In terms of their global health impact on children and pregnant women, as well as on adults engaged in subsistence farming, human hookworm infection (known as 'hookworm') and schistosomiasis are two of the most common and important human infections1,2. Together, their disease burdens exceed those of all other neglected tropical diseases3,4,5,6. They also trap the world's poorest people in poverty because of their deleterious effects on child development and economic productivity7,8,9. Until recently, the importance of these conditions as global health and economic problems had been underappreciated. Even the United Nations Millennium Development Goals for sustainable poverty reduction did not specifically mention these two conditions10. An important reason for this 'neglect' is that hookworm and schistosomiasis typically affect health without resulting in mortality, with infections such as HIV or malaria causing tenfold more deaths (Table 1). However, when the chronic morbidities of these two infections are fully considered according to disability-adjusted life years (DALYs; years of life lost owing to disability, ill health or death), hookworm and schistosomiasis combined rank among the most important diseases in developing countries, resulting in 4.5–92 million DALYs annually, the upper limit of which is greater than the DALYs due to malaria or HIV/AIDS4,5,6. Current efforts to control hookworm and schistosomiasis are inadequate and new tools are needed. Here we describe efforts to develop vaccines that combat hookworm and schistosomiasis, with an emphasis on disease caused by Necator americanus, the major hookworm of humans, and Schistosoma mansoni, the primary cause of intestinal schistosomiasis.

Table 1 Impact of hookworm, schistosomiasis, HIV/AIDS and malaria

Global distribution and pathobiology

Hookworms are roundworm parasites that belong to the phylum Nematoda. They share phylogenetic similarities with the free-living nematode Caenorhabditis elegans and with the parasitic nematodes Nippostrongylus brasiliensis and Heligmosomoides polygyrus, which are often used by immunologists to study T helper 2 (TH2) cell and related responses in mice11 (Box 1). Schistosomes are platyhelminths (flatworms) that belong to the order Trematoda (commonly called the trematodes or flukes). Human infections with hookworms and schistosomes occur predominantly in areas of rural poverty in sub-Saharan Africa, Southeast Asia and tropical regions of the Americas. The epidemiology and pathobiology of both hookworm and schistosomiasis have been extensively reviewed recently12,13,14,15,16,17 and are only briefly discussed here.

Hookworm. The global distribution and life cycle of hookworms are shown in Fig. 1. An estimated 600 million to 700 million people are infected worldwide, with the most infections occurring in the Asian countries of Indonesia, Bangladesh and India (60 million to 70 million people in each), followed by Nigeria and the Democratic Republic of the Congo in Africa, and Brazil (30 million to 40 million people in each)15,18,19,20. Approximately 85% of hookworm infections are caused by N. americanus, and the remainder are caused by Ancylostoma duodenale21. Microscopic infective larvae (third-stage larvae, or L3) live as non-feeding, non-replicating, developmentally arrested environmental stages in the soil, where they survive for a few days to weeks, depending on the temperature and the level of moisture. Hookworm L3 phenotypically resemble the developmentally arrested dauer larvae of C. elegans22. Human infection occurs when L3 come into contact with the skin. Larvae actively penetrate skin and migrate in the afferent vasculature to the lungs, where they ascend the pulmonary tree to the pharynx, are swallowed and moult to become adult male and female hookworms 1 cm long. Adult hookworms burrow deep into the mucosa and submucosa of the small intestine, eventually rupturing capillaries and arterioles12,13. Blood ingestion ensues, followed by the lysis of erythrocytes, an ordered enzymatic digestion of host haemoglobin23,24 and haeme detoxification25,26. Almost all of the pathology and morbidity owing to hookworm is the result of intestinal blood loss13. Female and male hookworms mate in the small intestine, and the females release microscopic eggs that exit the body in host faeces. The eggs hatch in the soil, resulting in a new generation of first-stage larvae, which feed on bacteria and other organic debris in the soil before they moult twice to the L3 stage and continue the life cycle.

Figure 1: Global distributions and life cycles of hookworms and schistosomes.
figure 1

a | The distribution of Schistosoma mansoni, which causes intestinal schistosomiasis, and Necator americanus, a hookworm, are shown13,14,18. b | The life cycles of hookworms and schistosomes.

