.1 bloodstream forms. 1 9 0 191 Fig. 26

Immunofluorescence patterns of formaldehyde-fixed or acetone-permeabilized Trypanosoma and Leishmcnia species using anti-PCF MAbs # 401 (A), 148 (B) and 91 (C). (1) formaldehyde-fixed L. donovani IS2D; (2) formaldehyde-fixed T. b. rhodesiense ViTat 1.1 PCF; (3) acetone-permeabilized L. donovani IS2D; (4) acetone-permeabilized'/1, b. rhodesiense ViTat 1.1 PCF; (5) acetone-permeabilized T. b. rhodesiense ViTat 1.1 bloodstream forms. 1 9 2

<

CO 1 9 3 Fig. 27

The Trypanosoma specific antigen profile recognized by MAb # 20 by immunoblotting. Lane A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b rhodesiense ViTat 1.1 PCF; Lane C, T. b. gambiense U2 PCF; Lane D, T. b. brucei 427.01 PCF; lane E, T. congolense 45/1 PCF; Lane F; L. major A2 promastigotes; Lane G, L. donovani IS2D promastigotes.

-14 1 9 4 Fig. 28

Detection of antigen in whole lysates of Trypanosoma and Leishmania species by immunoblotting using MAb # 148. Lan<* A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. gambiense U2 PCF; Lane D, T. b. brucei 427.01 PCF; Lane E, T. congolense 45/1 PCF; Lane F; L. major A2 promastigotes; Lane G, L. donovani IS2D promastigotes.

G F E D C B A '2 0 0 KOa -9 7 -68 4 3

-2 5

-1 8

-14 Detection of antigen in whole lysates of Trypanosoma and Leishrmnia species by immunoblotting using MAb # 236. Lane A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. gambiense U2 PCF; Lane D, T. b. brucei 427.01 PCF; Lane E, T. congolense 45/1 PCF; Lane F; L. major A2 promastigotes; Lane G, L. donovani IS2D promastigotes. 1 9 6 Fig. 30

Detection of antigen in whole lysates of Trypanosoma and Leishmama species by immunoblotting using MAb # 65. Lane A, T. b, rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. gambiense U2 PCF; Lane D, T. b. brucei 427.01 PCF; Lane E, T. congolense 45/1 PCF; Lane F; L. major A2 promastigotes; Lane G, L. donovani IS2D promastigotes.

G F E D C B A - 2 0 0 KOa -9 7 -68 -4 3 -2 5

-18

-14 1 9 7 Fig. 31

Detection of antigen in whole lysates of Trypanosoma and Leishmanic species by immunoblotting using MAb # 91. Lane A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. gambiense U2 PCF; Lane D, T. b. brucei 427.01 PCF; Lane E, T. congolense 45/1 PCF; Lane F; L. major A2 promastigotes; Lane G, L. donovani IS2D promastigotes.

G F E D C B A -200 KDa

-68 -43 -25 -18

-14 1 9 8 Fig. 32

Detection of antigen in whole lysates of Trypanosoma and Leishmania species by immunoblotting using MAb # 401. Lane A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. gambiense U2 PCF; Lane D, T. b. brucei 427.01 PCF; Lane E, T. congolense 45/1 PCF; Lane F; L. major A2 promastigotes; Lane G, L. donovani IS2D promastigotes.

D C B

-2 0 0 KDa

-43

-25

-18

-14 1 9 9 Fig. 33

Analysis of antigens recognized by the six selected anti-PCF monoclonal antibodies after heat, protease and chemical treatment. Lysates of T. b. rhodesiense ViTat 1.1 PCI7 receiving no treatment (open bars), heat treatment (closed bars with white hedged lines), Proteinase K treatment (open bars with horizontal lines), pronase treatment (open bars with dark hedged lines) and TFMS treatment (closed bars). (A) MAb # 20; (B) MAb # 148; (C) MAb#236; (D)MAb#65; (E)MAb#91; (F) MAb#401; (G) anti-"procyclin>’MAb# 137; (H) anti-urokinase MAb U17.23.26. Absorbance (405 nM) 1.5 ABCDEFGH

oolnl antibodies Monoclonal 200 2 0 1 Fig. 34

Enzyme-linked immunosorbent assay of T. b. rhodesiense ViTat 1.1 PCF after passing over Con-A agarose using MAb # 20. Results for the breakthrough and eluted fractions were placed in the same graph for ease of direct comparison. Con-A breakthrough fractions (open circles); eluted fractions (closed circles). po N> o

0. 8 " 9\ _ l _

L_ ii. o

Absorbance ( 405 nM ) K> Ul Ul - (A

Fraction Number 2 0 3 Fig. 35

Enzyme-linked immunosorbent assay of T. b. rhodesiense ViTat 1.1 PCF after passing over Con-A agarose using MAb # 65. Con-A breakthrough fractions (open triangles); eluted fractions (closed triangles). ro -p>- o -4

0.61

0.5 H i P U) L. Absorbance ( 405 nM ) N> N> __ (A

Fraction 2 0 5 Fig. 36

Extraction of trypanosomal antigens by various solubilization buffers. Trypanosome lysates: T. b. rhodesiense v iTat 1.1 PCF ( closed bars); T. b. rhodesiense ViTat 1.1 bloodstream forms (hedged bars); solubilization buffers (open bars). Solubilization buffers: (A) water, (B) 1% Chaps; (C) 0.1% SDS; (D) 0.5% NP-40; (E) 1% TX-114 aqueous phase; (F) 1% TX-114 detergent phase; (G) 1% TX-100; (H) 1% Butanol; (I) 2% Butanol; (J) 3% Butanol. TBRP1 monoclonal antibodies used in indirect ELISA: (A) MAb #20; (B)MAb# 148; (Q MAb#236; (D)MAb#65; (E) MAt #91; CE) MAb# 4C1. Absorbance (405 nM) Absorbance (405 nM) 0.0 0 0.41 0.6 0.8 . 2 - ABCDEFGHI J ABCDEFGHI ABCDEFGHI ouiiain Buffer Solubilization ouiiain Buffer Solubilization ouiiain Buffer Solubilization J J J 2 0 7

v>

8 § l i J3tm cn© A < h

Solubilization Buffer

B oV) ■ 't u B es •k. f i w© £ «<

ABCDEFGHI Solubilization Buffer

S s V) Tfe O© «B £k. ceO A <

ABCDEFGHI Solubilization Buffer 2 0 8 Fig. 37

Enzyme-linked immunosorbent assay on T. b. rhodesiense ViTat 1.1 PCF lysates: eluted fractions from the MAb # 20 immunoadsorbent. 2 0 9 Absorbance (405 nM)

Fraction number 210 Fig. 38

Silver stained SDS-PAGE profile of ELISA positive, affinity purified fractions from the MAb # 20 immunoadsorbent. Lanes A-D, represent various dilutions: 1/8 (A); 1/16 (B); 1/32 (C) and 1/64 (D) of the total pooled samples from the immunoadsorbent; Lane E, sample loading buffer. The arrow indicates the antigen identified by subsequent immunoblotting.

A B C D E 2 1 1 Fig. 39

Enzyme-linked immunosorbent assay on T. b. rhodesiense ViTat 1.1 PCF lysates: eluted fractions from the MAb # 148 immunoadsorbent. Absorbance (405 nM) 2 1 3 Fig. 40

Silver siained SDS-PAGE profile of the ELISA positive, affinity purified fraction.* from the MAb # 148 immunoadsorbent. Lane A-D, represents various dilutions: 1/10 (A); 1/20 (B); 1/40 (C) and 1/80 (D) of the total pooled samples from the immunoadsorbent; Lane E, sample loading buffer. Arrows indicate the antigens identified by subsequent immunoblotting.

A B C D E 2 1 4 Fig. 41

Binding of different biotinylated anti-trjpanosome monoclonal antibodies to trypanosome procyclic water lysates in indirect ELISA. The biodnylated monoclonal andbodies were assayed at a dilution of 1/2000 as described in Materials and Methods section. Absorbance (405 nM)

© o o o © © M ON 0 0

MAb#20

MAb #148

MAb#236

MAb#65

MAb#91

MAb#401

MAbs mix

PBS 5 1 2 216 Fig. 42

Detection of trypanosomal antigens in water lysates of parasites by double antibody sandwich ELISA. T. b. rhodesiense ViTat 1.1 PCF (open triangles); T. b. gambiense U2 PCF (open circles); T. b. brucei 427.01 PCF (open squares); T. congolense 45/1 PCF (closed triangles); L. donovani IS2D (closed circles); L. major A2 (closed squares); PBS (open circles; dotted lines). Concentration of parasites used in the ELISA: 10? (A); 5 x 10^ (B); 106 (C); 5 x 105 (D); 10$ (E); 5 x 10* (F); 104 (G); 103 (H); 5 x 102 (I); 102 (J); 10 (K) parasites per ml. Absorbance (405 nM) 0 1.4 0 1 0.4- 0 1 o.o . . . . . 2 0 8 6 2-1 - - - - B A aaie pr el (equivalents) well per Parasites E G I K J I H G F E D m 7 1 2 218 Fig. 43

ELISA measurement of trypanosomal antigens in sera and urine from T. b. rhodesiense - infected mice. A 1/2000 dilution of a mixture of six selected MAbs was used as capture antibody and a 1/2000 dilution of the biotinylated MAb mix was used as the detecting reagent. Each data point represents the mean value of the maximum absorbance ratios obtained with dilutions of mouse sera (open squares) and mouse urine (open triangles). Absorbance Ratios Absorbance Ratios 4 0 2 0 4 3 2 3 1 1 05 0 5 0 5 30 25 20 15 10 5 0 5 0 5 as Post-infection Days as Post-infection Days 1 1 2 2 30 25 20 15 10 5 A B 21S 2 2 0 Fig. 44

ELISA measurement of trypanosomal antigens in sera and urine from T. b. rhodesiense - infected mice using rabbit anti-TBRPl whole lysate antibodies. A 1/4000 dilution of rabbit antibodies was used as capture reagent and a 1/400 dilution of biotinylated rabbit antibodies was used as the detecting reagent. Each data point represents the mean value of the maximum absorbance ratios obtained with dilutions of mouse sera (open circles) and mouse urine (open triangles) found in infected mice. Absorbance Ratios Absorbance Ratios as Post-infection Days

CO o 221 2 2 2 Fig. 45

Immunoblottirg profiles of MAb # 148 on various T. brucei lysates separated on a reducing 10% SDS-PAGE gel. Lune A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. rhodesiense WraTat 3 PCF; Lane D, T. b. gambiense TH-1 PCF; Lane E, T. b. rhodesiense 1799 PCF; Lane F, T. b. gambiense TH-17 PCF; Lane G, T. b. gambiense U2 PCF; Lane H, T. b. brucei 10-26 PCF; Lane I, T. b. bruce '427.01 PCF; Lane J, T. b. brucei AmTat 1.1 PCF; Lane K, T. b. brucei MiTat 1.2 PCF; Lane L, L. ^ nwr promasttgotes; Lane M, L. donovani IS2D promastigotes.

MLKJ I H e F E D C B A

-3 1

-3 1

-14 2 2 3 Fig. 46

Lnmunoblotting profiles of MAb # 148 on various T. brucei species lysates separated in a non-reducing 10% SDS-PAGE gel. Lane A, T. b. rhodesiense ViTat 1.1 bloodstream forms; Lane B, T. b. rhodesiense ViTat 1.1 PCF; Lane C, T. b. rhodesiense WraTat 3 PCF; Lane D, T. b. gambiense TH-1 PCF; Lane E, T. b. rhodesiense 1799 PCF; Lane F, T. b. gambiense TH-17 PCF; Lane G, T. b. gambiense U2 PCF; Lane H, T. b. brucei 10- 26 PCF; Lane I, T. b. brucei 427.01 PCF; Lane J, T. b. brucei AmTat 1.1 PCF; Lane K, T. b. brucei MiTat 1.2 PCF; Lane L, L. major promastigotes; Lane M, L. donovani IS2D promastigotes.

KDa a r t . d e f GH l J K l M

14 *- 2 2 4 Fig. 47

Detection of Trypanosomal antigens in parasitologically-confirmea sleeping sickness sera from Kenyan patients using rabbit anti-trypanosome whole lysate antibodies (Closed bars) or MAb mixture (open bars with dots). Human sera from trypanosome-infected Kenyan patients (A-L); sera from healthy North Americans (NHS 1, NHS 2); T. b. rhodesiense ViTat 1.1 PCF lysate (TBRP1). 2 2 5

TBRP1

NHS 2

NHS 1

Absorbance ratios 2 2 6 Fig. 48

Detection of Trypanosomal antigens in uninfected human sera from Kenyans (-ve controls) using rabbit anti-trypanosome whole lysate antibodies (closed bars) or MAb mixture (open bars with dots). Sera from non-infected Kenyans (A-L); sera from healthy North Americans (NHS 1, NHS 2); T. b. rhodesiense ViTat 1.1 PCF lysate (TBRP1). 2 2 7

TBRP1

NHS 2

NHS 1

L

K

J

I e cs H E a G X F

E

D

C

B

A

0 10 15

Absorbance ratios Table 19

Binding characteristics of selected monoclonal antibodies to detergent lysatesa of Trypanosoma species and Leishmania species

Antibody6 Isotypec T. b. T. b. T. b. T. b. T. L. L. major human rhodesiense rhodesiense gambiense brucei congolense donovani A2 transferrind ViTat 1.1 ViTat 1.1 U2PCF 427.01 45/1 PCF IS2D bloodstream PCF PCF forms

20 IgG2b + + + + + 65 IgM + + + + - - -- 91 IgM + + + + -- -- 148 IgGl + + + + - - - - 236 IgGl + + + + - --- 401 IgM + + + + - “ anti- gonococus IgM “ - - MAb anti- ' urokinase IgGl ------MAb

to coat ELISA wells. 6 Derived against extracted gel fraction # 1 of T. b. rhodesiense ViTat 1.1 PCF. c Determined by antigen-capture ELISA. d Human transferrin was used as negative control antigen and was assayed at 5 pg/well in indirect ELISA.

ro ro GO )

Table 20

Detection of trypanosomal antigens by anti-PCF monoclonal antibodies in trypanosome supernatants and pellets3

Antibody1* Trypanosome Trypanosome Trypanosome procyclic Trypanosome procyclic bloodstream foim bloodstream form culture form culture form supernatants0 pelletsc supernatants0 pellets0

1 K 10 K 100 K 1 K 10 K 100 K 1 K 10 K 100K 1 K 10 K 100 K

20 4+ 4+ 4+ 2+ 2+ 2+ 3+ 4+ 4+ + 3+ 2+ 65 2+ + + 2+ 3+ 3+ 2+ 2+ 2+ 2+ 2+ + 91 3+ 4+ 4+ 2+ 2+ -r 3+ 3+ 3+ + + + ' 148 2+ 3+ 4+ 3+ 2+ 2+ 4+ 4+ 4+ 2+ 2+ 2+ 236 2+ 2+ 3+ + 4- + 2+ 2+ 2+ + + + 401 3+ 4+ 4+ + 2+ + 3+ 2+ 2+ + + + Anti- procyclin + 2+ 2+ + - - 4+ 3+ 2+ 3+ 4+ 4+ MAb 137 Anti­ urokinase ““ * “ “ — MAb

3 Sonicated T. b. rhodesiense ViTat 1.1 bloodstream or procyclic lysates (lOtyml 50 mM Tris-saline, 0.5 % NP-40) were centrifuged at xlOOOg (IK), xlOOOOg (10K) and xlOOOOOg (100K) for 60 miw at 49C. ti Antibodies were derived against extracted gel fraction # 1 of T. b. rhodesiense ViTat 1.1 PCF. 0 Supernatants and pellets were assayed in indirect ELISA. d + O.D. reading for the test samples were 2-3 X above those of the negative control (anti-urokinase MAb). 2+ O.D. reading for the test samples were 3-5 X above those of the negative control (anti-urokinase MAb). 3+ O.D. reading for the test samples were5-7 X above those of the negative control (anti-urokinase MAb). 4+ O.D. reading for the test samples were >8 X above those of the negative control (anti-urokinase MAb). 9 2 2 2 3 0 Table 21

The amino acid composition8 of immunoaffinity purified trypanosomal antigen from MAb # 20 immunoadsorbent. f Amino acid Number of residues*1

Ala 56 Arg 49 Asn + Asp 39 Cys - Gin + Glu 68 Gly 72 His 14 lie 31 Leu 71 Lys 55 Met 11 Phe 20 Pro 39 Ser 68 Thr 42 Trp - Tyr 26 Val 39

Total residues 700

Molecular weight (KD) 76

8 24 hr. acid hydrolysis was used. b Values are presented as number of residues found in the sequence of protein. One residue of amino acid = 76 pmoles; using Alanine as reference. 2 3 1 Table 22

Amino terminal amino acid sequence8 of T. b. rhodesiense ViTat 1.1 PCF antigen recognized by monoclonal antibody # 20

Cylcle number PTHAA Pmol

1 Pro 5.59 2 Ala 4.07 3 Lys 2.36 4 Asp 2.58 5 De 2.05 6 ? 7 Phe 2.78 8 Gly 2.56 9 Thr 1.84 10 Glu 1.74 11 Ala 1.82 a Determined by gas-phase microsequencing. 2 3 2 Table 23

The amino acid composition3 of immunoaffniity purified trypanosomal antigens from MAb # 148 immunoadsorbent

Amino acid Number of residues** Number of residues**

Ala 36 41 Arg 48 58 Asn + Asp 50 56 Cys - - Gin + Glu 71 88 Gly 66 60 His 14 14 lie 19 21 Leu 49 53 Lys 47 53 Met 6 11 Phe 15 16 Pro 30 33 Ser 66 74 Thr 36 44 Trp - - Tyr 21 24 Val 27 26

Total residues 601 675

Molecular weight (KD) 65 75 a 24 hr. acid hydrolysis was used. ** Values are presented as number of residues found in the sequence of protein. One residue of amino acid =128 pmoles; using Alanine as reference. 2 3 3 Table 24

Amino terminal amino acid sequence3 of T. b. rhodesiense ViTat 1.1 PCF antigen recognized by monoclonal antibody # 148

Cycle number PTHAA Pmol PTHAA** Pmol**

1 Val 6.58 2 Leu 2.70 3 Ser 1.73 4 Pro 1.79 thr 1.07 5 Ala 2.95 6 Asp 0.91 7 Glu 1.17 Lys 0.80 8 Lys 1.21 Thr 0.58 9 Asn 0.68 Ser (Val) 0.31 (0.27) 10 lie 0.65 Leu 0.59 11 Ala 0.48 12 Ser 0.75 13 Thr 0.23 14 Leu 0.39 15 Asp 0.31 16 Lys 0.24 Gly (Val) 0.20 (0.15) 17 lie 0.30 18 Val 0.35 19 Asn 0.29 Gly 0.28 20 Val 1.12 Leu 0.19

a Determii A by gas-phase microsequencing. ** Amino acids present in minor quantities detected by microsequencing. Table 25

Competitive solid-phase radioimmunometric assay of anti-trypanosome monoclonal antibodies8

Unlabelled antibodies13 125i-iabelled 20 65 91 148 236 401 Negative Anti-TBRPl MAbs** control0 rabbit serum**

20 4+ ------3+ 65 - 4+ - -- -- 2+ 91 -- 4+ - - - - 2+ 148 -- - 4+ --- 3+ 236 --- 4+ -- 4+ 401 * "• 44- 4+

8 The table shows the inhibition by unlabelled monoclonal antibodies of binding of radiolabelled MAbs to lysates of procyclic culture forms of T. b. rhodesiense ViTat 1 .1 .4+, over 80% inhibition; 3+, 50-60% inhibition; 2-f, 30-50% inhibition; 1+, 10- 30% inhibition; -, no inhibition, b Monoclonal antibodies made against extracted gel fraction # 1 of T. b. rhodesiense ViTat 1.1 PCF. c Anti-urokinase monoclonal antibody 9.22.8. ** Made against whole sonicated lysate of T. b. rhodesiense ViTat 1.1 PCF. 4 3 2 Table 26 Binding of biotin-labeled MAbs to trypanosomal antigens trapped by homologous or heterologous MAbs in double antibody sandwich ELISA

Detector antibodiesa Trapping 20 65 91 148 236 401 Rabbit serum*5 antibodiesa

16 3+ - -- - - 3+ 65 - 3+ ---- 2+ 91 -- 3+ - - - 2+ 148 - -- 3+ - - 2+ 236 - •- - 3+ - 3+ 401 - - -- - 3+ 3+

Rabbit serum 2v 2+ 2+ 2+ 2+ 3+ 2+

Negative ------control0 Buffer — * a Monoclonal antibodies made against extracted gel fraction # 1 of T. b. rhodesiense ViTat 1.1 PCF. b Made against the whole lysate of T. b. rhodesiense ViTat 1.1 PCF. c Anti-urokinase monoclonal antibody 9.22.8. 3+ Absorbance > 0.7 2+ Absorbance > 0.4 1+ Absorbance >0.1 - Absorbance < 0.05 5 3 2 2 3 6 Discussion

Antibodies produced by trypanosome-infected mammalian hosts have been used in a few studies to estimate the numbers of different trypanosomal antigens recognized by the host during infection (Gray, 1961; Le Ray, 1975; Poupin et al., 1976; Shapiro and Murray, 1982; Gardiner et al., 1983; Taylor and Smith, 1983). However, the detailed nature of these antigens is still largely unknown. In the current study, three methods were employed to identify trypanosomal antigens that might have diagnostic potential. These included HPLC separation, immunoblotting and immunoaffinity purification. Pooled sera from trypanosome-infected patients were used as a probe for identifying trypanosomal antigens with diagnostic potential. The use of HSSS was based on the assumption that during the course of the infection infected hosts will produce antibodies to those immunogenic trypanosomal components that are circulating in the host.

