**The Impact of Immune Responses on the Asexual Erythrocytic Stages of** *Plasmodium* **and the Implication for Vaccine Development**

Elke S. Bergmann-Leitner, Elizabeth H. Duncan and Evelina Angov *Department of Molecular Parasitology, Division of Malaria Vaccine Development, Walter Reed Army Institute of Research, USA* 

#### **1. Introduction**

200 Malaria Parasites

Wambua, S, Mwangi, TW, Kortok, M, Uyoga, SM, Macharia, AW, Mwacharo, JK,

Weatherall, DJ (1997). Thalassaemia and malaria, revisited. *Ann. Trop. Med. Parasitol.* 91(7):

Weiss, W, Sedegah, M, Beaudoin, R, Miller, L and Good, M (1988). CD8+ T cells

Williams, TN, Mwangi, TW, Wambua, S, Alexander, ND, Kortok, M, Snow, RW and Marsh,

Williams, TN, Mwangi, TW, Wambua, S, Peto, TE, Weatherall, DJ, Gupta, S, Recker, M,

Williams, TN, Wambua, S, Uyoga, S, Macharia, A, Mwacharo, JK, Newton, CR and

Wilson, DW, Crabb, BS and Beeson, JG (2010). Development of fluorescent *Plasmodium* 

Winkler, S, Willheim, M, Baier, K, Schmid, D, Aichelburg, A, Graninger, W and Kremsner,

Winkler, S, Willheim, M, Baier, K, Schmid, D, Aichelburg, A, Graninger, W and Kremsner,

groups with *Plasmodium falciparum* malaria. *J. Infect. Dis.* 179(1): 209-216. Yuthavong, Y, Bunyaratvej, A and Kamchonwongpaisan, S (1990). Increased susceptibility

thalassemia and the sickle cell trait. *Nat Genet* 37(11): 1253-1257

*falciparum* for in vitro growth inhibition assays. *Malar J* 9: 152

malaria sporozoites. *Proc. Natl. Acad. Sci. USA. 1988 Jan; 85(2): 573-6*  Williams, TN (2009). Human genetic resistance to malaria. *Adv Exp Med Biol* 634: 243-253 Williams, TN, Mwangi, TW, Roberts, DJ, Alexander, ND, Weatherall, DJ, Wambua, S,

protection by the sickle cell trait. *PLoS Med* 2(5): e128

childhood diseases. *J Infect Dis* 192(1): 178-186

Kenya. *Blood* 106(1): 368-371

6040-6044.

*Blood* 16: 602-604

the coast of Kenya. *PLoS Med* 3(5): e158

885-890.

Weatherall, DJ, Snow, RW, Marsh, K and Williams, TN (2006). The effect of alpha+ thalassaemia on the incidence of malaria and other diseases in children living on

(cytotoxic/suppressors) are required for protection in mice immunized with

Kortok, M, Snow, RW and Marsh, K (2005a). An immune basis for malaria

K (2005b). Sickle cell trait and the risk of *Plasmodium falciparum* malaria and other

Penman, BS, Uyoga, S, Macharia, A, Mwacharo, JK, Snow, RW and Marsh, K (2005c). Negative epistasis between the malaria-protective effects of alpha+-

Maitland, K (2005d). Both heterozygous and homozygous alpha+ thalassemias protect against severe and fatal *Plasmodium falciparum* malaria on the coast of

PG (1998). Reciprocal regulation of Th1- and Th2-cytokine-producing T cells during clearance of parasitemia in *Plasmodium falciparum* malaria. *Infect. Immun.* 66(12):

PG (1999). Frequency of cytokine-producing T cells in patients of different age

of malaria-infected variant erythrocytes to the mononuclear phagocyte system.