Schistosomiasis. Approximately 90% of the world's 207 million cases of schistosomiasis occur in sub-Saharan Africa, with the most in Nigeria, Tanzania, the Democratic Republic of the Congo and Ghana17. In Africa, Schistosoma haematobium is the cause of urinary tract schistosomiasis (accounting for approximately two-thirds of the world's cases of schistosomiasis), whereas S. mansoni is the main cause of intestinal schistosomiasis (approximately one-third of total cases) (Fig. 1). S. mansoni also causes intestinal schistosomiasis in Latin America, with most of the cases occurring in Brazil, whereas Schistosoma japonicum and Schistosoma mekongi cause fewer than one million cases of intestinal schistosomiasis in Asia17. Schistosomiasis is a fresh-water-borne disease, and humans become infected when free-swimming microscopic cercariae penetrate the skin. These larvae shed their tails to become schistosomulae, which enter the vasculature and lungs before relocating to the venous system, where they become sexually mature adults that pair and mate16. Adult S. haematobium schistosomes migrate to the venous plexus, which drains the bladder and reproductive organs, whereas S. mansoni and S. japonicum go to the mesenteric veins draining the intestine16. Female schistosomes produce eggs that are equipped with a spine to facilitate penetration through blood vessels and into the urinary tract and genitals (S. haematobium) or into the intestine and liver (S. mansoni and S. japonicum). During chronic infection, which can last 5–7 years, much of the pathology from schistosome infection is a product of the immune response to parasite eggs in host tissues, and the resulting granulomatous lesions lead to fibrosis, which, in turn, can cause severe circulatory impairment of the affected organs6,16,27.

Hookworms, schistosomes and anaemia

Both hookworm and schistosomiasis cause chronic anaemia that, over the long term, can manifest as impaired neurological and cognitive functioning in children, diminished work capacity in adults, and adverse outcomes of pregnancy in both mother and child28. The WHO defines anaemia as a blood haemoglobin concentration of below 11–13 g per 100 ml, depending on age, sex and pregnancy status; severe anaemia in pregnancy is defined as a haemoglobin concentration of below 7 g per 100 ml29. Approximately 50% of anaemia cases result from iron deficiency (iron deficiency anaemia (IDA)), with nearly three-quarters of the morbidity from IDA occurring in the poorest regions of Africa, Asia and the Americas30. IDA is associated with 841,000 deaths and 35 million DALYs annually, mostly by contributing to both maternal and perinatal mortality and through adverse effects on childhood development and cognition30,31. Young children are particularly susceptible to IDA, because of their increased iron requirements during growth periods, as are women of reproductive age, because of menstrual losses and the high iron demands of a growing fetus during pregnancy28.

Hookworm-associated anaemia. IDA is the hallmark of hookworm and results from intestinal blood loss caused by the feeding of adult worms at the site of parasite attachment in the intestine13,32. Infection with 25–30 adult N. americanus hookworms results in at least 1 ml of blood loss per day, a volume containing an amount of iron roughly equivalent to the daily requirement of an adolescent boy or girl and slightly more than the daily requirement of a younger child in order for them to grow32,33. As well as causing IDA, intestinal blood loss can result in protein malnutrition12,13. Numerous epidemiological studies confirm the substantial contribution of hookworm to the global burden of IDA34. Among school-aged children in Zanzibar, Tanzania, 41% of IDA and 57% of moderate to severe anaemia is attributable to hookworm34. Hookworm has been shown to be an important risk factor for anaemia in Brazilian schoolchildren35, and it has been identified as a key determinant of IDA in preschool children in Kenya36, Tanzania37 and Malawi38. Similarly, studies have identified hookworm as an important cause of IDA in non-pregnant women in Zanzibar and Vietnam34,39, whereas it accounts for 41–54% of the moderate to severe anaemia in pregnant women in Nepal34,40. The positive association between intensity of the hookworm infection and anaemia in children and during pregnancy was confirmed in recent meta-analyses41,42. For both children and women, anaemia is far more likely to be present in those with moderate to heavy hookworm infections34,41, which are defined on the basis of quantitative faecal egg counts exceeding 1,999 eggs per gram (epg) of faeces compared with counts of individuals with no or light infection (<2,000 epg of faeces)3.