Crude lysates of T. b. rhodesiense ViTat 1.1 PCF were initially separated by standard chromatographic techniques based on differences in molecular size, charge and hydrophobicity. Because antigenicity was used to identify materials of interest, the use of harsh conditions was minimized in these techniques. The sequence of purification steps was chosen on the basis of sample handing capacity, the amount of sample recovered and the speed of separation.

Because large amounts of materials have to be handled at the beginning of a separation, methods with higher capacity such as gel permeation and ion exchange chromatography are most appropriate (Pharmacia, 1980). At the later stages of separation, as the availability of sample is limited and the activity of sample becomes less stable, techniques with higher sample recovery and speed such as RP-HPLC are more useful (Pharmacia, 1980).

The complex mixture of trypanosome crude lysate was fractionated into five major peaks based on relative molecular sizes using a Sephacryl S-200 superfine column (Fig. 15). ELISA testing of the GPC fractions using pooled HSSS from Kenya as a probe indicated that antigens were distributed mainly in the high molecular weight peak (Fig. 15). Fractionation of this peak using a DEAE column was achieved by increasing the ionic strength of the mobile phase (Fig. 16). Because sample components were eluted from the ion-exchange column in the order of increasing net charge (Kopaciewicz and Regnier, 237 1983), the elution positions of the ELISA positive fractions at the late stages of the DEAE separation (85% 1.0 M NaCl) (Fig. 16) suggested that antigens in these mixtures were negativly charged at pH 8.0. Silver stained SDS-PAGE gels revealed antigens with molecular weights ranging between 31 KD - 92 KD (Fig. 17). The distribution of most immunoieactive trypanosomal components at the higher molecular weights was also determined using immunoblotting (Figs. 19 and 20) and immunoaffinity purification (Fig. 21) techniques. Together these results identified at least 20 different trypanosomal components as compared to the previously reported 7-8 antigens recognized by trypanosome infected mammals (Shapiro and Murray, 1982; Gardiner et al., 1983; Burgess and Jerrells, 1985). Differences in the identification methods and the sources and numbers of sera that were used may have contributed to these different results.

Although both HPLC and affinity chromatography allow the isolation of trypanosomal antigens, these methods are cumbersome and difficult to perform in large enough scale to isolate sufficient materials for immunization and hybridoma derivation. Since the majority of these antigens were found to be in the higher molecular weight range (>62 kD), gel fractions containing defined molecular sizes of macromolecules were thought to be an easier alternative for providing sufficient materials for the immunization of mice. This protocol was designed to focus on diagnostically promising antigens and to circumvent the possibility of making antibodies to only a few immunodominant molecules of procyclic trypanosomes. Such a result was encountered in a previous study in which all ten MAbs derived against live PCF parasites bound to a single immunodominant molecule - "Procyclin" (Richardson et al., 1986), and the present study in which all 8 MAbs made against gel fraction # 2 that contained the "Procyclin" also recognized the glu-pro dipeptide repeats of "Procyclin". Additionally, immunization with different gel fractions allows greater insights into the possible pathological role of some of the trypanosome antigens. For example, all six mice injected with gel fraction # 3 (Fig. 22) died while mice injected with whole lysate or other gel fractions survived. These results suggest gel fraction # 3 contained toxic molecule(s). The action of these molecules may be masked by components in the whole lysate or alternatively, their concentrations were too low to produce any immediate effects in animals immunized with the whole lysate. It is possible that these molecules are responsible for some of the pathological effects observed with trypanosome infections in animals. 238 Six of the MAbs (# 20,65,91,148,236 and 401) are potentially useful for the diagnosis of African sleeping sickness. These MAbs bind to both bloodstream and procyclic stages of trypanosomes. Five of six MAbs recognized T. brucei species specific antigens (Table 19), while MAb # 20 showed Trypanosoma species specificity by indirect ELISA and immunoblotting. The specificity of these MAbs will ensure a minimal cross­ reactivity with other parasite antigens when they are used as diagnostic reagents. Cross­ reactivity was detected in only one experiment where 2 of the 6 selected MAbs (# 65 and 236) showed very weak binding to acetone permeabilized L. donovani IS2D promastigotes in immunofluorescence (Fig. 24 and Fig. 25). However, binding to Leishmania promastigotes was not observed using these 2 MAbs in immunofluorescence with live, glutaraldehyde-fixed or formaldehyde-fixed parasites, in indirect ELISA with water lysates or detergent extracted lysates, and in immunoblotting. The cause of these discrepancies is unknown. It is possible that the weak binding of MAbs # 65 and 236 to acetone permeabilized Leishmania is due to non-specific binding of these Abs.

The six selected MAbs appeared to recognize protein epitopes of African trypanosomes as shown by their extreme sensitivities to proteases (Fig. 33). Chemical deglycosylation of glycoproteins by TFMS results in a complete destruction of carbohydrate side chains and the retension of the protein moiety (Sojar and Bahl, 1°87). Deglycosylation of PCF trypanosome lysates for a 1.0 hr TFMS treatment did not significantly alter the binding activities of 5 MAbs (# 20,65,91,148 and 401) (Fig. 33) indicates that epitopes recognized by these MAbs are not carbohydrates. The greater decline in binding activity (37% reduction) observed with MAb # 236 could be due to the destruction of carbohydrate epitope(s) recognized by this MAb or due to the partial inactivation of proteins by TFMS as reported previously (Sojar and Bahl, 1987). The protease sensitivity exhibited by MAb # 236, however, denotes the latter. With the exception of MAb # 65, none of the trypanosomal antigens recognized by the selected MAbs bound to Con-A agarose (Fig. 34) suggesting that these antigens are not glycoproteins containing alpha-D- mannose or alpha-D-glucose. The fact that only a portion of the antigens recognized by MAb # 65 bound to the Con-A column (Fig. 35) supports the possibility of varying types of carbohydrate side chains attached to these antigens.

Although the locations of the trypanosome antigens recognized by these 6 MAbs are unclear, the antigens are not likely to be surface exposed on the trypanosomes since no 239 fluorescence binding was observed on live parasites with any of the MAbs. Only MAbs # 20 and 65 showed very weak binumg to formaldehyde-fixed trypanosomes (Fig. 23 and Fig. 25, respectively) indicating that antigens recognized by these two antibodies could be closely associated with the surface membrane of PCF trypanosome^. The distribution of antigens recognized by MAb # 65 in the Triton X-l 14 detergent phase and the 10^x g centrifuged bloodstream trypanosome pellet further indicate the membrane association of these antigens. Antigens recognized by the remaining MAbs (# 20,91,148,236 and 401) are likely to be cytoplasmic non-membnjie components since they are found mainly in the 10^x g parasite supernatants (Table 20). MAb # 236 appeared to recognize antigens that localized at a specific organelle which could be the nucleus.

MAbs # 20 bound to a Trypanosoma specific antigen with a molecular weight of 76 KD in PCF of T. brucei species and 69 KD in the bloodstream form of T. b. rhodesiense and in the PCF of T. congolense. This slight shift in molecular weight could represent different post-translation modifications in these organisms. Amino acid sequence analysis of the immunoaffinity purified 76 KD protein revealed the 11 amino acid sequence with the N-terminal Proline (Table 22). Similarities between the newly determined amino acid sequence and the protein sequences already available in the database at the National Biomedical Research Foundation (NBRF) protein library were scanned using a FAS IP computer program (Pearson and Lipman, 1985). Although human complement C5 fragment gave the best alignment with 62.5% identity in 8 overlapping amino acids, it is difficult to deduce any significance of this finding considering the limited sequence data (11 amino acids) available.

MAb # 148 recognized 2 antigen bands (65 KD and 75 KD) in the reducing SDS PAGE gels of T. b. rhodesiense bloodstream stage and procyclic stage lysates. However, only one antigen band (75 KD) was observed in gels of the procyclic stages of T. b. brucei (Fig. 28) and two closely overlapping antigen bands at 75 KD in T. b. gambiense. This difference in immunoblot patterns suggests the possibility of using MAb # 148 for distinguishing T. b. rhodesiense from the other two 7. brucei subspecies. Subspecies identification has been difficult with the 7. brucei species. Even with the aid of specific DNA probes and isoenzyme techniques (Godfrey et al., 1987), 7. b. rhodesiense and 7. b. brucei remain indistinguishable with the exception of their relative sensitivities to lysis by human serum (Rickman and Robson, 1970; 1972; Rifkin, 1978). The use of MAb # 148 for subspecies identification was explored by immunoblotting with lysates of other frozen 240 stabilite clones of T. b. gambiense, T. b. brucei and T. b. rhodesiense. The same immunoblotting patterns were seen as described above with all three subspecies of trypanosomes (Fig. 45). Interestingly, a slightly different antigen pattern was observed with non-reducing SDS-PAGE gels (Fig. 46). Two bands of similar intensity were observed with T. b. rhodesiense at 65 KD and 75 KD and two closely overlapping bands with equal intensity at 75 KD and a fainter band at 65 KD for T. b. gambiense. An intense band at 75 KD and a lighter band at 65 KD was seen with T. b. brucei. Together these results suggest a possible role for MAb # 148 to distinguish T. brucei subspecies. Further testing of well characterized trypanosome clones from other laboratories may confirm the utility of this MAb in subspecies identification.

Although MAbs # 65,91 and 401 showed similar immunoblotting patterns (2 bands in the 110 KD region) (Fig. 30, Fig. 31 and Fig. 32), they appear to recognize different antigens as shown by Con-A binding, by their relative sensitivity to heat and proteases, by their differential solubility. These antibodies recognize distinct epitopes as determined by the competitive solid-phase RIA in which the binding of radiolabelled MAbs to trypanosome lysates were inhibited only by their homologous unlabelled Abs (T able 25). In addition, positive binding was observed only with homologous antibodies and noi with heterologous MAbs in an antigen trapping experiment (Table 26), further confirming that the different antigens were recognized by the 3 MAbs. Similarly, results from the competitive solid-phase RIA and the sandwich ELISA allowed the designation of distinct epitopes on different antigens recognized by the 3 remaining MAbs (# 20,148 and 236). Furthermore, the positive results obtained with homologous MAbs in an antigen trapping sandwich ELISA indicated the presence of repeated epitopes present on the antigen molecules. It is therefore not surprising that by mixing the different biotinylated MAbs to form a detection mixture, an increased signal was achieved (Fig. 41).

The use of an unlabelled MAb mixture as antigen capture reagent and biotinylated MAb mixture as detecting reagent allowed the optimization of the sandwich ELISA for sensitive detection of trypanosome antigens. In an artificial system where known quantities of antigen were added to normal human serum, the assay was capable of detecting as little as 1-3 ng/well, equivalent to 300 trypanosomes/well (i.e. 10-30 ng/ml or 3000 trypanosome s/ml). The MAbs used in the assay are specific to Trypanosoma spp. and showed no cross-reactivity with Leishmania lysates. This implies that the assay is 241 potentially useful in the diagnosis of both T. b. rhodesiense and T. b. gambiense infections of humans, T. b. brucei infections of cattle and possibly T. congolense infections of cattle.

Circulating parasite antigens were detected in the sera of mice as early as 3 days post-infection. Trypanosomes were not observed microscopically in these sera until the 10th day post-infection. Antigens levels remained at a low but detectable level until the 17th day post-infection when a rapid increase in detectable antigens in sera was observed (Fig. 43A). In contrast, antigen in urine was not detected until the 10th day post-infection and remained detectable in urine thereafter (Fig. 43B). An interesting temporal relationship was observed between the detectable antigen levels in mouse sera and urine (Fig. 43). The rise in antigen levels in sera was followed by an increase of antigen levels in urine. These results may represent the continous clearance of circulating, soluble parasite antigens from mouse sera via excretion into urine.

Although circulating trypanosome antigens were not detected in mouse sera until the 3th day post-infection when rabbit antibodies made against the whole lysate of T. b. rhodesiense PCF were used in the sandwich ELISA (Fig. 44), a persistently high antigen level was observed thereafter. However, parasite antigens were not detected in the urine of mice until 22 days post-infection (Fig. 44B), when the death of the animals was imminent The different antigen profiles obtained using the rabbit antibodies and the MAb mixture may reflect the different types of trypanosome antigens recognized by these antibodies. The higher detectable antigen signals obtained using rabbit antisera as compared to the MAbs might also reflect the greater variety of antigens recognized by the rabbit antibodies.

The data show that the six selected MAbs (# 20,65,91,148,236 and 401) used in the double antibody sandwich ELT3 A are potentially useful diagnostic tools for detection of active T. brucei spp. infections. Indeed, a preliminary study on a number of sera from humans infected with T. b. rhodesiense has shown the presence of trypanosomal antigens in these sera (Fig. 47). Although the rabbit antibodies made against the whole lysate of PCF trypanosomes gave better detection signals in general, both the MAb mixture and the rabbit antibodies gave good correlated results in their detection of parasite antigens in both positive and negative control sera (Fig. 47 and Fig. 48). Positive results obtained in sera from all 39 T. b. gambiense-mfected patients from Daloa using the MAb mixture (data not shown) thus confirm the diagnostic potential of these selected MAbs in both the Gambian form and the Rhodesian form sleeping sickness. The fact that these MAbs can detect 242 parasite antigens in urine of trypanosome-infected animals suggests a potential alternative procedure of using these MAbs in the diagnosis of African human sleeping sickness. Since urine specimens are usually available regardless of a patient's medical condition and the collection procedure is simple (Gooch, 1985), considerable interest has developed concerning the application of rapid antigen detection using aliquots of urine in diagnosis of infectious diseases including bacterial meningitis (Ingram et al., 1983), Schistosomiasis (Gold et al., 1969; Ripert et al., 1988) and Chagas’ disease (Katzin et al., 1989). The better signals obtained in urine samples when compared to sera samples using the 6 selected MAbs in the sandwich ELISA (Fig. 43) infers a viable alternative for diagnosis of human sleeping sickness by antigen detection in urine. The present results thus indicate that these MAbs and their antigens are of value in the diagnosis of both Gambian and Rhodesian forms of human sleeping sickness. By screening the expression library of the procyclic trypanosomes using these MAbs, recombinant clones containing antigens of interest could be identified. These clones would be useful for mass production of these antigens that may facilitate the development of simplified field diagnostic tests for human sleeping sickness. 243 Discussion

The objective of my thesis research was to develop effective serodiagnostic tests for both Gambian and Rhodesian forms of African human sleeping sickness. Non-variant antigens shared among bloodstream and procyclic stages of the two Trypanosoma brucei subspecies were targeted as being the most widely conserved parasite molecules available to host immune responses, and therefore, being of most potential diagnostic value. My initial attempt was to utilize procyclic culture form (PCF) trypanosomes in a simple agglutination assay for detecting anti-trypanosome antibodies in sera of infected animals - die Procyclic Agglutination Trypanosomiasis Test (PATT). PCF trypanosomes were targeted because they are devoid of the variant surface glycoproteins (VSG) and they share many common antigens with bloodstream trypanosomes (Shapiro and Pearson, 1986). Some of these common antigens are found on the surface of procyclics and are shown to elicit an antibody response in experimentally infected animals (Gardiner et al., 1983). The cell-surface expression of these non-variant antigens make procyclic stages potentially ideal probes for the detection of anti-trypanosome antibodies in host sera. Indeed, results from the present study have proven the value of the PATT in detecting anti-trypanosome antibodies in the sera of vervet monkeys experimentally infected with T. b. rhodesiense and in sera of T. b. gambiense -infected patients from the Cote d'Ivoire (Chapter 1). Anti- procyclic antibodies were detected by the PATT in vervet monkey sera as early as 7 days post-infection with T. b. rhodesiense. Positive agglutination titres were obtained with sera from monkeys with active, untreated infections and with sera taken from monkeys soon after successful drug cure. Sera from T. b. gambienese-infectcd humans also had anti- procyclic antibodies. Similar positive agglutination results were also observed using the PATT with additional sera from trypanosome-confirmed patients from Sudan and Kenya (chapter 3). No agglutination was observed with preinfection sera from vervet monkeys, control human sera from uninfected Canadians and control sera from Americans working in the Lambwe valley, Kenya. Together these results confirm the diagnostic value of procyclic trypanosomes in human African sleeping sickness.

The persistence of anti-trypanosomal antibodies after the elimination of parasites by trypanocidal drug therapy complicates the accurate assessment of the infection status of individuals (Luckins et al., 1978). This was observed with the PATT using sera collected from trypanosome-infected vervet monkeys (Chapter 1) and humans (Chapter 3) after successful drug treatment. Therefore, serodiagnostic tests based on antibody detection may 244 not be indicative of active infection. A considerable improvement in the diagnosis of sleeping sickness would be to detect, in the sera of patients, the presence of circulating antigens. A specific and sensitive detection assay would confirm the presence of parasite molecules in the patients indicating an active infection. It could possibly eliminate the need for parasitological confirmation of infection (WHO, 1981). I set out to develop an antigen detection method which would allow immunodiagnosis of infections with both Rhodesian and Gambian forms of sleeping sickness. The strategy was to detect parasite antigens using antibodies specific for non-VSG molecules of African trypanosomes which might be released into the blood of infected animals after lysis of organisms by host effector mechanisms, both induced and otherwise. The PATT results demonstrated the presence of anti'procyclic surface antibodies in trypanosome-infected vervet monkeys and humans (Chapter 1). It seemed rational therefore to attempt to use monoclonal or polyclonal antibodies raised against surface membranes of T. brucei spp. PCF for antigen detection in sera from infected animals. Circulating trypanosomal antigens were indeed detected in sera from trypanosome-infected mice (Chapter 2) and sera from parasite-infected patients by an antigen-trapping sandwich ELISA. However, limited success was obtained using this sandwich ELISA when a larger number of sera from confirmed sleeping sickness patients were tested. This is not surprising since we now know that the monoclonal antibodies (MAbs) employed in my assay recognized a developmentally regulated glycoprotein 'procyclin' (Richardson, 1988). Procyclin is apparently present in minute and possibly in sporadic quantities in sera of chronically infected individuals. Sandwich ELISA employing these monoclonal antibodies would therefore be unlikely to attain the required sensitivity for serodiagnosis of human sleeping sickness. Interestingly, rabbit antibodies made against crude membranes of T. b. rhodesiense PCF showed similar sensitivities as those obtained with the monoclonal antibodies when rabbit antibodies were used as detecting agents in sandwich ELISA (chapter 2). It is possible that the majority of the antibodies in the rabbit antiserum bound to 'procyclin* which is thought to be immunodominant (Richardson et al., 1986,1988). This would explain the similar results attained with these different antibodies. These results suggest that polyclonal anti-PCF antibodies in a double antibody sandwich ELISA may have potential for detection of circulating trypanosomal antigens in the sera of parasite-infected individuals.