Natural immunity against malaria is acquired after repeated infections for an extended period of time resulting in a state of immunological non-responsiveness against the malaria parasite. This state ultimately prevents the onset of severe disease such as cerebral malaria thereby reducing the risk of death from malaria. Individuals with acquired natural immunity to malaria still harbor parasites (albeit in low densities) in the blood, and therefore natural immunity against malaria is not sterile. For this reason, natural immunity to malaria cannot be compared to immunity achieved against other diseases where the immune response neutralizes and eliminates the pathogen. The second hallmark of natural immunity to malaria is that protection wanes once a "protected" individual leaves the malaria-endemic area indicating that protection depends on continued antigen exposure. Immunity to malaria is stage- and species-specific and distinct immune mechanisms confer protection against the different developmental stages of the parasite. In the case of blood stage infection, passive transfer experiments with purified human immunoglobulins derived from immune individuals living in malaria-endemic areas have demonstrated that antibodies can mediate protection (Cohen et al., 1961; Butcher et al., 1970; Bouharoun-Tayoun et al., 1990). Mechanistic studies revealed that the effect of blood stage-specific antibodies on the asexual erythrocytic parasite depends on their antigen-specificity; antibodies can bind to merozoites, opsonize and target them towards phagocytic cells of the host (Groux and Gysin, 1990), or prevent invasion of new erythrocytes (Perkins, 1991). Once infected, antibodies against asexual blood stage antigens such as Pf332 or MSP-1 inhibit the intra-erythrocytic development of *Plasmodium falciparum* (Ahlborg et al., 1996; Siddique et al., 1998; Bergmann-Leitner et al., 2009). Antibodies directed to antigens expressed by sexual erythrocytic stages (gametocytes) have been shown to prevent transmission of malaria by blocking either the infection of the mosquito or the development in the mosquito (Lavazec and Bourqouin, 2008). As will be outlined in this chapter, it becomes increasingly clear that a blood stage vaccine may never be able to induce sterile protection, but can prevent mortality. Why bother developing blood stage vaccines? Extensive studies characterizing

The Impact of Immune Responses on the Asexual

methods as well as their predictive value.

**specific antibodies** 

to significant artifacts.

**2.1.1 Microscopic analysis** 

Erythrocytic Stages of *Plasmodium* and the Implication for Vaccine Development 203

hours. Whether this will be achievable to prevent morbidity or whether blood stage vaccineinduced antibodies will only be able to prevent severe disease and mortality is currently the focus of erythrocytic vaccines. Ideally, *in vitro* methods and/or preclinical models would assist in the down-selection of vaccine candidates. However, currently there is no preclinical model that adequately simulates human malaria. In contrast, there are several *in vitro* methods that have been used for analyzing functional antibodies against blood stage parasites *in vitro*. The following methods have been employed extensively in the evaluation of blood stage vaccines and we outline the advantages and drawbacks of the various

**2.1 Functional assays capturing the biological activity of blood stage, parasite-**

Various methods have been developed that allow the measurement of the biological functionality displayed by either the antibodies themselves (growth inhibition assays (GIA) and invasion inhibition assays (IIA)) or antibodies in collaboration with immune cells expressing Fc-receptors (receptors that naturally bind certain immunoglobulin isotypes in order to eliminate antigens bound by antibodies). Antibodies bind their specific pathogen and form "immune complexes". Depending on the class and isotype of the antibody in the immune complex, this interaction leads to different outcomes: (1) the immune complex activates the complement system, *i.e.* complement fixation (which ultimately results in the lysis of the attached pathogen) or (2) the immune complex binds to Fc-receptors, *i.e.* opsonization (mediating uptake of the bound pathogen by the immune cells) or (3) antibody binds to the antigen and inactivates (neutralizes) the pathogen and thus prevents infection. The functional analysis of anti-blood stage antibodies, especially when obtained from individuals in endemic areas, is complicated by the presence of toxic factors or lipids in sera or plasma (such as anti-malarial drugs and anti-coagulants, or oxidized lipids due to long or inadequate sample storage). Eliminating nonspecific toxicities requires purification of the antibodies from serum or plasma using either bulk enrichment techniques (such as precipitation with ammonium sulfate and caprylic acid) or immunoglobulin purification methods (*e.g*., the use of Protein A, G , L-columns). It should be noted that most anti-malarial drugs can be removed by simply dialyzing the samples (Sy et al., 1990; Persson et al., 2006) which allows the testing of samples where only small volumes are available as is often the case for pediatric specimen. When comparing various methods in regards to yield, purity, integrity and retention of functional activity of recovered antibodies, we found that some methods lead to the selective enrichment of certain isotypes or in some cases fairly unstable immunoglobulin preparations (Bergmann-Leitner et al., 2008b). Also, not all purification methods perform equally well when purifying immunoglobulins from different species. The choice of purification method greatly influences the results obtained in the functional assay and can lead