Schistosomiasis-associated anaemia. All of the major forms of human schistosomiasis are associated with anaemia4,6,43,44,45,46,47,48,49,50. As found for hookworm, children and pregnant women infected with schistosomes are especially susceptible to anaemia43,44,45,46,47,48,49,50,51,52,53, and reduced haemoglobin concentrations in both have been associated with moderate to high faecal or urine schistosome egg counts28,44,45,47,48,49,51. The anaemia associated with schistosomiasis has been attributed to several mechanisms, including iron deficiency due to blood loss in the intestine or urine, splenic sequestration and destruction of erythrocytes, autoimmune haemolysis, and the chronic inflammatory response to schistosome eggs deposited in host tissues28,46.

Co-infections. In Africa and South America, co-infections with hookworm and schistosomes are common, and there are at least a dozen countries with more than five million cases of each helminth infection2. It has been proposed that co-infection with N. americanus and S. mansoni is synergistic with respect to worm burdens and the resulting pathologic sequelae, including anaemia9,35,54,55,56,57,58,59,60,61,62. A similar relationship has been suggested between hookworm and S. japonicum infection in East Asia62. In sub-Saharan Africa, there is also extensive geographical overlap among areas of hookworm, schistosomiasis, and malaria transmission resulting from infection with Plasmodium falciparum, another notable cause of anaemia63,64. In Kenya and Tanzania, the anaemias resulting from hookworm and P. falciparum co-infections have been shown to be additive63. There are conflicting data on whether hookworm or schistosomiasis increase host susceptibility to malaria or adversely affect the clinical course of the disease5. In Africa, female genital schistosomiasis caused by S. haematobium was shown to increase the odds ratio of acquiring HIV/AIDS threefold65, and it has been suggested that S. mansoni may also affect susceptibility to HIV66,67. Hookworm and other intestinal nematode infections are immunomodulatory and may increase viral loads and the progression of HIV/AIDS68, but larger studies are needed to confirm this relationship. Anaemia itself may adversely affect the course of HIV/AIDS69, so hookworm and schistosomiasis may indirectly impact the progression of this disease. The global public health impact of anaemia and other pathological sequelae resulting from hookworm and schistosomiasis have stimulated efforts to develop more effective control strategies for these conditions. Such measures include the development of new control tools, foremost among these being anthelmintic vaccines. The current status of the development of vaccines is outlined below, first for hookworm and then for schistosomiasis.

Rationale for a human hookworm vaccine

Hookworm and other common intestinal helminth infections such as ascariasis and trichuriasis are strongly associated with poverty and poor sanitation. However, improvements in sanitation or other environmental and personal protective control measures (such as footwear) frequently have a minimal impact on parasite prevalence or the intensity of infection, especially without accompanying programmes of health education and economic development70,71. The WHO has proposed annual mass treatment (that is, deworming of an entire population or age stratum in an endemic area) with a benzimidazole such as albendazole or mebendazole as the most cost-effective means of reducing the childhood morbidity related to chronic intestinal helminth infection (although the impact on improving childhood cognition is variable)42,72,73,74,75. There is evidence that both albendazole and mebendazole interfere with invertebrate tubulin and microtubules and reduce the number of adult worms in the intestine76. In 2001, the 54th World Health Assembly committed to providing annual deworming for school-aged children wherever the prevalence in this age group exceeds 50%42,72,73,74,75, although as of 2008 only 9% of school-aged children and 21% of preschool-aged children at risk of acquiring intestinal helminth infections have benefited from this intervention73,75.

New information indicates that annual deworming may be less effective for hookworm than other intestinal helminth infections. For example, single-dose albendazole or mebendazole typically achieves either cure or substantial reductions in worm burdens and faecal egg counts for ascariasis75,77. However, for hookworm high rates of drug failure have been reported for mebendazole, with an average cure rate of only 15%77. Furthermore, after repeated administration in the same population, the efficacy of mebendazole has been reported to diminish over time78, raising concerns about possible drug resistance. Indeed, drug failures have been shown to occur with benzimidazoles when used ubiquitously in livestock and have been associated with specific point mutations in the parasite gene encoding β-tubulin79. Although the same mutations have not yet been associated with drug failure in humans, efforts are underway to determine whether benzimidazole failures for hookworm and other helminth infections result from similar resistance mechanisms79.