Rabbit antibodies made against whole lysates of T. b. rhodesiense PCF were then used in an antigen-trapping sandwich ELISA. It was thought that the sandwich ELISA could be improved by using reagents which bound to a larger repertoire of trypanosomal 2 4 5 antigens. Results presented in Chapter 3 show that the sandwich ELISA using rabbit anti- PCF whole lysate antibodies was effective in revealing the infection status of T. b. rhodesiense-infected vervet monkeys and patients from the Lambwe Valley, Western Kenya. Trypanosomal antigens were detected as early as 7 days post-infection in experimentally infected vervet monkeys. Antigens were detected in sera from monkeys or Kenyan patients with active, untreated infections while all preinfection sera collected from vervet monkeys or mice or control sera from uninfected individuals were negative. Results from the sandwich ELISA thus correlated with the infection status of vervet monkeys or humans infected with T. b. rhodesiense. Further testing of well-documented sera from T. b. gambiense-infected patients from Daloa (Cote d'Ivoire) and from Sudan have clearly demonstrated the diagnostic potential of this ELISA assay for T. b. gambiense infections. Trypanosomal antigens were detected in all sera from T. b. gambiense sleeping sickness patients from Daloa and parasitologically-confirmed Sudanese patients (Chapter 3), but not in control sera from uninfected North Americans or sera from Toxoplasma- and Leishmania- infected patients.

The Card Agglutination Trypanosomiasis Test (CATT) is an antibody detection assay that depends on agglutination of bloodstream forms of T. b. gambiense of a particular antigenic phenotype (Magnus et al., 1978). Because it has been used successfully in the field for mass serodiagnosis of the Gambian form of sleeping sickness (Van Niewenhove et al., 1983), parallel testing of my double antibody sandwich ELISA using the rabbit anti-PCF whole lysate antibodies, the PATT and the CATT would allow a direct comparison of the sensitivities of these assays. Both the sandwich ELISA and the PATT were shown to be as effective as the CATT in detecting T. b. gambiense-infected patients from Cote d'Ivoire and Sudan (Chapter 3). In addition, the ELISA and the PATT exhibited higher sensitivities (100% in ELISA and 98.6% in PATT) than the CATT (97.3%) using confirmed parasite-positive human sera (Chapter 3). The antigen-detection sandwich ELISA excelled over the antibody-detection PATT and CATT in monitoring the trypanocidal drug treated patients. A drawback of PATT and CATT is that a greater number of patients showed persistently higher antibody titres than persisting trypanosomal antigen levels as detected by the ELISA (Chapter 3). Antibody-targeted PATT and CATT therefore cannot distinguish recently drug-cured patients from those with active infections. This is particularly important because a substantial number of drug-treated patients subsequently redevelop active parasitemias, either due to relapse, or to reinfection (Evans, 1981). In an effort to minimize the incidence of trypanosomiasis relapse, patients often receive 246 prolonged treatment with trypanocidal drugs which have severe side effects (Gutteridge, 1982). Antigen-targeted sandwich ELISA will likely be more suitable for evaluation of the efficacy of chemotherapeutic intervention than the PATT and the CATT.

Parasitological techniques are usually used in diagnosis of African sleeping sickness (WHO, 1986). Results are often equivocal due to the small number of parasites present in infected individuals (particularly with T. b. gambiense infections) and due to the fluctuations in detectable parasite numbers (Croft, 1985). This deficiency was confirmed by my results. Even with techniques such as the microhaematocrit centrifugation technique (MHCT) and mini-anion exchange centrifugation technique (m- AECT), which concentrate parasites from the blood in order to facilitate microscopic examination, trypanosomes were not detected in blood samples from some T. b. gambiense -infected patients from Cote d'Ivoire and from Sudan (Chapter 3). The infection status of patients was then confirmed parasitologically by microscopic examination of patients' CSF or lymph node biopsy materials. More than one parasitological technique was often required for the confirmation of trypanosome infections. The higher sensitivities attained by the sandwich ELISA using the rabbit anti-PCF whole lysate antibodies (100%) and the PATT (98.6%) indicated that they could be useful alternatives in pinpointing active T. b. gambiense infections. Parasitological diagnosis of Rhodesian form sleeping sickness patients with no previous history of African trypanosomiasis (primary cases) was shown to be effective when rodent inoculation of patients' blood (98.5% +) and Giemsa-stained thick blood smears (90% +) were used (Wellde et al. 1989d). However, the effectiveness of these tests dropped to 20 - 50% in patients who relapsed. Clinical diagnosis based on cerebrospinal fluid (CSF) examination was often necessary to confirm an active infection (Wellde et al., 1989d). Parallel testing of sera from both the primary cases and relapse patients from Lambwe Valley, Western Kenya, using the sandwich ELISA and the P IT gave respective positives of 98.7% and 97% for the primary case patients and 98% and 90% for the relapse patients. The indirect fluorescent antibody test (EFAT) and the complement fixation test (CFT) are the currently preferred serological tests for Rhodesian form sleeping sickness (Voller, 1977). Both of these tests are laboratory based assays. When these tests were used for mass diagnosis of Rhodesian form sleeping sickness patients from the Lambwe Valley, the IFAT was positive in 89% of the primary sleeping sickness patients and 77% of relapsed patients (Wellde et al., 1989d). Seventy-nine % of the primary patients were positive in a CFT, and 77% of the relapsed patients were considered positive (Wellde et al., 1989d). The sensitivity of both the sandwich ELISA 247 and the PATT are comparatively higher than the IFAT and the CFT. These results confirm the utility of the sandwich ELISA and the PATT for diagnosis of b *man sleeping sickness.

The sandwich ELISA using the rabbit anti-PCF whole lysate antibodies and the PATT gave highly reproducible results in three repeated experiments. The sensitivities of both assays varied only slightly with different sources of sera with an overall sensitivity of 97.4% for the sandwich ELISA and 94.3% for the PATT in diagnosis of Gambian and Rhodesian form sleeping sickness. Presently, the only other test that has a similar level of sensitivity is the CATT (90% - 98%) (Vervoort et al., 1983) and animal inoculation with patient's blood (98.5%) (Wellde et al. 1989d). Although the CATT is efficient in detecting T. b. gambiense infection, it does not detect T. b. rhodesiense infection at all. On the contrary, the animal inoculation test is effective in detecting T. b. rhodesiense but T.b. gambiense does not grow well in laboratory animals. Animal inoculation tests are costly and require longer time periods. Recently, an antigen-trapping sandwich ELISA using monoclonal antibodies against common surface plasma membrane antigens of T. b. rhodesiense PCF (Nantulya et al., 1987) has been reported (Nantulya, 1988). A prelimary study using this assay indicated a sensitivity of 93.5% in 108 Rhodesian sleeping sickness tested cases (Nantulya, 1988). However, no information regarding the assay system, the source of sera and the antigen(s) recognized by this monoclonal antibody was available. Katende et al.(1987a; 1987b) have also reported a new method for fixing and preserving procyclic trypanosomes for use in the in IFAT for diagnosis of African human trypanosomiasis. The IFAT detected 41 of 42 parasitologically diagnosed T. b. rhodesiense sleeping sickness patients and 29 of 41 confirmed T. b. gambiense -infected patients. The diagnostic value of this IFAT and the sandwich ELISA still awaits further assessment. Nevertheless, the PATT and the sandwich ELISA using the rabbit anti-PCF whole lysate antibodies appears to have a better overall sensitivity for both Gambian and Rhodesian forms of sleeping sickness than any of these reported assays.

One of the major considerations for developing diagnostic tests is the specificity of the assay. Cross-reactivity of sleeping sickness patient's sera with sera from patients with leishmaniasis (WHO, 1986) and malaria (I>. Milton Tam, Programme for Alternative Technology in Health, Seattle, personal communication) are common problems with some of the existing diagnostic assays for human sleeping sickness. The CATT was reported to have a minimum of 5% false positives and may be higher with sera from patients with other diseases (Van Meirvenne and Le Ray, 1985). When the sandwich ELISA and the PATT 248 were tested on a total of 30 sera from patients with other parasitic diseases, only 1 serum gave a very weak ELISA reaction (ELISA ratio = 2.64) while 3 sera gave weak agglutination titres in the PATT (1/20 -1/40). Although none of the sera from the uninfected North Americans or from Americans working in the endemic areas gave positive reactions in the sandwich ELISA or in the PATT, a few false positives were observed with sera from apparently uninfected individuals in endemic areas (Chapter 3). The overall specificity of the sandwich ELISA and the PATT was shown to be 95.5% and 84.5%, respectively, after testing approximately 250 human sera from uninfected individuals from Sudan and Kenya. Obviously, one of the difficulties is to confirm that the negative sera were obtained from individuals who were truly not infected with trypanosomes. Aysmptomatic sleeping sickness patients have long been suspected. In a study conducted in the Congo, infections were diagnosed on serological or parasitological grounds in three individuals who refused treatment but were found four years later to be well, without clinical symptoms (Maurice and Pearce, 1987). Some of the "negative" control endemic sera were from individuals that had no previous history of African sleeping sickness, yet their sera gave high antigen and antibody titres in the sandwich ELISA and the PATT, casting some doubt on their negative infection status. It is possible that they had been injected with trypanosomes (e.g. T. b. brucei) which are not infective in humans. Multiple tsetse fly bites over extended periods could result in significant inoculation with such trypanosomes.

Although both the double antibody sandwich ELISA and the PATT showed diagnostic value for African sleeping sickness, I thought that these assays could be improved by making monoclonal antibodies to well characterized antigens that have diagnostic potential, at least by allowing a renewable standardized set of reagents. Identification of diagnostically useful antigens was attempted to facilitate tk. idaptation of these diagnostic assays to a simpler format for field application. Pooled se \ obtained from trypanosome-infected patients was used as a probe for parasite antigen detection after parasite fractionation using high performance liquid chromatography (HPLC), immunoaffinily and immunoblotting techniques (Chapter 4). Most of these trypanosome PCF antigens appeared to be in the higher molecular weight range (> 62 KD). Immunization of mice with the target antigens yielded six trypanosome specific monoclonal antibodies. These antibodies v ere successful in trapping circulating parasite antigens in seta from trypanosome-infected mice as early as 3 days post-infection using a double antibody sandwich ELISA (Chapter 4). In addition, these antibodies also captured 249 trypanosomal antigens from the urine of infected mice. Antigen-capture tests for the detection of parasite antigens in patient's urine has been reported for other parasitic diseases including Schistosomiasis (Gold et al., 1969; Ripert et al., 1988) and Chagas' disease (Katzin et al., 1989). Such a non-invasive technique might be more acceptable to patients than one that requires the drawing of blood for serological testing. However, an extra concentration step is often required for urine testing. This might be too costly for a serodiagnostic field test of African sleeping sickness.

When the six selected MAbs were used in the antigen-capture sandwich ELISA on sera from trypanosome-infected mice, sera from human sleeping sickness patients and sera from uninfected humans from North America and Kenya, they gave similar results to those obtained with the rabbit anti-PCF whole lysate antibodies (Chapter 4). Positive ELISA reactions were observed in sera from trypanosome-infected mice and humans while no binding was obtained with the negative controls. However, the signals observed with the MAbs mixture were generally lower than those obtained with the rabbit antibodies. This implies that these MAbs are not detecting all the circulating antigens recognized by the rabbit antibodies. Alternatively, more antibodies that recognize different epitopes of the same antigens might be present in the rabbit antisera. It may be possible to make more MAbs to different epitopes of the identified antigens for use in the sandwich ELISA in order to generate a higher signal. Nevertheless, the positive results obtained with these MAbs in an antigen-trapping sandwich ELISA clearly show their utility in the diagnosis of both Gambian and Rhodesian forms of human sleeping sickness.

Presently, both the antigen-trapping sandwich ELISA and the PATT are laboratory- based assays. However, these tests need to be simplified for wider application in field tests. One approach to simplification of the antibody-dependent PATf is to use freeze- dried, fixed and stained procyclic trypanosomes as employed in the CATT. Agglutination assays using fixed parasites have been successful in field studies of other parasitic diseases including visceral leishmaniasis (Harith etal., 1988,1990). Experiments with the various fixation and staining methods described for the preparation of bloodstream trypanosomes for CATT (Magnus et al., 1978) resulted in insufficiently stained parasites with complete lost of antigenicity. However, procedures employed by Harith et al. (1988) gave better fixed ard stained procyclic trypanosomes with moderate loss of antigenicity (Chrystal McNabb, unpublished). Further testing of different fixatives, such as those described by 250 Katende et al. (1987a) for use in the IFAT for diagnosis of African sleeping sickness, and of other staining methods are necessary to establish a simplified format for the PATT.

Antibody detection assays may also be simplified by devising new formats. One such assay is performed by spotting procyclic trypanosomes, trypanosome lysates or purified antigens onto the combs of a dipstick. Incubation of test sera could be performed by placing these combs into ELISA wells containing different test sera. The presence of anti-trypanosome antibodies could be detected by protein-A conjugated colloidal gold which would give a red spot for a positive reaction. A similar format has been used for the diagnosis of malaria (Dr. Milton Tam, personal communication). A collaborative project using this proprietary assay with Dr. Tam is currently underway in Dr. Pearson's laboratory.

One of the constraints in producing a field diagnostic test for African sleeping sickness is the cost of the assay. Presendy, the World Health Organization recommends a cost of about 25 cents (U.S.) per test Since it is cosdy to cultivate procyclic trypanosomes for antigen preparation, one possible way to minimize the cost is to produce the diagnostically relevant antigens using recombinant DNA technology. One of the advantages of recombinant antigens is that they could be produced in high levels. The six selected MAbs made against the gel fractions of procyclic lysates have shown to be diagnostically useful. By screening a PCF cDNA expression library using these MAbs, recombinant clones containing antigens of interest could be identified. These clones would be useful for mass production of the antigens which could then be adapted to a suitable format for antibody detection. Alternatively, these antigens could be produced as a fusion protein conjugated with an enzyme (e.g. B-galactosidase) and used in an antigen detection assay. In this assay, MAbs could be coupled to a solid phase matrix such as ELISA wells or polystyrene beads. Test sera would be added to the antibody coated solid-phase along with the fusion antigen. The addition of appropriate enzyme substrate would be used to determine the amount of bo' J B-galactosidase coupled antigens. The amount of bound enzvme labeled antigen would be inversely proportional to the amount of circulating antigen in the test sera. A cDNA expression library for both procyclic and bloodstream trypanosomes was made (M. Liu, unpublished) and will be used in Dr. Pearson's laboratory for the assay development described above. 251 Recombinant DNA methods can be made highly sensitive and specific. These properties have promoted a rapid proliferation of DNA applications in disease diagnosis. DNA probes for diagnosis of infectious diseases have been most extensively employed for viral infections where current diagnostic methods are cumbeisome and slow (Caskey, 1987). In situ hybridization of infected tissues or cultured cells with diagnostic probes has facilitated diagnosis of infections caused by human immunodeficiency viruses (Harper et al., 1986) and herpes viruses (Caskey, 1987). Nucleic acid amplification by polymerase cnain reactions (PCR) (Saiki et al., 1985,1988) and the development of non-isotope labeled probes (Leary et al., 1983) have significantly improved the sensitivity and the handling safety of the DNA-based detection methods. Recently some of these techniques have been applied to diagnosis of parasitic diseases, including malaria (Lanar et al., 1989; Delves et al., 1989), leishmaniasis (Barker et al., 1986) and Chagas' disease (Sturm et al., 1989). The usefulness of the DNA techniques is directly dependent on the specificity of the probes employed. Many of the antigens recognized by the selected MAbs in this present study are trypanosome species-specific. Identification and sequencing of the genes coding for these antigens would allow the production of specific DNA probes, either hybridization fragments or primers for PCR, thereby allowing the use of DNA technology for diagnosis of African sleeping sickness. DNA techniques, especially PCR, greatly enhance the sensitivity of diagnostic assays. However, they are technically and fiscally demanding and are unlikely to be widely used in field tests in the Third World.

In the introduction I detailed the large number of tests currently employed in the diagnosis of African human sleeping sickness. They include a wide variety of assay systems that differ greatly in their degree of technical sophistication and sensitivity. Because of the poorly developed health infrastructure in the endemic areas an idealized diagnostic test for African trypanosomiasis has to combine simplicity, specificity, and sensitivity at a realistic cost. According to the WHO (1986) a cost-effective test for this pathogen should not exceed $1 (U.S.) per assay and ideally should cost less than 25 cents (U.S.), thereby ruling out practically all of the sophisticated laboratory-based assay techniques. The most successful test to date has undoubtedly been the CATT which elegantly packages the sensitivity of an antibody mediated agglutination test in an inexpensive format. It costs approximately 50 cents per assay and has been widely applied in West Africa. Its main shortcomings stem from its non-applicability to T.b. rhodesiense infections and from the fact that it is an antibody detection assay. Nevertheless, the CATTs success gives valuable pointers to the optimal design of improved field 252 serodiagnostic tests. For instance, once the appropriate fixation and staining regimens have been elucidated, procyclic T.b. rhodesiense trypanosomes could be employed in a PATT of similar design to the CATT that would detect antibodies to both subspecies of T. brucei. The PATT assay could also be adapted to a dipstick format as described above. Culturing procyclic trypanosomes for the PATT assay is, however, relatively expensive. An alternative approach for an improved antibody-based test would be to produce sufficient amounts of diagnostically useful antigens by recombinant DNA techniques to allow the use of antigen-coated latex beads in agglutination tests. Latex beads coated with anti- trypanosome MAbs would provide a simple agglutination test for the detection of circulating parasite antigens. Collaborative research presently under way with the Programme for Alternative Technologies in Health (Seattle, Washington, U.S.A.) aims to adapt the PATT and antigen detecting assays detailed in this present work to a simplified format suitable for field conditions.

African human trypanosomiasis is one of a number of tropical parasitic diseases that still pose serious health threats for much of humanity. They include malaria, schistosomiasis, Chagas' disease, leishmaniasis and filariasis. All have complex life cycles involving invertebrate vectors and are also diagnostically liiTicult for various reasons. Schistosomes and filarial worms occur in the patient's tissues or body fluids but are frequently inaccessible and present in minute amounts, hindering parasitological detection. The remainder are also difficult to detect parasitologically due to their mainly intracellular distribution (Maurice and Pearse, 1987). As with African trypanosomes, serodiagnosis promises to be the most effective diagnostic approach to these pathogens. However, differences in host-parasitc interaction between parasite species imply that the criteria for choosing diagnostically valuable antigens will vary. For instance, Chagas1 disease is caused by Trypanosoma cruzi, but this parasite differs profoundly from its African congeners in that it is largely intracellular in the mammalian host and it lacks a variable antigen coat (Miles, 1983). Development of an antigen detecting assay for T. cruzi is critically important for assessing active infections but it is a formidable project at present due to the low numbers of parasite antigens in the bloodstream of chronically infected patients (Araujo et al., 1984; Afffanchino et al., 1989). Attention is therefore being focused on antibody-based (Van Meirvenne and Le Ray, 1985) and PCR-based (Sturm et al., 1989) assays although the latter technique is unlikely to achieve widespread use in endemic areas due to its low cost effectiveness and technical complexity. In malaria, antibodies develop quickly after invasion of the blood by parasites and high antibody levels 253 are soon established (Voller et al., 1975). Prevalence and level of anti-malarial antibodies can therefore be used not only diagnostically but also as an index of malaria transmission pattern and these data are especially valuable when used in conjunction with conventional parasitological examination (Voller and DeSavigny, 1981). However, different species of malaria parasites differ somewhat in their clinical course, pathology and treatment. Major problems with the^i antibody-targeted tests are that they do not necessary denote an active infection and can not identify which species of malaria parasite is responsible for the infection (WHO, 1988). The emergence of drug resistant malaria parasites (Wilson et al., 1989) has renewed the interest in development of more effective diagnostic tests for malaria. Diagnostic techniques based on DNA probes (Barker et al., 1989) or antigen capture for detection of parasites in patient's blood and sporozoites in mosquitos have been developed for field use. As with malaria, visceral leishmaniasis is characterized by easily detectable anti-Leishmania antibodies and thus antibody-detection based serodiagnostic assays are generally used (Bray, 1976; Voller and De Savigny, 1981). However, their extensive use has been limited due to cross-reactions with other coendemic diseases (e.g. Chagas' disease and malaria) and poor sensitivities (Bray, 1980). Much effort has been concentrated on improving assay specificities and recently on the development of antigen- detection assays and species-specific DNA probes (Barker et al., 1986) because the cutaneous and mucocutaneous forms of leishmaniasis are not easily diagnosed by antibody assays (Bray, 1980). A simple agglutination assay using L. donovani promastigotes has recently been developed for field diagnosis of visceral leishmaniasis and is presently undergoing large-scale field evaluation (Harith etal., 1987). The main difficulty in developing effective diagnosis tests for filariasis stem* from the extensive cross-reactivity of filarial antigens which make it difficult to produce specific immunological probes (Voller and De Savigny, 1981). Nevertheless, progress has been made, especially with species- specific and stage-specific MAbs and specific DNA probes for these parasites (Lai et al., 1987; Nanduri and Kazura, 1989; Bradley, 1987). Similarly, diagnosis of schistosomiasis has been enhanced by identifying specific target antigens (Simpson etal., 1983; Knight et al., 1984) and antigen-detection methods using MAbs have been developed for capturing parasite antigens in patient's sera and urine (Newport et.al., 1988; Ripert et al., 1988).

As with the recently developed diagnostic assays for other parasitic diseases, the PATT and the sandwich ELISA have improved the sensitivity and the specificity of the currendy available diagnostic tests for African sleeping sickness. Identification of 2 5 4 diagnostically important trypanosomal antigens has allowed the production of trypanosome-sperific MAbs (Chapter 4) which were shown to useful in an antigen-capture ELISA. The use of recombinant DNA techniques in the search for candidate antigens could lead to the simplification of these assays for a field format and allow the identification of additional diagnostic antigens for overall improvement of the assay sensitivity. The possibility of applying simplified DNA techniques such as PCR will undoubtly open different avenues for specific diagnosis of African sleeping sickness. The challenge would be to simplify these sophisticated techniques to a format that is cost-effective and could be reproduced by untrained field technicians.

Immunodiagnosis of African sleeping sickness represents only one minor part of the ongoing immunological investigation of this disease. The main thrust of such investigations has been directed towards a better understanding of host-parasite interactions with the objectives of identifying the effector mechanisms important for vaccine development and for the pathological manifestations of this disease. We now understand in some detail, the role of the variant surface glycoproteins in the survival of the bloodstream trypanosomes. However, very little is known about the nature and the role of other trypanosomal antigens in the disease process. Identification of trypanosomal antigens in the bloodstream of parasitized individuals serves as a doorway to understanding the host- trypanosome interactions which occur in African sleeping sickness. 2 5 5 Literature Cited

Ad£m, E., N&sz, I., Lengyel, A., Erdel, J. and Fachet, J. (1986) Multiple copies of identical eptiopes on the adenovirus hexon. Molecular Immunology 23,755-759. Afffanchino, J. 1., Ibanez, C. F., Luquetti, A. O., Rassi, A., Reyes, M B., Macina, R. A., Aslund, L., Pettersson, U. and Frasch, A. C. C. (1989) Identification of a Trypanosoma cruzi antigen that is shed during the acute phase of chagas' disease. Molecular and Biochemical Parasitology 34,221-228. Aiyedum, B. A. and Amodu, A. A. (1973) The capillary tube-agglutination (CA) test A proposed aid for the diagnosis of trypanosomiasis in the field. Tropenmedizin und Parsitologie 24, 67-71. Aiyedum, B. A., Amodu, A. A. and Bidwell, D. E. (1976) Parrallel evaluation of serological tests applied in African trypanosomiasis: a WHO collaborative study. Bulletin of the World Health Organization 54,141-147.

Allen, J. (1985) Immunodiagnostics in clinical microbiology. In: Rapid detection and identification of infectious agents. (Kingsbury, D. T. and Falkow, S., eds.), pp. 279- 292, Academic Press, Inc. New York.

Almeida, J. D., Rubenstein, D. and Stott, E. J. (1971) New antigen-antibody system in Australia antigen positive hepatitis. Lancet 2,1225 -1227. Anderson, N. L., Parish, N. M., Richardson, J. P. and Pearson, T. W. (1985) Comparison of African trypansomes of different antigenic phenotypes, subspecies and life cycle stages by two-dimensional gel electrophoresis. Molecular and Biochemical Parasitolgy 16,299-314.

Apted, F. I. C. (1970a) Clinical manifestations and diagnosis of sleeping sickness. In: The African trypanosomiases (Mulligan, H. W., ed.), pp. 661-683, Allen and Unwin, London.

Apted, F. I. C. (1970b) Treatment of human trypanosomiasis. In: The African trypanosomiasis (Mulligan, H. W., ed.) pp. 684-710, Allen and Unwin, London. Apted, F. I. C. (1980) Present status of chemotherapy and chemoprophylaxis of human trypanosomiasis in the eastern hemisphere. Pharmacology and Therapeutics 11,391- 413.

Araujio, F. G., Heilman, B. and Tighe, L. (1984) Antigens of Trypanosoma cruzi detected by different classes and subclasses of antibody. Transactions of the Royal Society of Tropical Medicine and Hygiene 78,672-677. Asahi, H., Hosaka, Y., Hayashi, S. and Torres, R. B. E. (1977) The use of counterelectrophoresis in the immunodiagnosis for paragonimiases. Japanese Journal of Parasitology 26, 36-41. 256 Ashcroft, M. T. (1960) A comparision between a syringe passed and a tsetse fly transmitted line of a sirain of Trypanosoma rhodesiense. Annals of Tropical Medicine and Parasitology 54,44-53. Askonas, B. A. (1985) Macrophages as mediators of immunosuppression in murine African trypanosomiasis, hr. The biology of trypanosomes (Hudson, L. ed.) Current Topics in Microbiology and Immunology, volume 117, pp. 119-127, Springer-Verlag; Berlin. Bacchi, C. J., Nathan, H. C., Hutner, S. H. McCann, P. P. and Sjoekdsma, A. (1980) Polyamine metabolism: a potential therapeutic target in trypanosomes. Science 210, 332-334. Bafort, J. M. and Schmidt, H. (1983) Experimental chronic Trypanosoma brucei rhodesiense infection in Microtus montanus. American Journal of Tropical Medicine and Hygiene. 32, 968-975. Baker, J. R. (1970) Techniques for the detection of trypanosome infections. In: The African trypanosomiases (Mulligan, H. W. ed.) pp. 67-96, Allen and Unwin, London. Balber, A. E. and Ho, L. M. (1988) Trypanosoma brucei gamnbiense: partitioning of glycopeptides of bloodstream and procyclic forms in Triton X-l 14. Experimental parasitology 65,290-293. Bancroft, G. J. and Askonas, B. A. (1982) Regulation of antibody production of protozoal infections. Clinical Immunology and Allergy. 2,511-540. Barker, D. C., Gibson, L.J., Kennedy, W. P. K., Nasser, A. A. A. A. and Williams, R. H. (1986) The potential of using recombinant probes for the identification of tropical Leishmania. Parasitology 91, S139-S174.

Ba*.y, J. D. (1986) Surface antigens of African trypanosomes in the tsetse fly. Parasitology Today 2, 52-54.

Basson, W., Parge, M. L. and Myburgh, D. P. (1977) Human trypanosomiasis in South Africa. South Africa Medir Journal 51,453-457. Beat, D. A., Stanley, H. A., Choromanski, L., MacDonald, A. B. and Honigberg, B. M. (1984) Nonvariant antigens limited to bloodstream forms of Trypanosoma brucei brucei and Trypanosoma brucei rhodesiense. Journal of Protozoology 31,541-548. Bernards, A., DeLange, T., Michels, P. A. M., Liu, A. Y. C., Huisman. M. J. and Borst, P. (1984) 1 wo modes of activation of a single surface antigenic gene of Trypanosoma bruce; Cell 36,163-170.

Bestagno, M. Cerina, A., Riva, S. and Ricotti, G. C. B. A. (1987) Improvements of western blotting to detect monoclonal antibodies. Biochemical and Biophysical Research Communications 146,1509-1514.

Bidwell, D. S. and Voller, A. (1981) Malaria diagnosis by enzyme-linked immunosorbent assays. British Medical Journal 282,1747-1748. 257

Bienen, E. J., Hammadi, E. and Hill, G. C. (1981) Trypanosoma brucei: biochemical and morphological changes during in vitro transformation of bloodstream to procyclic- trypanomastigotes. Experimental Parasitology 51,408-417. Bienen, G. L., Hill, G. C. and Shin, K. (1983) Elaboration of mitochondrial function during Trypanosoma brucei differentiation. Molecular and Biochemical Parasitology 7, 75-80. Binz, G. (1972a) Observations on levels of immunoglobulin M in confirmed cases of Trypanosoma rhodesiense infection in the Lambwe Valley. Bulletin of the World Health Organization 47,751-755.

Binz, G. (1972b) An evaluation of the capillary and latex agglutination and heterophile antibody tests for the detection of Trypanosoma rhodesiense infections. Bulletin of the World Health Organization 47,773-778. Binz, G., Timperman, G.and Hutchinson, M. P. (1968) Elevation of serum immunoglobulin M as a screening technique for trypanosomiasis. Bulletin of the World Health Organization 38,523-545. Birk, H.-W. and Koepsell, H. (1987) Reaction of monoclonal antibodies with plasma membrane proteins after binding on nitrocellulqse: renaturation of antigenic sites and reduction of nonspecific antibody binding. Analytical Biochemistry 164,12-22.

Black, S. J., Sendashonga, C. N., O'Brien, C., Borowy, N. K., Naessens, M., Webster, P. and Murray, M. (1985) Regulation of parasitemia in mice infected with Trypanosoma brucei. Current Topics in Microbiology and Immunology 117,93-118. Bond, G. J. and Charlier, J. (1975) L'hdmagglutination indirecte en capillaire: une mdthode de diagnostic de la trypanosomiase applicable sur le terrain. Annales de la Socidtd Beige de Mddecine Tropicale 55,559-569.

Boothroyd, J. C. (1985) Antigenic variation in African irypanosomes. Annual Review of Microbiology 39,475-502.

Bordier, C. (1981) Phase separation of intergral membrane proteins in Triton X I14 sloution. Journal of Biological Chemistry 256,1604-1607. Boreham, P. F. L. and Facer, C. A. (1974) Fibrinogen and fibrinogen-fibrin degradation products in African trypanosomiasis. International Journal of Parasitology 4,143- 151.

Borst, P. (1986) Discontinuous transcription and antigenic variation trypanosomes. Annual Review of Biochemistry 46,701-732. Borst, P., Fase-Fowler, F., Frasch, A. C. C., Hoeijmaker, J. H. J. and Weijer, P. J. (1980) Characterization of DNA from Trypanosoma brucei and related trypanosomes by restriction endonuclease digestion. Molecular and Biochemical Parasitology 1, 221-246. 253 Borst, P., Van der Ploeg, M., Van Hoek, J. F. M., Tas, J. and James, J. (1982) On the DNA content and ploidy of trypanosomes. Molecular and Biochemical Parasitology 6, 13-23. Bowman, I. B. R. and Flynn, I. W. (1976) Oxidative metabolism of trypanosomes. In: Biology of the Kinetoplastida. vol. 1 (Lumsden, W. H. R. and Evans, D. A., eds.), pp. 435-476. Academic Press, New York. Bowtell, D. D. L., Saint, R. B., Rickard, M. D. and Mitchell, G. F. (1984) Expression of Taenia taeniaeformis antigens in Escherichia coli. Molecular and Biochemical Parasitology 13,173-185. Bradford, M. (1976) A rapid and sensitive method for the quantitation of microgram and quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72,248-254. Bradley, D. (1987) DNA probes for river blindness. Nature 327, 365-366. Bray, R. S. (1976) Immunodiagnosis of leishmaniasis. In: Immuology of Parasitic Infections (Cohen, S. and Sadun, E., eds.), pp. 65-76. Blackwell Scientific Publications, Oxford. Bray, R. S. (1980) Leishmaniasis. In: Immunological investigation of Tropical Parasitic Diseases (Houba, V., ed.), pp. 65-74, Churchill Livingstone, Edinburgh. Brooks, G. F. and York, M. E. (1985) Technological advances in the clinical microbiology laboratory: sensitivity, specificity and cost effectiveness. In: Rapid detection and identificationof infectious agents (Kingsburg, D. T. and Falkow, S., eds.), pp. 3-17., Academic Press, Inc., New York. Brown, J. R. (1983) Human trypanosomiases and their therapy. Pharmacy International (Netherlands) 4, 53-57.

Brown, K. N. and Williams, J. (1962) Antigens of brucei trypanosomes. Nature 194, 1253-1255. Brown, K. N. and Williams, J. (1964) The chemical composition of trypanosomes. IV. Location of antigens in subcellular fractions of Trypanosoma rhodesiense. Experimental Parasitology 15,69-86. Brown, R. C., Evans, D. A. and Vickerman, K. (1973) Changes in oxidative metabolism and ultrastructure accompanying differentiation of the mitochondrion in Trypanosoma brucei. International Journal for Parasitology 3,691-704. Brun, R. and Jenni, L. (1985) Cultivation of African and South American trypanosomes of medical or veterinary importance. British Medical Bulletin 41,122-129. Brun, R. and Schonenberger, M. (1979) Cultivation and in vitro cloning of procyclic culture forms of Trypanosoma brucei in a semi-defined medium. Acta Tropica 36, 289-292. 259 Brun, R. and Schdnenberger, M. (1979) Cultivation and in vitro cloning of procyclic culture forms of Trypanosoma brucei in a semi-defined medium. Acta Tropica 36, 289-292. Buckland, R. M. (1986) Strong signals from streptavidin-biotin. Nature 320,557-558. Burges, D. and Jerrells, T. (1985) Molecular identity and location od invariant antigens on Trypanosoma brucei rhodesiense defined with monoclonal antibodies reactive with sera from trypanosomiasis patients. Infection and Immunity 50,893-899. Butler, J. E., Peterman, J. H., Suter, M. and Dierks, S. E. (1987) The immunochemistry of solid-phase sandwich enzyme-linked immunosorbent assays. Federation Proceedings 46,2548-2556.

Biilow, R. and Overath, P. (1985) Synthesis of a hydrolase for the membrane-form variant surface glycoprotein is repressed during transformation of Trypanosoma brucei. Federal European Biochemical Societies letters 187,105-110. Biilow, R. and Overath, P. (1986) Purifiation and characterization of the membrane-form variant hydrolase of Trypanosoma brucei. Journal of Biological Chemistry 261, 11918-11923. Biilow, R., Griffiths, G., Webster, P., Stierhof, Y.-D., Opperdoes, F. R. and Overath, P. (1989b) Intracellular localization of the glycosyl-phosphatidylinositol-specific phospholipase C of Trypanosoma brucei. Journal of Cell Science 93,233-240. Biilow, R., Nonnengasser, C. and Overath, P. (1989a) Release of the variant surface glycoprotein during differentiation of bloodstream to procyclic forms of Trypanosoma brucei. Molecular and Biochemical Parasitology 32, 85-95.

Campbell, G. H., Esser, K. M. and Philips, S. M. (1978) Trypanosoma rhodesiense infection in cogenitally athymic (nude) mice. Infection and Immunity 20,714-720.

Campbell, G. H., Griswold, S., Cain, G., Giorgi, J. V. and Warner, N. L. (1981) Monoclonal antibodies to antigens of Trypanosoma rhodesiense. In: Monoclonal antibodies and T cell hybridoma (Hammerling, G. H., Hammerling, U. and Keamer, J. T., eds.), pp.323-328, Elsevier, Amsterdam.

Capbem, A., Biroud, C., Baltz, T. and Maltem, P. (1977) Trypanosoma equiperdum: eute des variations antigeniques au cours de la trypansome experimentale du lapin. Experimental Parasitology 42,6-13. Cardoso de Almeida, M. L. and Turner, M. J. (1983) The membrane form of variant surface glycoproteins of Trypanosoma brucei. Nature 302,349-352. Carrington, M., Biilow, R., Reinke, H. and Overath, P. (1989) Sequence and expression of the glycosyl-phosphatidylinositol-specific phospholipase C of Trypanosoma brucei. Molecular and Biochemical Parasitology 33,289-296. Caskey, C. T. (1987) Disease Diagnosis by recombinant DNA methods. Science 236, 1223-1229. 260 Cattand, P. (1987) Advances in the field diagnosis of human African trypanosomiasis. La Medicina Tropicale nella cooperazione alio sviluppo 3,105-108. Cattand, P., Miezan, B. T. and de Raadt, P. (1988) Human African trypanosomiasis: use of the double centrifugation of cerebrospinal fluid to detect trypanosomes. Bulletin of the World Health Organization 6,83-86. Chaiet, L. and Wolf, F. J. (1964) The properties of streptavidin, a biotin-binding protein produced by Stretomycetes. Archives of Biochemistry and Biophysis 106,1-5. Chandler, H. M., Cox, J. C., Healey, A., Macgregor, A., Premier, R. P. and Hurrell, J. G. R. (1982) An investigation of the use urease-antibody conjugates in enzyme immunoassays. Journal of Immunological Methods 53,187-194. Chou, S. and Merigan, T. C. (1983) Rapid detection and quantitation of human cytomegalovirus in urine through DNA hybridization. New England Journal of Medicine 308,921-926. Christensen, P., Kahlmeler, G., Jonsson, S. and Kronvall, G. (1973) New method for the serological grouping of streptococci with specific antibodies adsorbed to protein A- containing Staphylococci. Infection and Immunity 7, 881-885 Clarke, M. W., Barbet, A. F., Pearson, T. W. (1987) Structural features of antigenic determinants on variant surface glycoproteins from Trypanosoma brucei. Molecular Immunology 7,707-713. Clarkson, M. J. (1976) Immunologlobulin M in trypanosomiasis. In: Patholophsiology of parasitic infection (Soulsby, E. J. L. ed.), pp. 171-182, Accademic Press, New York. Clarkson, M. J. and Awan, M. A. Q. (1969) The immune response of sheep to Trypanosoma vivax. Annals Tropical Medicine and Parasitology 63,515-527. Clayton, C. E., Sacks, D. L., Ogilvie, B. M. and Askonas, B. A. (1979) Membrane fractions of trypanosome mimic the immunosuppressive and mitogenic effects of living parasites on the host. Parasite Immunology 1,241-249. Clayton, C. E., Selkirk, M. E., Corsini, A. C., Ogilvie, B. M. and Askonas, B. A. (1980) Murine trypanosomiasis: cellular proliferation and function and functional depletion in the blood, peritoneum and spleen related to changes in bone marrow stem cells. Infection and Immunity 28,824-831. Clegg, J. A. In vitro cultivation of Schistosoma mansoni. Experimental Parasitology 16, 133-147.