Evidence of widespread mebendazole drug failure indicates that the global control of human hookworm depends solely on the continued efficacy of albendazole. However, although treatment with albendazole cures existing infections, it does not confer protection from reinfection, which often occurs as early as 6 months after treatment in areas of high transmission80. Concerns about mebendazole drug failure, possible emerging resistance to the benzimidazoles and the rapid reinfection after treatment provided the impetus for establishing the Human Hookworm Vaccine Initiative, a non-profit product development partnership based at the Sabin Vaccine Institute (Washington DC, USA) that was established in 2000 to develop and test new vaccines for hookworm1,2,12,81,82,83,84,85,86. Prior efforts to develop human hookworm vaccines were limited to basic research conducted in university laboratories, although a live attenuated canine hookworm vaccine was developed by industry and briefly marketed for pet owners in the early 1970s82,83,87. The HHVI is the only partnership in the world that is currently working on vaccine development for hookworm and consists of investigators from the Sabin Vaccine Institute, the George Washington University (Washington DC, USA), The Fundação Oswaldo Cruz (FIOCRUZ; Brazil), the Instituto Butantan (São Paulo, Brazil), James Cook University (Cairns, Australia) and the London School of Hygiene and Tropical Medicine (UK). The ultimate aim of vaccine development efforts is to prevent moderate and heavy hookworm infections (that is, infections associated with faecal egg counts exceeding 1,999 epg of faeces), which are associated with substantial intestinal blood loss. Such a vaccine could be administered to very young preschool-aged children in a programme of 'vaccine-linked chemotherapy' (Ref. 88) before their exposure to infective larvae in the environment, or to both preschool-aged and school-aged children who may have already been exposed and even infected.

Targeting hookworm blood feeding

The vaccines that are currently under development by the HHVI target the nutritional and metabolic requirements of the adult hookworm. The main approach has been to identify the essential components involved in parasite blood feeding, to genetically engineer these components as recombinant proteins and then to combine the recombinant components with one or more adjuvants to elicit protective antibodies on vaccination82. These protective antibodies would either directly neutralize the parasite macromolecules required for blood feeding and nutrition or indirectly damage important parasite structures. Protective immunity in vaccinated individuals would manifest as diminished hookworm-related blood loss and reduced numbers of hookworms in the intestine compared with levels in unvaccinated people. Based on prior experiences with live attenuated helminth vaccines for veterinary use, including those for canine hookworm and bovine lungworm, sterilizing immunity is not considered an attainable — or necessary — goal for anthelmintic vaccines83,87,89 (Table 2). However, as the morbidity associated with hookworm is proportional to the number of worms harboured by individuals, a vaccine that prevents most moderate and heavy infections would be sufficient to have a major impact on the worldwide burden of hookworm and the associated anaemia.

Table 2 Successful* vaccines against helminth infections

Over the past decade, the molecular basis by which hookworms ingest and derive nutrition from host blood has been elucidated. As with all haematophagous parasites, N. americanus depends on host haemoglobin and serum proteins for survival. Adult hookworms ingest blood, lyse erythrocytes, degrade haemoglobin and serum proteins, and then absorb the digested peptides and amino acids23,24 (Fig. 2). Following haemolysis, adult N. americanus hookworms use a hierarchical cascade of haemoglobinases (haemoglobin-degrading proteases), beginning with the cleavage of intact haemoglobin by an aspartic protease, APR1, followed by further proteolysis through the action of several cysteine proteases and metalloproteinases, all of which are expressed in the brush border membrane of the parasite's digestive tract24. The resultant small peptides and free amino acids are possibly absorbed through the parasite gut through a homologue of the membrane-spanning amino acid transporter of the free-living nematode C. elegans90. After their cleavage from digested globin, both iron-containing haeme and iron-containing haematin are potentially toxic to hookworms, because these compounds can generate oxygen radicals that may damage parasite structures91. Therefore, in addition to haemoglobinases, all blood-feeding parasites such as hookworms and P. falciparum have evolved mechanisms to detoxify and transport haeme25,92,93,94. In the case of N. americanus and other hookworms, one putative mechanism for neutralizing these toxic moieties involves the pairing of glutathione S-transferase 1 (GST1) molecules as homodimers to create specific pockets capable of binding haeme and haematin25,92,95 (Fig. 2).

Figure 2: Necator americanus degradation of host blood components and potential vaccine targets.
figure 2

Adult worms in the gut ingest blood, and parasite haemolysins drill pores into the erythrocytes, releasing haemoglobin into the parasite gut lumen (step 1). Haemoglobin is digested by the hierarchical and ordered cascade of haemoglobinases (APR1, an aspartic protease, CP3, a cysteine protease, and MEP1, a metalloproteinase) lining the brush border membrane of the parasite gut (step 2). The globin peptides and free amino acids that are released following haemoglobin digestion are absorbed into the gut cells, putatively being transported by OPT1 (step 3). Free haeme is detoxified by the action of glutathione S-transferase (GST) (step 4). Antibodies that could be induced by vaccination to neutralize the function of target proteins and interrupt blood feeding are shown. Question marks indicate steps that have not been experimentally confirmed. GSH, glutathione; GSSH, glutathione disulphide.