Collins, A. B. (1988) Immunofluorescence. In: Diagnostic immunopathology (Colvin, R. B., Bhan, A. K. and McCluskey, R. T., eds.), pp. 421-435., Raven Press, New York.

Coons, A. H., Creech, H. J., Jones, R. N. and Berliner. E. (1942) The demonstration of pneumococcal antigens in tissues by the use of fluorescent antibody. Journal of Immunology 45,159-170. 2 6 1

Croft, S. L. (1985) Serodiagnosis tests for West African trypanosomiasis. Parasitology today 1,115. Cross, G. A. M. (1975) Identification, purification and properties of clone-specific glycoprotein antigens constituting the surface coat of Trypanosoma brucei. Parasitology 71,391-417. Cross, G. A. M. (1984) Structure of the variant glycoproteins and surface coat of Trypanosoma brucei. Proceedings of the Royal Society of London (Biological Sciences) 307,3-12. Cunningham, I. (1973) New culture medium for maintainance of tsetse tissue and growth of trypanosomatids. Journal of Protozoology 21,325-329. Cunningham, M. P., Bailey, N. M. and Kimber, C. D. (1967) The estimation of IgM immunoglobulin in dried blood, for use as a screening test in the diagnosis of human trypanosomiasis in Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene 61, 688-695. Czichos, J., Nonnengasser, C. and Overath, P. (1986) Trypanosoma brucei: cis-aconitate and temperature reduction as triggers of synchronous transformation of bloodstream to procyclic trypomastigotes in vitro. Experimental Parasitology 62,283-291. Davis, J. M., Pennington, J. E., Kubleer, A.-M. and Conscience, J.-F. (1982) A simple single-step technique for selecting and cloning hybridomas for the production of monoclonal antibodies. Journal of Immunological Methods 50,161-171. De Bruijn, M. H. L. (1988) Diagnosis DNA amplification: no respect for the elusive parasite. Parasitology Today 4,293-295. De Raadt, P. (1974a) Immunity and antigenic variation: clinical observations suggestive of immune phenomena in African trypanosomiasis. In: Trypanosomiasis and Leishmaniasis, pp. 199-216 Ciba Foundation symposium 20 (new series) Associated Scientific Publishers, Amersterdam. De Raadt, P. (1974b) Trypanosomiasis and leishmaniasis with special reference to chagas' diseases. In: Ciba foundation (ed.) pp. 199-244. Associated Scientific Publishers, Amsterdam.

De Raadt, P. and Seed, J. R. (1977) Trypanosmes causing diseases in man in Africa. In: Parasitic Protozoa, volume 1; Taxonomy, Kinetoplastids and flagellates of fish (Kreier, J. P., ed), pp. 175-237, Academic Press, New York. DeLange, T. and Borst, P. (1982) Genomic environment of the expression-linked extra copies of genes for surface antigens of Trypanosoma brucei resembles the end of a chromosome. Nature 229,451-453. Delves, C. J., Goman, M., Ridley, R., G., Matile, H., Lensen, T, H. W., Ponnudurai, T. and Scaifi, J. G. (1989) Identification of Plasmodium falciparium-Mected mosquitoes using a probe containing repetitive DNA. Molecular and Biochemical Parasitology 32, 105-112. 262

Donelson, J. E. (1987) Antigenic variation in African trypanosomes. Contrib. in Microbiology and Immunology 8, 138-175. Donelson, J. E. (1988) Unsolved mysteries of trypanosome antigenic variation. In: The biology of Parasitism (Englund, P. T. and Sher, A., eds.), pp. 371-400, Alan R. Liss, Inc., New York. Donelson, J. E. and Rice-Ficht, A. C. (1985) Molecular biology of trypanosome antigenic variation. Microbiological Reviews 49,107-125. Doyle, J. J. (1977) Antigenic variation in the sailvarian trypanosomes. In: Immunity to blood parasites of animals and man (Miller, L. H., Pino, J. A. and McKelvey, J. J. Jr., eds.), pp. 31-63, Plenum Press, New York. Doyle, J. J., Hirumi, H., Hirumi, K., Lupton, E. N. and Cross, G. A. M. (1980) Antigenic variation in clones of animal infective Trypanosoma brucei derived and cloned in vitro. Parasitology 80, 354-370.

Duerden, B. I., Reid, T. M. S., Jewsbury, J. M., Turk, D. C. (1987) Microbial and Parasitic infection, pp. 160-185. Hodder and Stroughton Educational, London. Duggan, A. J. (1970) An historical prepective. In: The African Trypanosomiases (Mulligan, H. W. ed.), pp. xli-xlxxxviii, Allen and Unwin LTD., London. Duggan, A. J. and Hutchinson, M. P. (1966) Sleeping sickness in Europeans: a review of 109 cases. Journal of Tropical Medicine and Hygiene 69,124-129. Dye, W. H. (1926) The serum-formalin reaction in T. b. rhodesiense infection. Transactions of the Royal Society of Tropical Medicine and Hygiene 20,74-78.

Egan, J. E., Weber, J. L., Ballou, R., Hollongdale, M. R., Majarian, W. R., Gordon, D. M., Maloy, W. L.. Hoffman, S. L., Wirtz, R. A., Schneider, I., Wollett, G. R., Young, J. F. and Hockmeyer, W. T. (1987) Efficacy of murine malaria sporozoite vaccine development. Science 236,453-456. Englund, P. T., Hereld: D., Krakow, J. L., Doering, T. L„ Masterson, W, J. and Hart, G. W. (1938) Glycolipid membrane anchor of the trypanosome variant surface glycoprotein: its biosynthesis and clevage. In: The biology of Parasitism (Erglund, P. T. and Sher, A., eds.), pp. 401-412, Alan R. Liss, Inc., New York. Engvall, E. and Perlmann, P. (1972) ELISA. HI. Quantitation of specific antibodies by enzyme linked anti-immunoglobulin in antigen coated tubes. Journal of Immunology 109, 129-135.

Esser, K. U. and Schoenbechler, M. J. (1985) Expression of two variant surface glycoproteins on individual African trypanosomes during antigen switching. Science 229, 190-193.

Esser, K., Schoenbechler, M. J. and Gingrich, j. B. (1982) Trypanosoma rhodesiense blood forms express all antigen-specificities relevant to protection against metacyelic (insect form) challenge. Journal of Immunology 129,1715-1" i 8. 263

Evans, D. A. (1979) Cyclical transmission of Trypanosoma brucei, Trypanosoma rhodesiense and Trypanosoma congolense by tsetse flies infected with culturs-form procyclic trypanosomes. Journal of Protozoology 26,425-431. Evans, D. A. (1981) African trypanosomes. Antibiotics and Chemotherapy 30,272- 287. Evans, D. A. and Brown, R. C. (1972) The utilization of glucose and proline by culture forms of Trypanosoma brucei. Journal of Protozoology 19,686-690. Evans, D. A. and Ellis, D. S. (1983) Recent observations on the behaviour of certain trypanosomes within their insect hosts. Advances in Parasitology 22,1-42. Fairban, H. and Godfrey, D. G. (1957) The local reaction in man at the site of infection with Trypanosoma rhodesiense. Annals of Tropical Medicine and Parasitology 51, 461-470. Fairlamb, A. H. and Bowman, I. B. R. (1980) Uptake of the trypanocidal drug suramin by bloodstream forms of Trypanosoma brucei and its effects on respiration and growth rate in vivo. Molecular and Biochemical Parasitology 1,315-333. Feldmeier, H., Nogueira-Queiroz, J. A., Peixoto-Queiroz, M. A., Doehring, E., Dessaint, J. P., De Alencar, J. E., DaFalla, A. A. and Capron, A. (1985) Detection, and quantification of circulating antigen in schistosomiasis by monoclonal antibody n. The quantification of circulating antigens in human Schistosoma mansoni and haematobium : relationship to intensity of infection and disease status. Clinical and Experimental Immunology 65, 232-243. Ferguson, B., Jones, N., Richter, J. and Rosenberg, M. (1984) Adenovirus Ela gene product expressed at high levels in Escherichia coli is functional. Science 224,1343- 1346. Ferguson, M. A. J. and Cross, G. A. M. (1984) Myristylatic f the m^nbrane form of a Trypanosoma brucei variant surface glycoprotein. Journal . Biological Chemistry 259, 3011-3015. Ferguson, M. A. J., Haider, K. and Cross, G. A. M. (1985) Trypanosoma brucei varinat surface glycoprotein has a sn-l,2-dimyristylg!ycerol membrane anchor at its COOH terminus. Journal of Biological Chemistry 260,4963-4968.

Ferguson, M. A. J., Homans, S. W., Dwek, R. A. and Rademacher, T. W. (1988) Glycosyl-phosphatidylinositol moiety that anchors Trypanosoma brucei variant surface glycoprotein to the membrane. Science 239,753-759. Fink, E. and Schmidt, H. (1979) Meningoencephalitis in chronic Trypanosoma brucei rhodesiense infection of the white mouse. Tropenmedizin und Paiasitologie 30,206- 211.

Reck, S. L. and Moody, A. H. (1988) Diagnostic techniques in medical parasitology. (Fleck, S. L. and Moody, A. H., eds.), pp. 113 - 124, Butterwoith and Co., Ltd. London. 264

Flynn, I. W. and Bowman, I. B. R. (1974) The action of trypanocidal arsenical drugs on Trypanosoma brucei and Trypanosoma rhodesiense. Comperative Biochemistry and Physiololgy 48B, 261-273. Focus (1985) Bethesda Research Laboratories Technical Infonnation. 7 (#4), 12. Fox, J. A., Duszenko, M., Ferguson, M. A. J., Low, M. G. and Cross, G. A. M. (1986) Purification and characterization of a novel glycan-phosphatidylinositol specific phospholipase C from Trypanosoma brucei. Journal of Biological Chemistry 261, 15767-15771. Franco, L. T. De A. (1960) A relocidale de sedimenta^aa dos eritrdcitos na tripanossomfase humana (T. gambiense). Anais Inst. Med. Trap. Lisboa 17,967-972.

Frank, E. (1905) Therapeutische versuche bei trypanosomen-enkrankung. Munch Medicine Wochenschrift 52,2059. Abstracted in the bulletin of sleeping sickness (1909) 1, 218. Frezil, J. L., Carrie, J. and Rio, F. (1974) Application et valeur da la technique d'immunoriuorescence indirecte an depistage et i la survellance da la trypanosomiase & Trypanosoma gambiense. Cashiers O.R.S.T.O.M. Entomologie Medicale et Parasitologie 12,111-126. Frdzil, J-L., Coulm, J. a id Alary, j. C. (1977) L'immunofluorescence indirecte et la strategic de luttte conure la t- ypanosomiase humaine en Afrique Centrale. Medecine Tropicale (Ma^ceilles) 37,285-289. Frommel, T. O., Kohler, M. F. and B riber, A. E. (1987) Trypanosoma brucei brucei and Trypanosoma brucei gambiense: s age specific differences in wheat germ agglutinin binding and in cndoglycosidase H sensitivity of glycoprotein oligosaccharides. Experimental Parasito’ogy 64,104-110. Galfrd, G. and Milstei.n, C. (1981) Preparation of monoclonal antibodies: strategies and procedures. In: Methods in Enzymology (Langone, J. J. and Van Vunakis, H., eds.) vol. 73, pp. 3-52, Academic Press Inc., NY. Gall, D., Hutchinson, M. P. and Yates, W. (1957) The erythrocyte sedimentation rate in sleeping sickness. Annals of Tropical Medicine and Parasitology 51,136-140. Gallo, J. M. and Anderson, B. H. (1983) A subpopulation of trypanosomal microtubles recognized by a monoclonal antibodies to tubulin. EMBO 2,479-483.

Gallo, J. M. and Schrevel, J. (1985) Homologies between paraflagella rod proteins from trypanosomes and euglenoids revealed by a monoclonal antibody. European Journal of Cell Biology 36,163-168. Gallo, J. M., Precigout, E., Schrevel, J. (1988) Subcellular sequestration ^f an antigenically unique beta tubulin. Celt Mot. Cytoskel. 9,175-183. 265 Gardiner, P. R., Finerty, J. F. and Dwyer, D. M. (1983) Icdination and Identification of surface membrane antigens in procyclic Trypanosoma rhodesiense. Journal of Immunology 131,454-457.

Gelfand, M. (1966) The early clinical features of Rhodesian trypanosomiasis with special references to the chancre (local reaction) Transactions of the Royal Society of Tropical Medicine and Hygiene 60,376-380. Ghrayeb, J., Kaplan, P. M., Mueller W. T. and Putney, S. D. (1989) Detection of HIV antibodies using recombinant envelope and core proteins. In: HIV detection by genetic engineering methods. (Luciw, P. A. and Steimer, K. S., eds.), pp. 41 - 53, Marcel Dekker, Inc. New York and Basel. Gibbons, I., DiNello, R. L„ Greenburg, R. R., Olson, J. and Ullman, E. F. (1985) Sensitive homogeneous enzyme immunoassays for microbial antigens. In: Rapid detection and identificationof infectious agents (Kingsburg, D. T. and Falkow, S., eds.) pp. 155-163. Academic Press, Inc., New York. Gibson, W. C., Mashall, T. F. de C., and Godfrey, D. G. (1980) Numerical analysis of enzyme polymorphism: a new approach to the epidemiology and taxonomy of the subgenus Trypanozoon. Advances in Parasitology 18,175-246. Gibson, W. G, Osinga, K. A., Michels, P. A. M. and Borst, P. (1985) Trypanosomes of subgenus Trypanozoon are diploid for housekeeping genes. Molecular and Biochemical Parasitology 6,231-242. Godfrey, D. G., Scc.t, C. M., Gibson, W. C., Mehlitz, D. and Zillmann, U. (1987) Enzyme polymorphism and identity of Trypanosoma brucei gambiense. Parasitology 94, 337-347.

Gold, R., Rosen, F. and Weller, T. H. (1969) A specific circulating antigen in hamsters infected with Schistosoma mansorti: Detection of antigen in serum and urine, and correlation between antigc.ic concentration and worm burden. The American Journal of Tropical Medicine and Hygiene 18,545-552.

Gooch, W. M. (1985) Immunologic diagnosis of infectious disease by antigen detection in urine. Journal of Medical Technology 2,762-765. Goodwin, L. G. (1985) Trypanosomiasis. British Medical Bulletin 41,103-104. Gottstein, B., Lengeler, C., Bachmann, P., Hagemann, P., Kocher, P., Brossard, M., Witassek, F. and Eckert, J. (1987) Sero-epidemiological survey for alveolar echinococcosis (by Em2-ELISA) of blood donors in an endemic area of Switzerland. Transactions of the Royal Society of Tropical Medicine and Hygiene 81,960-964. Gottstein, B., Schantz, P. M. and Wilson, J. F. (1985) Serological screening for Echinococcus multibcularis infections with ELISA. Lancet i, 1097-1098. Grab, D. J. and Bwayo, J. J. (1982) Isopycnic isolation of African trypanosomes on percoll gradients formed in situ. Acta Tropica 39,3oi 366. 266 Gray, A. R. (1960) Precipitation antibody in trypanosomiasis of cattle and other animals. Nature 186,1058-1059. Gray, A. R. (1961) Soluble antigens of Trypanosoma vivax and other trypanosomes. Immunology 4, 253-261. Gray, A. R. (1965) Antigenic variation in a strain of Trypanosoma brucei transmitted by Glossina morsitans and G. palpalis. Journal of General Microbiology 41,195-214. Gray, A. R. (1967) Some principles of the immunology of trypanosomiasis. Bulletin of World Health Organization 37,177-193. Gray, A. R. (1972) Variable agglutinogenic antigens of Trypanosoma gambiense and their distribution among isolates of trypanosomes collected in different places in Nigeria. Transactions of the Royal Society of Tropical Medicine and Hygiene 66, 263-270. Gray, A. R. (1974) Antigenic similarity among isolates of Trypanosoma gambiense from different countries in Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene 68,150-151. Gray, A. R. and Luckins, A. G. (1976 < Antigenic variation in salivarian trypanosomes. In: Biology of the Kinetoplastida (lumsden, W. H. R. and Evans, D. A. eds.), vol. 1, pp. 493-452, Academic Press, N. Y. Greenblatt, H. C., Diggs, C. L. and Aikawa, M. (1983) Antibody-dependent phagocytosis of Trypanosoma rhodesiense by murine macrophages. American Journal of Tropical Medicine and Hvgiene 32,34-45. Greenwood, B. M. and Whittle, H. C. (1973) Cerebrospinal fluid IgM in patients with sleeping sickness. Lancet II, 525-527. Greenwood, B. M. and Whittle, H. C. (1980) The pathogenesis of sleeping sickness. Transactions of the Royal Society of Tropical Medicine and Hygiene 74,716-725.

Greenwood, F. C., Hunter, W. N. and Glover, J. S. (1963) The preparation of ^ 5 j. labelled human growth hormone of high specific radioactivity. Biochemical Journal 89, 114-123. Guesdon, J. L., Temyck, T. and Avrameas, S. (1979) The use of avidin-biotin interaction in immunoenzymatic techniques. Journal of Histochemistry and Cytochemistry 27, 1131-1139.

Gumett, A. M., Raper, J. and Turner, M. J. (1986) Solution properties of the variant surface glycoprotein of Trypanosoma brucei. Molecular and Biochemical Parasitology 18, 141-153.

Gutteridge, W. E. (1982) Chemotherapy. In: Modem Parasitolgy (Cox, F. E. G. ed.) pp. 287-318, Blackwell Scientific, Oxford. Gutteridge, W. E. (1985) Existing chemotherapy and its limitations. British Medical Bulletin 41,162-168. 267

Hadjuk, S. and Vickerman, K. (1981a) Antigenic variation in cyclically transmitted Trypanosoma brucei variable antignic type composition of the first parasitemia in mice bitten by trypanosome-infected Glossina moristans. Parasitoloy 83,609-621. Hajduk, S. (1984) Antigenic variation during the developmental cycle of Trypanosoma brucei. Journal of Protozoology 31,41-42. Hajduk, S. L. and Vickerman, K. (1981b) Absence of detectable alteration in the kinetoplast DNA of a Trypanosoma brucei clone following loss of ability to infect the insect vector. Molecular and Biochemical Parasitology 4,17-28. Hajduk, S. L., Klein, V. A. and Englund, P. T. (1984) Replication of kinetoplast DNA maxicircles. Cell 36,483-492. Hall, T. and Esser, K. (1984) Topologic mapping of protective and nonprotective epitopes on the vaiant surface glycoprotein of the WRATat 1 clone of Trypanosoma brucei rhodesiense. Journal of Immunology 132, 2059-2063. Harith, A. E., Kolk, A. H. J., Kager, P. A., Leeuwenburg, J., Muigai, R., Kiugu, S. and LaarmanJ. J. (1990) A simple and economical direct agglutination test for serodiagnosis and sero-epidemiological studies of visceral leismaniasis. Transactions of the Royal Society of Tropical Medicine and Hygiene (in press). Harith, A. E., Kolk, A. H. J., Leeuwenburg, J., Muigai, R., Huigen, E., Jelsma, T. and Kager, P. A. (1988) Improvement of a direct agglutination test for field studies of visceral Leishmaniasis. Journal of Clinical Microbiology 26,1321-1325. Harper, M. E., Marselle., L. M., Gallo, R. C. and Wong-Staal, F. (1986) Detection of lymophocytes expressing human T-lymphotropic virus type in lymph nodes and peripheral blood from infected individuals by in situ hybidization. Proceedings of the National Academy of Science USA 83,772-776.

Hart, D. T., Misset, O., Edwards, S. W. and Opperdoes, F. R. (1984) A comparision of the glycosomes (microbodies) isolated from Trypanosoma brucei bloodstream form and cultured form procyclic trypomastigotes. Molecular and Biochemical Parasitology 12, 25-35.

Hart, D., Langridge, A., Barlow, D. and Sutton, B. (1989) Antiparasitic drug design. Parasitology Today 5,117-126.

Hawkin, F. (1975) The human plasma resistance test for the differentiation of Trypanosoma brucei rhodesiense from T. b. brucei. Transactions of the Royal Society of Tropical Medicine and Hygiene 69,271-272.

Heisch, R. B., McMahon, J. P. and Manson-Bahr, P. E. C. (1958) The isolation of Trypanosoma rhodesiense from a bushbuck. British Medical Journal 15,1203-1204. Herbert, W. J., Parratt, D., Van Meirvenne, N. and Lennox, B. (1980) An accidental laboratory infection with trypanosomes of a defined stock. II. Studies on the serological responses 'f the patient and the identity of the infecting organism. Journal of infection 2,113-124. 2 6 8

Hereld, D., Krakow, J.-L., Bangs, J.-D., Hart, G. W. and Englund, P. T. (1986) A phospholipase C from Trypanosoma brucei which selectively cleaves the glycolipid on the variant surface glycoprotein. Journal of Biological Chemistry 261, 13813-13819. Hide, G., Gray, A., Harrison, C. M. and Tait, A. (1989) Identification of an epidermal growth factor receptor homologue in trypanosomes. Molecular and Biochemical Parasitology 36, 51-60. Hill, G. C., Shimer, S. P., Caughey, B. and Sauer, L. S. (1978) Growth of infective forms of Trypanosoma rhodesiense in vitro, the causative agent of African trypanosomiasis. Science 202,763-765. Hirumi, H., Doyle, J. J. and Hirumi, K. (1977) African trypanosomes: cultivation of animal-infected Trypanosoma brucei in vitro. Science 196,992-994. Hoare, C. A. (1970) Systemic description of the mammalian trypanosomes of Africa. In: The African Trypanosomiases (Mulligan, H. W., ed.), pp. 41-48, Allen and Unwin, Inc., London.

Hoare, C. A. (1972) The trypanosomes of mammals. A zoological monography, Blackwell Scientific Publications, Oxford and Edinburg. Hoeppli, I. R. (1959) Parasites and Parasitic Infections in Early Medicine and Science. University of Malaya Press, Singpapore.

Holder, A. A. (1983) Carbohydrate is linked through ethanolamine to the C-terminal amino acid of Trypanosoma brucei variant surface glycoprotein. Biochemical Journal 209, 261-262. Holder, A. A. and Cross, G. A. M. (1981) Glycopeptides from variant surface glycoproteins of Trypanosoma brucei. C-terminal location of antigenically cross­ reactivity carbohydrate moieties. Molecular and Biochemical Parasitology 2,135-150. Hollins, C. and Lewis-Faning, E. (1947) The sedimentation rue in the African peasant with specific reference to trypanosomiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 41,217-220.

Honigberg, B. M., Cunningham, I., Stanley, H. A., Su-Lin, K. E. and Luckins, A. G. (1976) Trypanosoma brucei: an antigenic analysis of bloodstream, vector and culture stages by the quantitative fluorescent antibody methods. Experimental Parasitology 33, 496-502.

Hoogenraad, N., Helman, T. and Hoogenraad, J. (1983) The effect of pre-injection of mice with pristane on ascites tumor formation and monoclonal antibody production. J. Immunological Methods 61,317-320. Hope-Gill, C. W. (1938) A study of the reaction rate of the serum-formalin reaction in Trypanosoma gambiense sleeping sickness. Transactions of the Royal Society of Tropical Medicine and Hy giene 31,507-509. 269 Houba, V. (1980a) African trypanosomiasis (sleeping sickness) In: Immunological Investigation of Tropical Parasitic Diseases (Houba, V. ed.) pp. 32-48, Churchill Livingstone, Edinburgh. Houba, V. (1980b) Introduction In: Immunological investigation of tropical parasitic diseases (Houba, V. ed.) pp. 1-5, Churchill Livingstone, Inc., Edinburgh, London. Hum, B. A. L. and Chantler, S. M. (1980) Production of reagent antibodies. In: Methods in Enzymology (Colowick, S. P. and Kaplan, N. O., eds.) vol. 70, pp. 104- 142. Academic Press, New York. ILRAD (1986) Annual report of the International Laboratory for Research on Animal Diseases, NsiroU, Kenya. Ingram, D. L., Pearson, A. W. and Occhiuti, A. R. (1983) Detection of bacterial antigens in body fluids with the Wellcogen Haemoplilus influenzae B, Streptococcus pneumoniae and Neisseria meningitidis (ACYW135) latex agglutination tests. Journal of Microbiology 18,1119-1121. James, M. a., Kakoma, I., Ristic, M. and Cagnard, M. (1985' Induction of protective immunity to Plasmodium falciparium in Saimiri sciureus monkeys with partially purified exoantigens. Infection and Immunity 49,476-480. Jawetz, E., Melnick, J. L. and Adelberg, E. A. (198 7s Principles of diagnostic medical microbiology. In: Review of medical microbiology (Jawetz, E. A., ed.) 7th edition, pp. 338-361, Appleton and Lange, Inc., Los Altos. Jayawardena, A. N., Waksman, B. H. and Eardley, D. D. (1978) Activation of distinct helper and suppressor T-cells in experimental trypanosomiasis. Journal of Immunology 121, 622-628. Jenni, L., Marti, S., Schweizer, J., Betschart, B., LePage, R. W. F., Wells, J. M., Tait, A., Paindavoive, P., Pays, E. and Steinert, M. (1986) Hybrid formation between African trypanosomes during cyclical transmission. Nature 322, 173-175.

Jennings, F. W. (1987) Chemotherapy of chronic trypanosomiasis: combination treatment with difluoromethylomithine and arsenicals. Tropical Medicine and Parasitology 38, 249-253.

Johnson, J. G, and Cross, G. A. M. (1977) Carbohydrate composition of variant specific surface antigen glycoprotein from Trypanosoma brucei. Journal of Protozoology 24, 587-591.

Kapaciewixz, W. and Regnier, F. E. (1983) Mobile phase selection for the high- performance ion-exchange chromatography of proteins. Analytical Biochemistry 57, 251-259.

Kapikian, A. Z., Kim, H. W , Wyatt, R. G., Cline, W. L., Arrobio, J. O., Brandt, C. D., Rodriquez, W., J., Sack, D. A., Chanock, R. M. and Parrott, R. H. (1976) Human reovirus-like agent as the major pathogen associated with "winter" gastroenteritis in hospitalized infants and young children. New England Journal of Medicine 294,965 -972. 270

Kar, S. K., Roelants, G. E., Mayor-whithey K. S. and Pearson, T. W. (1981) immunodepression in trypanosome-infected mice. VI. Comparison of immune response of different lymphoid organs. European Journal of Immunology 11,100- 105. Katende, J. M., Musoke, A. J., Nantulya, V. M. and Goddeeris, B. M. (1987a) A new method for fixation and perservation of trypanosomal antigens for use in the indirect immunofluorescence antibody test for diagnosis of bovine trypanosomiasis. Tropical Medicine and Parasitology 38,41-44. Katende, J. M., Nantulya, V. M. and Musoke, A. J. (1987b) Comparison between bloodstream and procyclic form trypanosomes for serological diagnosis of African human trypanosomiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 81, 607-608. Katzin, A. M., Marcipar, A., Freilij, H., Corral, R and Yanovsky, F. (1989) Rapid determination of Trypanosoma cruzi-urinary antigens in human chronic chagas' disease by agglutination test. Experimental Parasitology 68,208-215. Kemp, D. J., Coppel, R. L., Cowan, A. F., Saint, R. B., Brown, G. V. and Anders, R. F. (1983) Expression of Plasmodium falciparum blood-stage antigens in Escherichia coli: Detection with antibodies from immune humans. Proceedings of the National Academy of Science USA. 80, 3787-3791. King, H., Lourie, E. M. and York, D, W. (1937) New trypanocidal substances. Lancet 2, 1360-1363. Kleine, F. K. and Fischer, W. (1922) Bericht iiber die priifung von'Bayer 205’in Africa. Deutsche Medizinische wochenschrift 48,1693-1696.

Klinkert, M. Q., Ruppel, A., Felleisen, R., Link, G. and Beck, E. (1988) Expression of diagnostic 31/332 kilodalton proteins of Schistosoma mansoni as fusions with bacteriophage MS2 polymerase. Molecular and Biochemical Parasitology 27,233- 240. Knight, M., Simpson, A. J. G., Payares, G., Chaudri, M. and Smithers, S. R. (1984) Cell-free synthesis of Schistosoma mansoni surface antigens: stage specificity of their expression. EMBO 3, 213-219.

Knowles, G., Abebe, G. and Black, S. (19‘J9) Detection of parasite peptidase in the plasma of heifers infected with Trypanosoma congolense. Molecular and Biochemical Parasitology 34,25-34. Knowles, G., Black. S. J. and Whitelaw, D. D. (1987) Peptidase in the plasma of mice infected with Trypanosoma brucei brucei. Parasitology 95,291-300. Kohler, G. and Milstein, C‘. (1975) Continous culture of fused cells secreting antibody of defined specificity. Nature 256,495-497. 271 Kowk, S., Mack, D. H. and Mullis, K. B. (1987) Identification of human immunodeficency virus sequence by using in vitro enzymatic amplification and oligomer cleavage detection. Journal of Virology 61,1690-1694. Kwok, S., Mack, D. H. Sninsky, J. J., Ehrlich, G. D., Poiesz, B. J., Dock, N. L., Alter, H. J., Mildvan, D. and Grieco, M. H. (1989) Diagnosis of human immunodeficiency virus in seropositive individuals: enzymatic amplification of HIV viral sequences in peripheral blood mononuclear cells. In: HTV detection by genetic engineering methods. (Luciw, P. A. and Steimer, K. S., eds.), pp. 41 - 53, Marcel Dekker, Inc. New York and Basel. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lai, R. B., Paranjape, R. S., Briles, D. E. Nutman, T. B. and Ottesen, E. A. (1987) Circulating parasite antigen(s) in lymphatic filariasis: use of monclonal antibodies to phosphocholine for immunodiagnosis. Journal of Immunology 138,3454-3460. Lambert, P. H., Bemey, M. and Kazyumba, G. (1981) Immune complexes in serum and in cerebrospinal fluid in African trypanosomiasis. Journal of Clinical Investigation 67, 77-85. Lanar, D. E., McLaughlin, G. L., Wirth, D. F., Barker, R. J., Zolg, J. M. and Chulay, J. D. (1989) Comparison of thick film, in vitro culture and DNA hybridization probes for detecting Plasmodium falciparium malaria. American Journal for Tropical Medicine and Hygiene. 40, 3-6.

Lanar, D. E., Pearce, E. J. and Sher, A. (1985) Expression in Escherichia coli of two Schistosoma mansoni genes that encode major antigens recognized by immune mice. Molecular and Biochemical Parasitology 17,45-60.

Lanham, S. M. and Godfrey, D. G. (1970) Isolation of salivarian trypanosomes from man and other mammals using DEAS-cellulose. Experimental Parasitology 28,521-534. LeRay, D. (1975) Structures antigdniques de Trypanosoma brucei brucei (Protozoa, Kinetoplastido). Analyse immunoelectrophoretique et 6tude comparative. Annales de la Socidtd Beige de M6decine Tropicale 55,129, 311. Le Ray, D., Van Meirvenne, N. and Jadin, J. B. (1973) Immunoelectrophoretic characterization of common and variable antigens of Trypanosoma brucei. Transactions of the Royal Society of Tropical Medicine and Hygiene 67,273-274.

Leary, J. J., Brigati, D. J. and Ward, D. C. (1983) Rapid and sensitive colorimetric methods for visualizing biotin-labeled DNA probes hybridized to DNA or RNA immobilized on nitrocellulose: Bio-blots. Proceedings of the National Academy of Science USA 80,4045-4049. Levine, R. A., Wardlaw, S. C. and Patton, C. L. (1989) Detection of hacmatoparasites using quantitative buffy coat analysis tubes. Parasitology Today 5,132-134.

Lobel, H. O. and Kagan, I. G. (1978) Seroepidemiology of parasitic diseases. Annual Review of Microbiology 32, 329-347. 272

Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) protein measurement with folin phenol reagent. Journal of Biological Chemistry 193,265- 272. Luckins, A. G. (1977) Detection of antibodies in trypanosome-infected cattle by means of a microplate enzyme-linked immunosorbent assay. Tropical animal Health and Production 9, 53-62. Luckins, A. G. and Mehlitz, D. (1976) Observations on serum immunoglobulin levels in cattle infected with Trypanosoma brucei, T. vivax and T. congolense. Annals of Tropical Medicine and Parasitology 70,479-480. Luckins, A. G., Gray, A. R. and Rae, P. F. (1978) Comparison of the diagnostic value of serum immunoglobulin levels, an enzyme-linked immunosorbent assay and a fluorescent antibody test in experimental infection with Trypanosoma evansi in rabbits. Annals of Tropical Medicine and Parasitology 72,429-441.

Lumsden, W. H. R., Kimber, C. D., Evans, D. A. and Doig, S. J. (1979) Trypanosoma brucei: miniature anion-exchange centrifugation technique for detection of low parasitemias: adaptation for field use. Transactions of the Royal Society of Tropical Medicine and Hygiene 73,312-317. Magnus, E., Van Meirvenne, N., Vervoort, T., Le Ray, D. and Wery, M. (1978a) Use of freeze-dried trypanosomes in indirect fluorescent antibody test for serodiagnosis of sleeping sickness. Annales de la Socidtd Beige de Mddecine Tropicale 58,169-176. Magnus, E., Vervoort., T. and Van Meirvenne, N. (1978b) A card-agglutination test with stained trypanosomes (CATT) for the seological diagnosis of T. b. gambiense trypanosomiasis. Annales de la Socidtd Beige de Mddecine Tropicale 58,177-185.

Mancini, G., Carbonara, A. O. and Heremans, J. F. (1965) Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry 2,235-238. Mancini, P. R., Strikler, J. E. and Patton, C. C. (1982) Identification and partial characterization of plasma membrane polypeptide of Trypanosoma brucei. Biochimica et Biophysica Acta 688, 399-406.

Mansfield, J. (1990) Immunology of African tryoanosomiasis. In: The Cell Biology. Molecular Biology and Immunology of Para ^ Vyler, D. J., Pereira, M. E. A. and Wirth D., eds.), W. H. Freeman Press (in p

Mansfield, J. M. (1981) Immunology and Immunopathology of African trypanosomiasis. In: Parasitic diseases (Mansfild, J. M. ed.), volume 1, pp. 167-226; Marcel Dekker, Basel.

Mansfie5- J. M. and Kreier, J. P. (1972) Autoimmunity in experimental Trypanosoma congotsnse infections of rabbits. Infection and Immunity 5, 648-656.

Marcus, D. M. and Schwarting, G. A. (1976) Immunochemical properties of glycolipids and phospholipids. Advances in Immunology 23,203-233. 273 Matsudaira, P. (1987) Sequence from piciomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. Journal of Biological Chemistry 262,10035- 10038. Mattem, P. (1964) Technique et intdr€t 6pid£miologique du diagnostic de la trypanosomiase humaine Africaine par la recherche de la beta-macroglobulin dans le sang et dans le L. C-R. Annales de L'Institut Pasteur 107,415-421. Mattem, P. (1968) Etat et r6sultats des techniques immunologiques utilis&s a l'lnstitut Pasteur de Dalkar pour le diagnostic etT ude de la trypanosomiase humaine Africaine. Bulletin of the World HeaM ‘organization 38,1-8. Maudlin, I. and Welbum, S. C. (1987) Lectin mediated establishment of midgut infections of Trypanosoma congolense and Trypanosoma brucei in Glossina moristans. Tropenmedizin und Parasitologie 38,167-170. Maudlin, I. and Welbum, S. C. (1988) The role of lectins and trypanosome genotype in the maturation of midgut infections in Glossina moristans. Tropenmedizin und Parasitologie 39,56-58. Maurice, J. and Pearce (1987) The African trypanosomiases. In: Tropical Disease Research: A Global Partnership (Maurice, J. and Pearce, M. A., eds.), pp. 75-86. 8th programme report by UNDP/World Bank/WHO special programme for research and training in Tropical Diseases. WHO, Geneva. McCarrcn, R. M., Kempski, O., Spaatz, M. and McFarlin, D. E. (1985) Presentation of myelin basic protein by murine cerebral vascular endothelical cells. Journal of Immunology 134, 3100-3103. McCarthy, L. R. (1985) Lattex Agglutination tests for the rapid diagnosis of infectious disease. In: Rapid detection and identification of infectious agents (Kingsburg, D. T. and Falkow, S., eds.), pp. 165-175., Academic Press, Inc., New York.

McConnell, J., Turner, M. J. and Rovis, L. (1983) Biosynthesis of Trypanosoma brucei variant surface glycoproteins- analysis of carbohydrate heterogeneity and timing of post-translational modifications. Molecular and Biochemical Parasitology 8,119-135.

McLaren, M. L., Lillywhite, J. E. and Au, A. C. S. (1981) Indirect enzyme-linked immunosorbent assay (ELISA): practical aspects of standardization and quality control. Medical Laboratory Science 38,245-251. McLaughlin, J. (1982) Subcellular distribution of particle-associated antigens in Trypanosoma rhodesiense. Journal of Immunology 128,2656-2663. McLaughlin, J. (1984) Evidence for lipid-protein interactions in the attachment of antigens to a low-density membrane fraction isolated from Trypanosoma rhodesiense. Infection and Immunity. 43, 294-301.

McLaughlin, J. (1987) Trypanosoma rhodesiense: antigenicity and immunogenicity of flagella pocket membrane components. Experimental Parasitology 64,1-11. 274 Mehiitz, D. (1979) Trypanosome infections in domestic animals in Liberia. Tropenmedizin und Parsitologie 30,212-219. Mehiitz, D., Zillman, U., Scott, C. M. and Godfrey, D. G. (1982) Epedemiological studies on the reservoir of Gambiense sleeping sickness. Part III. Characterization of Trypanozoon stocks by isoenzyme and sensitivity to human serum. Tropenmedizin und Parasitologie 33,113-118. Michelotti, E. F. and Hadjuk, S. L. (1987) Developmental regluation of trypanosome mitochrondrial gene expression. Journal of Biological Chemistry 262,927-932. Miles, M. A. (1983) The epidemiology of South American trypanosomiasis biochemical and immunological approaches and their relevance to control. Transactions of the Royal Society of Tropical Medicine and Hygiene. 77,5-23. Miller, E. N., Allan, L. M. and Turner, M. J. (1984a) Topological analysis of antigenic determinants of a variant surface glycoprotein of Trypanosoma brucei. Molecular and Biochemical Parasitology 13,67-81. Miller, E. N., Allan, L. M. and Turner, M. J. (1984b) Relationship of antigenic determinants to structure vithin a variant surface glycoprotein of Trypanosoma brucei. Molecular and Biochemical Parasitology 13,309-322. Miller, L. 11, Russell, J. H., Carter, R , Food, M. F., Nussenzweig, V. and Nussenzweig, R. S. (1986) Research toward malaria vaccines. Science 234,1349- 1356. Molyneux, D. H., De Raadt, P. and Seed, J. R. (1984) African human trypanosomiasis. In: Recent advances in Tropical Medicine (Gills, H. M. ed.), pp. 53-72, Churchill Livingstone, New york.