Antigens of the human hookworm vaccine

From approximately two-dozen proteins that are putatively involved in the hookworm blood-feeding process12,82, two lead candidate antigens have been selected for clinical development. The antigen selection programme of the HHVI is based on a ranking system that includes several key criteria, such as efficacy in animal trials, immuno-epidemiological observations in individuals resident in endemic areas, and the feasibility of protein expression and scaled-up manufacture using low-cost expression systems such as yeast, bacteria, or plants82 (Table 3). In addition, antigens are prioritized if there is a plausible mechanism of protection associated with them, such as triggering the production of antibodies that inhibit crucial parasite enzymes or target important surface antigens. Both GST1 and APR1 are involved in parasite blood feeding, and it is thought that each antigen induces antibodies that interfere with the function of the respective protein and impair worm survival. Both antigens are therefore being considered for eventual combination in a human hookworm vaccine.

Table 3 Ranking of lead candidate Necator americanus vaccine antigens

GST1 is a 24 kDa polypeptide that is expressed as a recombinant protein in the yeast Pichia pastoris to generate the antigen used for vaccines. Both GST1 from N. americanus and its orthologue from the canine hookworm Ancylostoma caninum have peroxidase activity that catalyses the conjugation of reduced glutathione to various electrophiles25,92,95. Both proteins belong to the Nu class of nematode GSTs, which is characterized by a reduced peroxidase activity relative to other classes of GSTs but an increased binding capacity for haeme and related products25,92,95,96,97. According to X-ray crystallography studies, N. americanus GST1 forms homodimers in solution to create atypically large binding cavities that are accessible to a diversity of ligands, including haeme95, to which GST1 from both A. caninum and N. americanus binds with high affinity in vitro25,92. Both haeme and haematin contain oxidative iron that can result in the formation of oxygen radicals, which damage helminth structures. In vivo, GSTs may protect hookworms by binding and detoxifying haeme and the haematin by-products that are generated during the blood digestion process25,92.

On the basis of their putative roles in hookworm blood feeding, GST1 from both N. americanus and A. caninum were tested in laboratory animal models of hookworm. In dogs, vaccination with recombinant A. caninum GST1 resulted in high levels of antibodies; following challenge with infective A. caninum larvae, the worm burdens and faecal egg counts of these vaccinated dogs were substantially lower than those observed in controls25. In hamsters, vaccination with recombinant A. caninum GST1 also resulted in cross-protection, with substantially lower worm burdens (less than half) following heterologous challenge with infective N. americanus larvae than those seen in controls25,98, as did vaccination with N. americanus GST1 followed by homologous larval challenge92. A recombinant GST from the nematode parasite Wuchereria bancrofti is also showing promise in a jird model as a protective antigen against lymphatic filariasis99. Because of these encouraging results, recombinant N. americanus GST1 (formulated with Alhydrogel (aluminium hydroxide)) has been produced according to current good manufacturing practices in preparation for clinical trials.

N. americanus APR1 is a 45 kDa protein. For stability and safety reasons, including concerns about injecting humans with an active proteolytic enzyme, the version of this aspartic protease used for vaccination has been inactivated by substituting alanines for the catalytic aspartic acid residues100. The recombinant protein has been expressed in multiple systems, with Escherichia coli100 and tobacco plants101 producing the highest yields. The rationale for selecting N. americanus APR1 for development was based on successful laboratory animal trials. In dogs vaccinated with either N. americanus APR1 or A. caninum APR1, high levels of antibody were induced that inhibited protease activity in vitro; this was associated with substantially diminished blood loss (measured by haemoglobin levels) and worm burdens following challenge with A. caninum larvae compared with those seen in controls100,102. Vaccination with A. caninum APR1 also resulted in a substantial reduction in the worm burdens of hamsters challenged with N. americanus compared with burdens of controls98. These findings also suggest that cross-reactive immunity between Necator spp. and Ancylostoma spp. hookworms may occur. Following vaccination, APR1-specific antibodies are ingested by the parasite during blood feeding and localize to the parasite gut, where they can inhibit parasite feeding by neutralizing enzyme activity98,102 (Fig. 2). Efforts to optimize both the yield and the solubility of N. americanus APR1 are in progress. A chimeric protein consisting of N. americanus GST1 fused to epitopes of N. americanus APR1, to generate neutralizing antibodies and inhibit parasite blood digestion, is already under development103. Additional key molecules involved in hookworm blood feeding have been identified from proteomic and transcriptomic analyses of the hookworm gut104. These molecules include a putative hookworm orthologue of a prolylcarboxypeptidase ('contortin') that protects sheep against Haemonchus contortus infection105, and the extracellular domain of an intestinal peptide transporter that is essential for nutrient uptake and growth in C. elegans90.

Previous vaccine antigens. ASP2 from N. americanus infective larvae was previously under consideration as a candidate vaccine antigen106,107,108,109. The canine hookworm orthologue of this molecule was determined to be a major immunogen associated with an effective vaccine consisting of live attenuated A. caninum L3 larvae106,109. In a Phase I trial conducted in healthy volunteers in the United States, ASP2 adjuvanted with Alhydrogel was shown to be safe and immunogenic107. However, in a second Phase I trial conducted at a hookworm endemic site in Brazil, some of the adult volunteers experienced generalized urticaria immediately after vaccination (D.J.D., unpublished observations). The study was halted, and it was found that individuals who developed urticaria had high levels of immunoglobulin E (IgE) against ASP2. This finding has led to testing for the levels of IgE specific for candidate vaccine antigens using sera from individuals resident in hookworm-endemic areas. No detectable levels of IgE specific to N. americanus GST1 (J.M.B. and D.J.D., unpublished observations) or N. americanus APR1 (Ref. 100) have been found in individuals living in hookworm-endemic areas of Brazil, thus permitting their continued development as candidate vaccine antigens. The reason neither recombinant adult hookworm protein induces IgE during natural infection is unknown, but it may be related to antigen structure or presentation to the immune system.

Vaccinating against schistosomiasis

In both Brazil and most of sub-Saharan Africa, N. americanus hookworm infections are co-endemic with intestinal schistosomiasis caused by S. mansoni2. Vaccines to combat each of these helminth infections are being developed because both diseases are associated with anaemia and malnutrition, especially in children. Ultimately, the two vaccines may be co-administered or combined in a multivalent anthelmintic vaccine (see below)2. The justification for developing vaccines against schistosomiasis has been reviewed recently88,110 and includes the high disease burden4,6, the high rates of post-treatment reinfection, the inability of chemotherapy-based morbidity control to interrupt transmission111, the exclusive reliance on praziquantel for control and concerns about emerging drug resistance without new drugs in the development pipeline88,110. An important additional stimulus to develop new preventive approaches to schistosomiasis is the observation of so-called 'rebound morbidity': up to 80% of children living in high-transmission areas can suffer recurrent aggressive inflammation following interrupted annual chemotherapy because of reinfection. The feasibility of developing vaccines for schistosomiasis has been extensively reviewed88,110,112. Humans living in endemic areas can become resistant or partially immune to reinfection over time113. Furthermore, irradiated cercariae can elicit high levels of protective immunity in laboratory animals, and several recombinant-protein vaccines have been shown to elicit comparable levels of protective immunity in immunized animals that were subsequently challenged with cercariae110.

Schistosome antigens under development

To date, one vaccine for urinary schistosomiasis has entered clinical trials. The Institut Pasteur and the French Institut National de la Santé et de la Recherche Médicale have taken a recombinant 28 kDa GST cloned from S. haematobium through both Phase I and Phase II clinical trials in Europe and West Africa110,114 (see also Le Project Bilhvax 3). Known as Bilhvax, this GST formulated with an aluminium hydroxide adjuvant has been reported to be immunogenic and safe after testing in healthy adults114. However, further information regarding its efficacy, the duration of protection and the progress towards licensure are not available in the published literature. In addition, other vaccine candidates for intestinal schistosomiasis caused by S. mansoni will soon be ready for clinical testing112. One candidate is a 14 kDa fatty acid-binding protein known as Sm14 (Ref. 115), which in experimental animals (mice and rabbits) elicits protection against S. mansoni as well as against Fasciola hepatica, another trematode fluke116. Recently, the group developing this vaccine reported success in stabilizing Sm14 by replacing a crucial cysteine residue in order to prevent dimerization117. Recombinant Sm14 is being developed as an anthelmintic vaccine by a partnership between private and government organizations in Brazil for use against both fascioliasis of livestock and human schistosomiasis caused by S. mansoni116. Another S. mansoni vaccine potentially moving into clinical development is Sm-p80, which is a DNA vaccine encoding the large subunit of a calcium-dependent neutral protease, providing levels of protection in baboons that are comparable to irradiated cercariae118. Finally, the schistosome molecule paramyosin is undergoing pilot-scale production in Asia for use against S. japonicum infection, possibly as a transmission-blocking vaccine, administered to water buffaloes119.