Moseley, S. L., Echevererria, P., Seri vatana, J., Tirapat, C. and Falkow, S. (1982) Identification of enterotoxigenic Escherichia coli by colony hybridization using three enterotoxin gene probes. Journal of Infectious Diseases 145,863-871. Moseley, S. L., Huq, I., Alim, A. R. M. A., So, M., Zamadpour-Motalebi, M. and Falkow, S. (1980) Detection of enterotoxigenic Escherichia coli by DNA colony hybridization. Journal of Infectious Diseases 142,892-898. Mowatt, M. R. and Clayton, C. E. (1987) Developmental regulated of a novel repetitive protein of Trypanosoma brucei. Molecular and Celluar Biology 7,2833-2844. Mowatt, M. R. and Clayton, C E. (1988) Polymorphism in the procyclic acidic repetitive protein (PARP) gene family of Tryp" osoma brucei. Molecular and Celluar Biology 8, 4055-4062.

Mowatt, M. R. and Clayton, C. E. (1989) The procyclic acidic repetitive proteins of Trypanosoma brucei: Purification and evidence of a glycosylphosphatidylinositol membrane anchor. Journal of Biological Chemistry 264,15088-15093. Mullis, K. B., Faloona, F. A. (1987) Specific synthesis of DNA in vitro via a polymerase catalyzed chain reaction. Method in Enzymology 155,335-350. 275

Murray, M. and Urquhart, G. M. (1977) Immunoprophylaxis against African trypanosomiasis. In: Immunity to blood parasites of animal and man (Miller, L. H., Pino, J. A. and McKeveley, J. J. Jr., eds.), pp. 209-241, Plenum Press, New York and London. Murray, M., Murray, P. K. and McIntyre, W. I. M. (1977) An improved parasitological technique for the diagnosis of African trypanosomiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 77,325-326. Muller, N., Gottstein, B., Vogel, M., Flury, K. and Seedbeck, T. (1989) Application of Echinococcus multilocularis antigen in an enzyme-linked immunosorbent assay for immunodiagnosis of human alveolar echinococcosis. Molecular and Biochemical Parasitology 36,151-160. Myler, P., Allison, J., Agabian, N. and Stuart, K. (1984a) Antigenic variation in African trypanosomes by gene replacement or activation of alternative telomeres. Cell 39,203- 211. Myler, P., Nelson, R. G., Agabian, N. and Stuart, K. (1984b) Two mechanisms of expression of a predominant variant antigenic gene of Trypanosoma brucei. Nature 309, 282-284.

Nanduri, J. and Kazura, J. W. (1989) Clinical and Laboratory aspect of filariasis. Clinical Microbiology Reviews. 2,39-50.

Nantulya, V. M. (1988) Immunodiagnosis of Rhodesian sleeping sickness: Detection of circulating trypanosomal antigens in sera and cerebrospinal fluid by enzyme immunoassay using a monoclonal antibody. Bulletin de la Socit€t£ de Pathologie Exotique 81,511-512.

Nantulya, V. M., Musoke, A. J., Rurangirwa, F. R., Saigar, N. and Minja, S. H. (1987) Monoclonal antibodies that distinguish Trypanosoma congolense, T. vivaxand 71 brucei. Parasite Immunology 8,421-431. Newport, G. R., Hedstrom, R. C., Kallestad, J., Tarr, P., Klebanoff, S. and Agabian, N. (1988) Identification, molecular cloning, and expression of a schistosome antigen displaying diagnostic potential. American Journal of Tropical Medicine and Hygiene 38, 540-546.

Newton, B. A. (1974) The chemotherapy of trypanosomiasis and leishmaniasis: towards a more rational approach. In: Trypanosomiasis and Leishmaniasis with special reference to chagas' disease (Ciba Foundation symposium 20; new series) pp. 28 j - 307, Associated Scientific Publishers, Amsterdam. Ngaira, J. M„ Nantulya, V. M., Mouskoe, A. J. and Hirumi, K. (1983) Phagocytosis of antibody-sensitive Trypanosoma brucei in vitro by bovine peripheral blood monocytes. Immunology 49, 393-400. Nozais, J. P., Giordano, C., Doucet, J. and Bertrand, E. (1975) Intdrdt de L'immunofluorescence indirecte dans le diagnostic de la trypanosomiases & 276 Trypanosomiases gambiense. Bulletin de la Sxi6t£ de Pathologie Exotique 68,390- 398. Oka, M., Ito, Y., Furuya, H. and Osaki, M. (1984) Trypanosoma gambiense: immunosuppression and polyclonal B-cell activation in mice. Experimental Parsitology 58, 209-214. Olafson, R. W., Clarke, M. W., Kielland, S. L., Pearson, T. W., Barbet, A. F. and McGuire, T. C. (1984) Amino terminal sequence homology among variant surface glycoprotein of African trypanosomes. Molecular and Biochemical Parasitology 12, 287-298. Olenick, J. G., Wolff, R., Nauman, R. K. and McLaughlin, J. (1988) A flagella pocket membrane fraction from Trypanosoma brucei rhodesiense: immunogold localization and non-variant immunoprotection. Infection and Immunity 56,92-98. Olubayo, R. (1978) Trypanosomiasis of game animals. In: Pathogenicity of trypanosomes. Proceedings of a workshop held at Nairobi, Kenya, 20-23 November, 1978 (Losos, G. and Chouinard, A. eds.) pp. 87-88, International development Research center (IDRC-132e), Ottawa. Opperdoes, F. R. (1985) Biochemical peculiarities of trypanosomes. British Medical Bulletin. 41,130-136. Opperdoes, F. R. (1987) Compartmentation of carbohydrate metabolism in trypanosomes. Annual Review of Microbiology 41,127-151.

Opperdoes, F. R. and Borst, P. (1977) localization cf nine glycolytic enzymes in the microbody-like organelle in Trypanosoma brucei: the glycosome. Federal European Biochemical Societies Letters 80,360-364. Ormerod, W. E. (1976) Development of Trypanosoma brucei in the mammalian host. In: Biology of the Kinetoplastia (Lumsden, W.H. R. and Evans, D. A., eds.), vol. 1, pp. 340-393, Academic Press, New York. Ouchterlony, O. (1949) An in vitro test of the toxin-produced capacity of Corynebacterium diphtheriae. Lancet 1, 346-348.

Oudin, J. (1946) Mdthode d'analyse immunochimique par precipitation sp6cifique en milieu gelifie. CR Academic Science 222, V 5-125. Overath, P., Czichos, J. and Haas, C. (1986) The effect of cilrate/cis-aconitate on oxidative metabolism during transformation of Trypanosoma brucei. European Journal of Biochemistry. 160,175-182.

Overby, L. R. (1985) Viral hepatitis - a model for rapid diagnosis of infectious disease. In: Rapid detection and identification of infectious agents. (Kingsbury, D. T. and Falkow, S., eds.), pp. 33 - 48, Academic Press, Inc. New York. Paindavoine, P., Pays, E., Laurent, M., Geltmeyer, Y., LeRay, D., Mehiitz, D. and Steinert M. (1988) The use of DNA hybridization and numerical taxonomy in 277 determining relationships between Trypanosoma brucei stocks and subspecies. °arasitology 92,31-50. Pener, E. G., So, M., and Davis, C. E. (1989) Identification of a developmentally regulated cysteine protease of Trypanosoma brucei. Molecular and Biochemical Parasitology 33,27-32. Paris, J. Murray, M. and McOdimba, T. (1982) A comparative evaluation of the parasitological technique currently available for the diagnosis of African trypanosomiasis in cattle. Acta Tropica 39,307-16. Paris, J., Murray, M. and Agure, R. (1980) An evaluation of the sensitivity of current trypanosome parasitological diagnostic technique. Bulletin of the World Health Organization 46,2-7. Parish, N. M., Morrison, W. I. and Pearson, T. W. (1985) Identification of an antigen specific to Trypanosoma congolense by using monoclonal antibodies. Journal of Immunology 134, 593-597. Parish, N. M., Morrison, W. I. and Pearson. T. W. (1985) Identification of an antigen specific to Trypanosoma congolense using monoclonal antibodies. Journal of Immunology 134,593-599. Parratt, D. and Herbert, W. J. (1976) Heterophi'e and autoantibodies in trypanosome infections. In: Biology of Kinetoplastida (Lumsden, W. H. R, and Evans, D. A., eds.), volume 1, pp. 524-545, Academic Press, New York. Patarroyo, M. E., Amador, R., Clavijo, P., Moreno, A., Guzman, F.r Romero, P., Tascon, R., Franco, A., Murillo, L. A., Ponton, G. and Trujillo, G. (1988) A synthetic vaccine protects humans against challenge with asexual blood stages of Plasmodiumfalciparium malaria. Nature 332,158-161.

Paulnock, D. M., Smith, C. and Mansfield, J. (1988) Antigen presentation cell function in African trypanosomiasis. In: Antigen Presenting Cells: Diversity, Differentiation and Regulation, pp. 135-143. Allan R. Liss Inc., New York. Pautrizel, R., Mattem, P. and Duret, J. (1960) Diagnostic seerologique de la maladie du sommeil. n. Les anticorps fixant le complement an cours de l'infection. Bulletin de la Soci6te de Pathologie Exotique 53,878-832. Pays, E. (1985) Gene conversion in trypanosome antigenic variation. Progress in Nucleic acids Research and Molecular Biology 32,1-26. Pays, E. M., Guyaux, M., Aerts, D., Van Meirvenne, N. and Steinert, M. (1985) Telomeric reciprocal recombination as a possible mechanism for antigenic variation in trypanosomes. Nature 318, 562-564.

Pearson, T. W. and Jenni, L. (1989) Detection of hybrid phenotypes in African trypanosomes by high resolution two-dimensional gel electrophoresis. Parasitology Research 76, 63-67. 2 7 8 Pearson, T. W., Kar, S., McGuire, T. C. and Lundin, L. B. (1981) Trypanosome surface antiges: studies using monoclonal antibodies and two-dimensional gel electrophoresis. Journal of Immunology 126,823-828. Pearson, T. W., Moolo, S. K. and Jenni, L. (1987) Culture form and tsetse fly niidgut procyclic Trypanosotm brucei express common proteins. Molecular and Biochemical Parasitology 25,273-278. Pearson, T. W., Pinder, M., Roelants, G. E., Kar, S. K., Lundin, L. B., Mayor-Withey, K. S. and Hewett, R. S. (1980) Methods for the derivation and detection of anti- parasite monoclonal antibodies. Journal of Immunological Methods 34,141-154. Pearson, T. W., Roelants, G.E., Lundin, L and Mayor-whitey, K. S. (1978) Immunodepression in trypanosome-infected mice. I. Depressed T-lymphocyte responses. European Journal of Immunology 8,723-727. Pearson, T. W., Saya, L. E., Howard, S. P. and Buckley, J. T. (1982) The use of aerolysin toxin as an aid for visualization of iow numbers of African trypanosomes in whole blood. Acta Tropica 39,73-77. Pearson, W. R. and Lipman, D. L. (1988) Improved tools for biological sequence comparison. Proceedings of the National Academy of Science USA. 85,2444-2448. Pentreath, V. W. (1989) Neurobiology of sleeping sickness. Parasitology Today 5,215- 218. Perie, N. M., Tinnemans-Anggawidjaja, T. and Zwart, D. (1975) A refinement of the immunofluorescent complement fixation test for trypanosome infections. Tropenmedizin und Parsitologie 26, 399-404.

Petersons, G. L. (1983) Determination of total protein. In: Methods in Enzymology. vol 92 (Hirs, C. H. W. and Timasheff, S. N., eds.), pp. 95-119, Academic Press, New York.

Pharmacia Separation News (1980) Protein purification step by step, volume 7. Parmacia, Uppsala, Sweden.

Pinder, M, Van Melick, A. and Vernct, G. (1987) Analysis of protective epitopes on the variant surface glycoprotein of a Trypanosoma brucei brucei (DiTat 1.3) using monoclonal antibodies. Parasite Immunology 9,395-400. Poltera, A. A. (1985) Pathology of human African trypanosomiasis with reference to experimental African trypanosomiasis and infections of the central nervous system. British Medical Bulletin 41,169-174.

Poltera, A. A. and Cox, J. N. (1977) Pancarditis with valvulitis in endomyocardial fibrosis (EMF) and in human African trypanosomiasis (HAT). A study comparative histological study of four Ugandan cases. Virchows Archiv A Pathological Anatomy and Histology 375, 53-70. 279 Poltera, A. A. and Sayer, P. D. (1983) Cardiac lymph drainage of experimental trypanosomiasis in vervet monkeys. Bulletin de la Soci£t£ de Pathologie Exotique 76, 614-621. Poltera, A. A., Cox, J. N., Owor, R. (1976) Pancarditis involving the conducting system and all values in human . frican trypanosomaisis. British Heart Journal 38,827-837. Poltera, A. A., Hochmann, A., Rudin, W. and Lambert, P. H. (1980) Trypanosoma brucei brucei: A model for cerebral trypanosomiasis in mice. Clinical and Experimental Immunology 40,496-507. Poltera, A. A., Owor, R. and Cox, J. N. (1977) Pathological aspects of human trypanosomiasis (HAT) in Uganda. A post-mortem survey of fourteen cases. Virchows Archiv A. Pathological Anatomy und Histology 373,249-265. Poupin, F., Gaillez, M„ Petithory, J. C. and Savel, P. (1976) Les possibilities de l'electrosynettise dans le diagnostic des trypanosomes Africaines. Bulletin da la Soci£t£ de Pathologie Exotique 71,76-83. Purcell, R. H. Dienstag, J. L., Feinstone, S. M. and Kapikian, A. Z. (1975) Relationship of hepatitis A antigen to viral hepatitis. American Journal of the Medical Sciences 270, 61 -71. Rae, P. F. and Luckins, A. G. (1984) Detection of circualting trypanosomal antigens by enzyme immunoassay. Annals of Tropical Medicine and Parasitology 78,587-596. Rice-Fioht, A. C., Chen, K. K., Donelson, J. E. (1981) Sequence homologies near the C-termini of the variable surface glycoproteins of Trypanosoma brucei. Nature 294, 53-57. Richardson, J. P., Beecroft, R. P., Tolson, D. L., Liu, M. K. and Pearson, T. W. (1988) Procyclin: an unusal immunodominant glycoprotein surface antigen from the procyclic stage of African trypanosomes. Molecular Biochemical Parsitology 31,203-216.

Richardson, J. P., Jenni, L., Beecroft, R. P. and Pearson, T. W. (1986) Procyclic tsetse fly midgut forms and culture forms of African trypanosomes share stage- and species- specific surface antigens identified by monoclonal antibodies. Journal of Immunology 136, 2259-2264.

Rickman, L. R. and Robson, J. (1970) The testing of proven Trypanosoma brucei and T. rhodesiense strains by the blood incubation infectivity test Bulletin of World Health Organization 42,911-916. Rickman, L. R. and Robson, J. (1972) Some supplementary observation on the blood incubation infectivity test Bulletin of the World Health Organization 46,403-404. Rifidn, M. R. (1978) Identification of trypanocidal factor in normal human serum: high- density lipoprotein. Proceedings of the National Academy of Science 75, 3450-3454.

Ripert, C., Combe, A., Daulou&le, S., Appriou, M., Tribouley-Duret, J., Tribouley, J. Moyou-Somo, R., Same-Ekobo, A. and Ambassa, P. (1988) Detection with a 280 monoclonal antibody of an antigen, characteristic of the genus Schistosoma, excreted in the urine. Tropical Medicine and Parasitolgy 39,131-135. Ritz, H. (1916) On relapses in experimental trypanosomiasis: 2nd communication. Arch. Schiffs. Trop. enhyg. 20,397. Abstracted in Tropical Disease Bulletin (1917) 9, 341. Roditi, I. and Pearson, T. (1990) The procyclin coat of African trypanosomes. Parasitology Today (in press). Roditi,!., Carrington, M. and Turner, M. (1987) Expression of a polypeptide containing a dipeptide repeat is confined to the insect stage of Trypanosoma brucei. Nature 325, 272-274. Roditi, I., Dobberlaere, D., Williams, R. O., Masterson, W., Beecroft, R. P., Richardson, J. P. and Pearson, T. W. (1989a) Expression of Trypanosoma brucei procyclin as a fusion protein in E. coli. Molecular and Biochemical Parasitology 34, 35-44. Roditi, I., Schwarz, H., Pearson, T. W., Beecroft, R. P., Liu, M. K., Richardson, J. P., Buhring, H. -J., Plsiss, J., Biilow, R., Williams, R. O. and Overath, P. (1989b) Procyclin gene expression and loss of the variant surface glycoprotein during differentiation of Trypanosoma brucei. Journal of Cell Biology 108,737-746. Roelants, G.E., Pearson, T. W., Tyrer, H. W., Mayor-whitey, K. S. , Lundin , L. B. (1979)Immune depression in trypanosome-infected mice n. Characterization of the spleen cell types involved. European Journal of Immunology 9,195-199. Roffi, J., Diallo, P. B., Dedet, J. P., Garre M. T. and Pradeau, F. (1979) Diagnostic immunoenzymatique (ELISA) de la trypanosomiase humaine Africaine. Etude compar6e de plusieurs antig€nes. M&fccine Tropicale (Marseilles) 39,637-641. Ross, J. D. J. (1971) The detection of circulating trypanosomal antibodies by capillary- tube agglutination test. Annals of Tropical Medicine and Parasitology 65,327-333. Rovis, L., Muskoe, A. S. and Moloo, S. K. (1984) Failure of tryanosomal membrane antigens to induce protection against tsetse-transmitted Trypanosoma vivax or T. brucei in goats and rabbits. Acta Tropica 41,277-236.

Rytel, M. (1979) Overview of newer diagnostic methods. In: Rapid diagnosis in infectious disease. (Rytel, M., ed.), pp. 7-16, CRC Press, Boca Raton, Florida. Sacks, D. L. (1984) Induction of protective immunity using antiidiotypic antibodies: immunization against experimental against exoerimental African trypanosomiasis. In: Idiotypy in Biology and Medicine (Kohler d ., Urbain, J. and Cazenave, P., eds.), pp. 401-432., Academic Press, Inc., New fork. Sacks, D. L. and Askonas, B. A. (1980) Trypanosome induced suppression of anti­ parasite responses during experimental African trypanosomiasis. European Journal of Immunology 10,971-974.

Sacks, D. L., Bancroft G. L., Evans, W H. and Askonas, B. A. (1982) Incubation of trypanosom i-derived mitogenic and immunosuppressive products with peritoneal 281 macrophages allows recovery of biological activities from soluble parasite fractions. Infection and Immunology 36,160-168. Sacks, D. L., Selkirk, M., Ogilvie, B. M. and Askonas, B. A. (1980) Intrinsic immunosuppresive activity of different trypanosome strains varies with parasitic virulence. Nature 283,476-478. Sadun, E. H., Duxburg, R. E., Williams, J. S. and Anderson, R. I. (1963) Fluorescent antibody test for the serodiagnosis of African and American trypanosomiasis in man. Journal of Parasitology 49,385-398. Saiki, R. K. (1985) Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230,1350-1354. Saiki, R. K., Gelfand, D. H., Stoffel, S., fcharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B. and Erlich, H. A. (1988) Primer directed enzymatic amplication of DNA with a thermostable DNA polymerase. Science 239,487-491. Sailyo, R. S., Gray, A. R. and Luckins, A. G. (1980) Use of antigens of cultured Trypanosoma brucei in tests for bovine trypanosomiasis. Tropical Animal Health Production 12,127-131. Schmidt, H. and Sayer, P. (1982a) Trypanosoma brucei rhodesiense infection in vervet monkeys. I. Parasitologic, hematologic immunologic and histologic results. Tropenmedizin und Parasitologie 33,249-254. Schmidt, H. and Sayer, P. (1982b) Trypanosoma brucei rhodesiense infection in vervnt monkeys, n. Provocation of the encephalitic late phase by treatment of infected monkeys. Tropenmedizin und Parasitologie 33,255-259. Schneider, A., Eichenberger, W. and Seedbeck, T. (1988b) A microtubule-binding protein of Trypanosoma brucei which contains covalently bound fatty acid. Journal of Biological Chemistry 263,6472-6476. Schneider, A., Hemphill, A., Wyler, T. and Seedbeck, T. (1988a) A high molecular weight microtubule-associated protein from T. brucei contains randomly repeated near­ identical sequences. Science 241,459-462.