The Sabin Vaccine Institute, in partnership with FIOCRUZ and the Instituto Butantan, is also working to transition a S. mansoni vaccine into clinical testing in Brazil (J.M.B., D.J.D. and P.J.H., unpublished observations). The primary targets of this vaccine development programme are schistosome membrane proteins identified by combined genomic, post-genomic and proteomic analyses of the adult S. mansoni outer surface, or tegument120 (Table 4). The tegument of adult schistosomes is a single syncytium covering the body wall and is thought to be a dynamic layer involved in several key physiological processes, including parasite nutrition, osmoregulation and evasion of host immunity120. Hence, the schistosome tegument is a potentially vulnerable target for immunological attack by host antibodies. However, analysis of the schistosome proteome predicts that surprisingly few membrane-spanning proteins of the tegument are accessible to the host immune response120,121. They include a family of tetraspanin integral membrane proteins122 and several outer-membrane proteins of unknown function, such as Sm29 (Refs 123, 124). The tetraspanins are so named because they contain four transmembrane domains, with two extracellular loops that are predicted to interact with exogenous proteins or ligands (Fig. 3). The second extracellular domain fragment of a schistosome tetraspanin known as TSP2, from S. mansoni, has been selected for development as a human vaccine antigen. When this 9 kDa extracellular domain was expressed in either P. pastoris or E. coli and formulated with several adjuvants (including Freund's complete adjuvant122, aluminium hydroxide, or aluminium hydroxide with CpGs) it provided high levels of protection in mice vaccinated with the antigen and then challenged with S. mansoni cercariae (A.L. and M.S.P., unpublished observations). In addition, evidence from human epidemiological studies indicates that putatively resistant individuals living in endemic areas of Brazil have elevated antibody responses to this protein compared with the responses of chronically infected individuals from the same endemic areas122. Recently, the S. japonicum orthologue of S. mansoni TSP2 was described and resulted in protection in mice that was similar to that described for S. mansoni TSP2, suggesting that this molecule may be effective against multiple human schistosome species125.

Table 4 Ranking of lead candidate Schistosoma mansoni vaccine antigens
Figure 3: Schistosoma mansoni tegument.
figure 3

a | A fluorescence micrograph of the tegument of an adult male Schistosoma mansoni probed with a mouse antibody raised to recombinant TSP2 (red), a tetraspanin. Nuclei, stained with 4′,6-diamidino-2-phenylindole (DAPI), are blue. b | A schematic representation of S. mansoni TSP2 in the tegument plasma membrane. Extracellular loops (ECs) are indicated, and cysteine residues are shown (the lines between them denote the disulphide bond pairing); transmembrane domains (TMs) are shown numbered from amino terminus to carboxyl terminus. c | The tegument of an S. mansoni schistosomula incubated for 7 days with S. mansoni tsp2 (left) or luciferase (control; right) double-stranded RNAs. Digitate extensions (arrows) are more abundant on the surface of the tegument incubated with tsp2 double-stranded RNA. Mus, muscle; Som, schistosomula; Teg, surface layer of the tegument. Part a image is reproduced, with permission, from Ref. 120 © (2007) Elsevier. Part c image is reproduced from Ref. 126.