Schneider, A., Lutz, H. U., Marugg, R., Gehr, P. and Seedbeck, T. (1988c) Spectrin­ like proteins in the paraflagella rod structure of Trypanosoma brucei. Journal of Cell Science 90, 307-315. Schoenaers, F., Ncujean, G. and Evens, F. (1953) Valeur pratique de le reaction de fixation du complement dans la maladie du sommeil & T. gambiense. 1. Le diagnostic de la maladie. Annales de la Societd Beige de Mddecine Tropicale 33,141-153.

Seed, J. R. (1963) Antigens isolated from T. rhodesiense. Journal of Protozoology ’0, 380-389.

Seed, J. R. (1977) The role of immunoglobulins in immunity to Trypanosoma brucei gambiense. International Journal for Parasitology 7,55-60. 282 Seed, J. R. (1978) Competition among serologically different clones of Trypanosoma brucei gambiense. Journal of Protozoology 25,526-529. Seedbeck, T., Kiing, V., Wyler, T. and Muller, M. (1988) A 60 KDa cytoskeleton protein from Trypanosoma brucei brucei can interact with membrane and with microtubules. Proceedings of the National Academy of Sciences (USA) 85,1101- 1104. Shapiro, S. T. and Murray, M. (1982) African trypanosome antigens recognized during the course of infection om N dama and Zebru cattle. Infection and Immunity 35,410- 416. Shapiro, S. Z. and Pearson, T. W. (1986) African Trypanosomiasis: Antigens and host- parasite interactions. In: Parasite Antigens (Pearson, T. W. ed.), pp. 215-276, Marcel Dekker Inc., New York. Simpson, A. J. G., James, S. L. and Sher, A. (1983) Idetification os surface antigens of Schistomula of Schistosoma mansoni recognized bu antibodies from mice immunized by chronic infection and by exposure to highly irradiated cercariae. Infection and Immunity 41, 591-597. Smith, A. E. R., Levine, R. F. and Mansfield, J. M. (1982) Cloning of African trypanosomes in mice suppressed by cyclophosphamide treatment American Journal of Tropical Medicine and Hygiene 31,1098-102. Smith, R. D., James, M. A., Ristic, M., Aikawa, M. and Vega, C. A. (1981) Bovine babesiosis: protection of cattle with culture-derived soluble Babesia bovis antigen. Science 212, 335-338.

Sojar, H. T. and Bahl, O. P. (1987) A chemical method for the deglycosylation of proteins. Archives of Biochemistry and Biophysis 259,52-57.

Ssenyonga, G. S. Z. and Adam, K. M, (1975) The number and morphology of trypanosomes in the blood and lymph of rats infected with Trypanosoma brucei and T. congolense. Parasitology 70,255-261. Stanley, H. A., Honigberg, B. M. and Cunningham, I. (1978) Analysis of the antigenic composition of Trypanosoma brucei brucei bloodstream and culture forms by the quantitative direct fluorescent antibody methods. Journal of Protozoology 25,245- 252.

Steiger, J. and Seedbeck, T. (1986) Monoclonal antibodies against a 60 KDa phenothiazine- binding protein from Trypanosoma brucei can discriminate between different trypanosome species. Molecular and Biochemical Parasitology 21,37-45. Steiger, R. F. (1973) On the ultrastructure of Trypanosoma (Trypanozoon) bruceiin the cour se of its life cycle and some related aspects. Acta Tropica 30,64-168. Sternberg, J. C., Turner, M. P., Wells, J. M., Ranford-Cartwright, L. C., LePage, R. W. F. and Tait, A. (1989) Gene exchange in African trypanosomes: frequency and allelic segregation. Molecular and Biochemical Parasitology 34,269-280. 283 Strickler, J. E. and Patton, C. L. (1980) Trypanosoma brucei brucei: inhibition of glycosylation of the major variable surface glycoprotein by tunieamycin. Proceedings of the National Academy of Science (USA) 77,1529-1533. Strum, N. R., Degrave, W., Morel, C. and Simpson, L (1989) Sensitive detection and schizodeme classification of Trypanosoma cruzi cells by amplification of kinetoplast minicircle DNA sequences: use in diagnosis of Ch inas' disease. Molecular and Biochemical Parasitology 33,205-214. Stuart, K. D. (1987) Regulation of mitochiondrial gene expression in Trypanosoma brucei. Bioassays 6, 178-181. Tait,A. (1980) Evidence for diploidy and mating in tiypanosomes. Nature 287,536- 538. Tail, A. (1983) Sexual processes in Kinetoplastida. Parasitology 86,29-57. Tanner, M., Jenni, L., Hecker, H. and Brun, R. (1980) Characterization of Trypanosoma brucei isolated from lymph nodes of rats. Parasitology 80,383-391. Taylor, A. E. R. and Smith, V. (1983) Micro-counterimmunoelectrophoresis: a rapid screening technique for trypanosomiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 77,481-486. Terry, R. J. (1985) In: Parasite antigens in protection, diagnosis and escape. (Parkhouse, R. M. E., ed.) pp. 1-6, Current Topics in Microbiology and Immunology, vol. 120, Springer-Verlag, Berlin. Tetley, L. and Vickerman (1985) Differentiation in Trypanosoma brucei: host parasite cell junctions and their persistence during acquisition of the variable antigen coat Journal of Ceil Science 87,363-372.

Tetley, L., Turner, C. M. R., Barry, J D., Crowe, J. S., and Vickerman, K„ (1987) Onset of expression of the surface glycoproteins of Trypanosoma brucei in the tsetse fly studied using immunoelectron microscopy. Journal of Cell Science 87,363-372. Thompkins, L. S. (1985) Summary of DNA probes. In: Rapid detection and identificationof infectious agents (Kingsburg, D. T. and Falkow, S., eds.), pp. 273- 278., Academic Press, Inc., New York.

Tizard, I., Neilson, K. H., Seed, J. R. and Hall, J. E. (1978) Biological active products from African trypanosomes. Microbiological Reviews 42,661-697.

Tompkins, L. S. (1985) Summary of DNA probes. In: Rapid detection and identification of infectious agents. (Kingsbury, D. T. and Falkow, S., eds.), pp. 273 - 278, Academic Press, Inc. New York. Towbin, H., Staehelin, T. and Cordon, J. (1970) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose: procedure and some applications. Proceedings of the National Academy of Science USA 76,4350-4354. 284 Trager, W. and Jensen, J. B. (1976) Human malaria parasites in continuous culture. Science 193,673-675. Turner, C. M. R., Barry, J. D. and Vickerman, K. (1988) Loss of variable antigen during transformation of Trypanosoma brucei rhodesiense from bloodstream to procyclic forms in the tsetse fly. Parasitology Research 74,507-511. Turner, M. J. (1982) Biochemistry of the variant surface glycoproteins of salivarian trypar osomes. Advances in Parasitology 21,69-153. Turner, M. J. (1984) Antigenic variation in its biological context Philosophical Transactions of die Royal Society of London (Biological Sciences) 307,27-40. Turner, M. J. (1985) Antigens of African trypanosomes. In: Parasite Antigens in Protection Diagnosis and Escape (Parkhouse, R. M. E., ed.), pp. 141-158. Current Topics in Microbiology and Immunology, volume 120, Springer-Verlag, Berlin. Turner, M. J. (1988) Trypanosome variant surface glycoprotein. In: The Eiology of Parasitism (Englund, P. T. rud Sher, A., eds.), pp. 349-369, Alan R. Liss, Inc., New York. Van der Ploeg, L. H. T., Valerio, D., DeLange, T., Bernards, A., Brost, P. and Grosveld, F. G. (1982) An analysis of cosmid clones of nuclear DNA from Trypanosoma brucei shows that the gene for VSGs are clustered in the genome. Nucleic Acid Research 10,5905-5923. Van Knappen, K., Buys, J. and Ruitenberg, E. J. (1977) Enzyme-linked immunosorbent assay (ELISA) and its application to the setodiagnosis of African trypanosomiasis. Annales de la Society Beige de Medecine Tropicale 57,231-292. Van Meirvenne, N. and Le Ray, D. (1985) Diagnosis of African and American trypanosomiases. British Medical Bulletin 41,156-161.

Van Meirvenne, N., Jassens, P. G. and Magnus, E. (1975) Antigenic variation in syringe passed populations of Trypanosoma (trypanozoon) brucei. 1. Rationalization of the experimental approach. Annales de la Soci6t6 Beige de Mldecine Tropicale 55,1-23. Van Meirvenne, N., Magnus, E. and Vervoort, T. (1977) Comparisons of variable antigenic types produced by trypanosome strains of the subgenus Trypanozoon. Annales de la Soci6t6 Beige de M&lecine Tropicale 57,409-423. Van Nieuwenhove, S. and Declerg, J. (1983) Mass serodiagnosis and treatment of serological positives as a control strategy in Trypanosoma gambiense. In: Proceedings of the Symposium: On the diagnosis of African sleeping sickness due to T. gambiense. Smith-Kline-RIT, Antwwerp, pp. 71-175.

Vervoort, T., Magnus, E. and Van Meirvenne, N. (1983) Serological test for sleeping sickness: importance of antigen selection. In: Symposium on die diagnosis of African sleeping sickness due to T. gambiense, Anvers-Antwerp. (Crooy, P. J. ed.), pp. 47- 50, Smith Kline-RIT Biological Division, Belgium. 285 Vervoort, T., Magnus, E , Van Meirvenne, N. (1978) Enzyme-linked immunosorbent assay (ELISA) with variable antigen for serodiagnosis of T.B. gambiense trypanosomiasis. Annales de la Socidtd Beige de Mddecine Tropicale 58,177-183. Vickerman, K. (1965) polymorphism and mitochondrial activity in sleeping sickness trypanofomes. Nature 208,762-766. Vickerman, K. (1969) On the surface coat and flagellar adhesion in trypanosomes. Journal of Cell Science 5,163-193. Vickerman, K. (1971) Morphological and physiological considerations of extracellular blood protozoa. In: Ecology and Physiology of parasites (Fallis, A.M., ed.), pp. 58- 91. University Press, Toronto. Vickerman, K. (1985) Developmental cylcles and biology of pathogenic trypanosomes. British Medical Bulletin 41,105-141. Vickerman, K. and Barry, J. D. (1982) African trypanosomiasis. In: Immunology of parasitic infections (Cohen, S. and Warren, K. S., eds.), 2nd edition, pp. 204-260. Blackwell Scientific, Oxford. Vickerman, K. and Luckins, A. G. (1969) Localization of variable surface coat of Trypanosoma brucei using ferritin conjugated antibody. Nature 224,1125-1126. Vickerman, K., Tetley, L., Hendry, K. A. K. and Turner, M. R. (1988) Biology of African trypanosomes in the tsetse fly. Biology of the Cell 64,109-119. Voller, A. (1977) Serology of African trypanosomiasis. Annales de la Soci6t6 Beige de Mddecine Tropicale 57,273-280. Voller, A. and De Savigny, D. (1981) Diagnostic serology of tropical parasitic diseases. Journal of immunological Methods 46,1-29. Voller, A. and Houba, V. (1980) Malaria. In: Immunological investigation of tropical parasitic diseases (Houba, V. ed.) pp. 6-31, Churchill Livingstone, Inc., Edinburgh, London.

Voller, A., Barlett, A. and Bidwell, D. E. (1976) Enzyme immunoassays for parasitic diseases. Transactions of the Royal Society of Tropical Medicine and Hygiene 70,98- 106. Voller, A., Barlett, A., Bidwell, D. E. and Edwards, R. (1977) A comparision of istopic and enzyme immunoassays for parasitic diseases. Transactions of the Royal Society of Tropical Medicine and Hygiene 71,431-437. Voller, A., Bidwell, D. and Bartlett, A. (1980) Enzyme-linked imn. jnosorbent assay. In: Manual of Clinical Immunology (Rose, N. R. and Freeman, H., eds.), pp. 359-371, American Society fo_ Microbiology, Washington, D. C.

Voller, A., Bidwell, D. E. and Bartlett, A. (1975) A serological study of human Trypanosoma rhodesiense infections using a micro-scale enzyme-linked immunosorbent assay. Tropenmedizin und Parsitologie 26,247-251. 2 8 6

Voller, A., Bidwell, D. E. and Bartlett, A. (1979) The enzyme-linked immunosorbent assay (ELISA) - a guide with abstracts of microplate applications. Billingshurt, Dynatech Europe. Walker, P. J. (1972) Capillary concentration technique applicable to infections of T. congolense in cattle. Transactions of the Royal Society of Tropical Medicine and Hygiene 66, 348. Walls, K.W. and Schantz, P. M. (1986) Immunodiagnosis of Parasite Diseases: Helminthic diseases, vol. 1. pp. 35-73, Academic Press, New York. Weitz, B. (1960) The properties of some antigens of Trypanosoma brucei. Journal of General Microbiology 23,589-600. Wekerle, H., Sun, D., Oropeza-Wekerle and Meyermann, R. (1987) Immune reactivity in the nervous system: modulation of T-lymphocyte activation by glial cells. Journal of Experimental Biology 132,43-57. Wellde, B. T. Chumo, D. A., Readon, M. J., Mwangi, J., Asenti, A., Mbwabi, D,, Abinya, A., Wanyama, L. and Smith, D. H. (1989a) Presenting features of Rhodesian sleeping sickness patients in the Lambwe Valley, Kenya. Annals of Tropical Medicine and Parasitology 83,73-89 (Supplement #1). Wellde, B. T., Chumo, D. A., Hockmeycr, W. T., Readon, M. J., Esser, K., Schoenbechler, M. J. and Olando, L. (,1989b) Sleeping sickness in the Lambwe Valley in 1978. Annals of Tropical Medicine and Parasitology 83,21-27. (supplement # 1) Wellde, B. T., Chumo, D. A., Readon, M. J., Abinya, A., Wanyama, L., Dalo, S., Mbwabi, D., SmiJi, D. K. and Signgok, T. A. (1989c) Treatment of Rhodesian sleeping sicloiess in Kenya. Annals of Tropical Medicine and Parasitology 83,99- 109. (supplement # 1)

Wellde, B. T., Chumo, D. A., Readon, M. J., Nawiri, J., Olando, J., Wanyama, L., Awala, J., Koech, D., Siongok, T. A. and Sabwa, C. (1989e) Diagnosis of Rhodesian sleeping sickness in the Lambwe Valley (1980-1984). Annals of Tropical Medicine and Parasitology 83,63-71. (supplement A 1) Wellde, B. T., Chumo, D. A., Readon, M. J., Waema, D., Smith, D. H., Gibson, W. C., Wanyama, L. and Siongok, T. A. (1989e). Epidemiology of Rhodesian sleeping sickness in the Lambwe Valley, Kenya. Annals of Tropical Medicine and Parasitology 83,43-62. (supplement # 1)

Wells, J. M., Prospero, T. D., Jenni, L. and LePage, R. W. F. (1987) DNA contents arid molecular karotypes of hybrid Trypanosoma bruce. Molecular and Biochemical Parasitology 249,103-116. Wery, M., Wery-Paskoff, S. and Van Wettere, D. (1970) The diagnosis of African trypanosomiasis (7\ gambiense) by the use of a fluorescent antibody test. 1. Standardization of an easy technique to be used in mass surveys. Annales de la Soci6t£ Beige de Mldecine Tropicale 50,613-634. 2 8 7

Wdry, M., Mulumba, P. M., Lambert, P. H and Kazyumba, L. (1982) Hematologic manifestations, diagnosis, and immunopathology of African trypanosomiasis. Seminars in Hematology 19,83-92. Whittle, H. C., Greenwood, B. M., Bidwell, D. E., Bartlett, A. and Voller, A. (1977) IgM and antibody measurement in the diagnosis and mangement of Gambian trypanosomiasis. American Journal of Tropical Medicine and Hygiene 26,1129- 1134. Wijers., D. J. B. and Willet, K. C. (1960) Factors that may influence the injection rate of Glossina palpalis with Trypanosoma gambiense: n. The number and the morphology of the trypanosomes present in the blood of the host at the time of the infected feed. Annals of Tropical Medicine and Parasitology 54,341-350. Williams, R. O., Young, J. R. and Majiwa, P. A. O. (1982) Genomic environment of Trypanosoma brucei VSG genes: presence of a minichromosome. Nature 229,417- 421. Williams, R. O., Young, J. R. and Majiwa, P. A. O. (1979) Genomic rearrangements correlate wiih antigenic variation in Trypanosoma brucei. Nature 282,847-849. Williamson, J. (1976) Chemotherapy of African trypanosomiasis. Tropical Disease Bulletin 73, 531-542. Wilson, C. M., Serrano, A. E., Walsey, A., Bogenschutz, M. P., Shankar, A. H. and Wirth, D. F. (1989) Amplification of a gene related to mammalian mdr genes in drug- resistant Plasmodium falciparium Science. 244,1184-1185.

Wisdom, G. B. (1976) Enzyme immunoassay. Clinical Chemistry 22,1243-1255. Woo, P. T. K. (1970) The haematocrit centrifugation techniques for the diagnosis of African trypanosomiasis. Acta Tropica 27,384- 386. Wood, J. N. (1984) Subclass analysis sind purification of monoclonal antibodies. In: Methods in Molecular Biology, vol 1 (Walker, J. M., ed.), pp. 287-293, The Humana Press, New Jersey. World Health Oragnization (1976) Parallel evaluation of serological tests applied in African trypanosomiasis: a WHO collaborative study. Bulletin of the World Health Organization 5^, 141-147.

World Health Organization (1975) Bulletin of World Health Organization 52,237-242. World Health Organization (1979) The African trypanosomiases. World Health Organization Technical Report Series 635. World Health Organization (1981) Scientific working group on African trypanosomiases. Antwerp. World Health Organization (1983) Trypanosomiasis control manual. World Health Organization Experimental edition. 2 8 8

World Health Organization (1986) Epidemiology and control of African trypansomiasis. World Health Organization technical report series 739. Yamamoto, K. M., Onodera, M., Kato, K., Kakinuma, M., Kimura, T. and Richards,T7. F. (1985) Involvement of suppressoi cells induced with membrane fractions of trypanosomei in immunosuppression of trypanosomiasis. Parasite Immunology 7, 95-101. Yolken, R. H. (1982) Enzyme immunoassays for the detection of infectious antigens in body fluids: current limitations and future prospects. Review of Infectious Diseases 4, 35 -68.

Yolken, R. H., Stopa, P. J. and Harris, C. C. (1980) Enzyme immunoassay for the detection of rotaviruse and antibody. In: Manual of Clinical Immunology (Rose, N. and Friedman, H., eds.), pp. 692-699, American Society for Microbiology, Washington, D.C. Young, J. F ., Hockmeyer, W. T., Cross, M. T., Ballou, Vvr. R., Wirtz, R. A., Trosper, J. H., Beadonin, R. L., Holingdale, M. R., Miller, L. H., Diggs, C. L. and Rosenberg, M. (1985) Expression of Plasmodium falciparium sporozoite proteins in Escherichia coli for potential use in ? human malaia vaccine. Science 228,958-962. Zillman, N.and Albiez, E. J. (1986) The TESTRYP CATT (Card Agglutination test for trypanosomiasis): a field study of gambiense sleeping sickness in liberia. Tropical Medicine and Parasitology 37, 390-392.