S. mansoni TSP2 is thought to have a crucial role in tegument development and maturation126. The ultrastructural morphology of adult worms and schistosomula treated in vitro with S. mansoni tsp2 double-stranded RNA (dsRNA) displays a distinctly vacuolated and thinner tegument compared with that of controls, suggestive of impaired closure of tegumentary invaginations126. Moreover, injection of mice with schistosomulae that had been pre-treated with S. mansoni tsp2 dsRNA resulted in 83% fewer parasites being recovered from the mesenteric veins 4 weeks later when compared with recovery from mice injected with untreated schistosomulae126. These results suggest that tetraspanins have important structural roles in tegument development, maturation or stability. Other tegument tetraspanins are attractive vaccine candidates; for example, S. mansoni tsp3 is the most highly upregulated mRNA in maturing schistosomula, a developmental stage that is widely accepted as being susceptible to damage by the human immune system127,128. In addition, Sj23 (23 kDa integral membrane protein of S. japonicum) is a tegument tetraspanin that is showing promise as a DNA vaccine for water buffaloes, which are an important reservoir host for S. japonicum in China129.

Future directions

By targeting both hookworms and schistosomes, human helminth vaccines are being developed to reduce parasite-induced morbidity, the symptoms of which include intestinal blood loss and inflammation2. Administered in early childhood, such vaccines could prevent the major paediatric sequelae of these infections, including anaemia, malnutrition, growth failure and impaired cognitive development.

The human hookworm vaccine is being developed as a bivalent product consisting of two co-formulated recombinant proteins (GST1 and APR1 from N. americanus) to prevent moderate and heavy hookworm infections caused by N. americanus. The vaccine is intended primarily for preschool- and school-aged children (<10 years of age) living in N. americanus-endemic regions. Similarly, TSP2 from S. mansoni is being developed as a recombinant-protein vaccine for the prevention of heavy-intensity infections of S. mansoni, the leading cause of intestinal schistosomiasis. A paediatric population will be targeted for both vaccines because this age group is at the greatest risk of developing the severe developmental, growth and cognitive impairments associated with these chronic infections1,12,13,14,15. Initially, vaccines containing the hookworm and schistosome antigens are being formulated with Alhydrogel; however, they will also be evaluated with an additional immunostimulant such as a lipid A derivative. The vaccines will be delivered by intramuscular injection, and the goal is to achieve the desired protection after one or two doses, depending on the number of doses required to achieve a protective response. The desired result is the prevention of moderate and heavy helminth infections, which would have a major impact on the anaemia, malnutrition and end-organ pathology associated with these parasitic infections in children6,34. The extension of protection into adulthood would also prevent the severe anaemia that is related to infection with these parasites during pregnancy, and reduce transmission. Such vaccines may also have an important impact on poverty reduction because of their anticipated effect on improving child and maternal health and development81.

Both vaccines are being developed with the ultimate goal that even the most impoverished populations will have access to them as soon as they are available. As such, a strategic road map is being followed to ensure that low-cost manufacturing processes are used and that vaccine manufacturers in middle-income disease-endemic countries are involved from the start. In the Americas, Brazil is the furthest advanced, with two major vaccine manufacturers — FIOCRUZ/Bio-Manguinhos and the Instituto Butantan — actively engaged in development130. Accurate forecasting of the eventual demand for licensed vaccines is essential: for hookworm, it is estimated that there are approximately one billion children at risk globally, so that covering a global birth cohort would require the vaccination of 100 million children annually73. Demand forecasting is underway for an intestinal schistosomiasis vaccine.

Substantial hurdles must be overcome during clinical development of hookworm and schistosomiasis vaccines, not least of which is securing adequate funding to conduct the clinical trials required for licensure. Additional obstacles include obtaining access to the novel adjuvants that may be required to induce an adequate immune response and the difficulty of conducting large-scale efficacy studies in endemic areas. As hookworm and schistosomiasis are prevalent in resource-limited, rural areas of the tropics and subtropics, this is where Phase III clinical trials must be conducted, which can be logistically challenging. Furthermore, because the clinical effects of these parasitic infections are chronic, with sequelae such as IDA often appearing only after months or years of infection, efficacy trials will be necessarily long.

Combining hookworm and schistosomiasis vaccines. In the future, vaccines for hookworm and intestinal schistosomiasis could be combined in a multivalent anthelmintic vaccine2, which may increase vaccine efficiency and reduce the timeframe for widespread distribution in affected areas of Africa and Latin America. The Sabin Vaccine Institute is currently working with the WHO to build consensus on the use of hookworm and schistosomiasis vaccines in resource-poor settings131. Health systems established through the integrated control of mass drug administration for neglected tropical diseases and expanded global deworming efforts for preschool- and school-aged children8,10 could ultimately provide an infrastructure for linking newly developed vaccines with anthelmintic chemotherapy.