Microscopic evaluation of blood smears stained with Giemsa is the classic method for detecting parasitized erythrocytes (pRBC) and remains the gold standard in clinical trials. It is also used for studying the effect of antibodies on the morphology of the intra-erythrocytic parasite. To detect parasitemia in the blood, thick film-smears are prepared and 100

the leading pre-erythrocytic vaccine RTS,S have shown approximately 40% sterile protection of vaccinated US individuals (Kester et al., 2007). Protection in the field is restricted to reduced mortality and morbidity and requires vaccination of young children that have continued exposure to the parasite. Under these circumstances, partial protection (when defined as time to first clinical episode) was approx 30% (Sacarlal et al., 2009). The limited efficacy of pre-erythrocytic vaccines appears to be the result of the fact that even a few sporozoites which escape the vaccine-induced immune response are still capable of establishing full blown blood stage infection. Combining a pre-erythrocytic vaccine such as RTS,S with a blood stage vaccine would assure that individuals, who did not develop sterile immunity (sterile immunity has never been achieved in field trials) would have a reduced risk of dying as a result of the added blood stage component.

This chapter will highlight the major readout methods that are currently used to gauge the efficacy of blood stage vaccines. In the absence of 'protection' models, the study of blood stage vaccines requires the development of *in vitro* assays that predict vaccine efficacy. In the absence of a definitive correlate of protection, the predictive value of these assays still awaits validation. Various target antigens for vaccine development will be highlighted with a special focus on the Merozoite Surface Protein (MSP)-1. Lastly, we will discuss modes of action of immune effector mechanisms against blood stage parasites and provide a preview of next-generation malaria vaccines.

#### **2. Methodological tools to study anti-parasite activities mediated by**  *Plasmodium***-specific antibodies**

The ability to culture blood stage parasites of *P. falciparum* has greatly assisted in our understanding of events associated with maturation of the intra-erythrocytic parasite (Trager and Jensen, 1976; Haynes and Moch, 2002; Haynes et al., 2002). This is underscored by the lack of knowledge for other *Plasmodium* species that have not been successfully established in culture. Once the liver schizonts have reached maturity, the hepatocyte membrane dissolves and the merosome, *i.e*., the parasitophorous vacuole which resembles a bag containing infectious merozoites, is released from the liver destined for the lung capillaries where infection of erythrocytes occurs (Baer et al., 2007). The erythrocytic life cycle of *P. falciparum* blood stage parasites ranges from 38-48 hrs. During the first 24 hrs the parasite has a ring-like morphology (ring stage parasite) which subsequently develops into a trophozoite that is clearly distinguishable by the increasing amount of DNA and cytoplasm. Further maturation of the parasite into the schizont stage is marked by DNA segmentation resulting in individual nuclei. Additionally, schizonts have increasing amounts of digested hemoglobin (*aka*, malaria pigment or hemozoin). The schizonts rupture after first dissolving the parasitophorous vacuole and then the erythrocyte membrane, thus releasing 16-32 infectious merozoites per new infected erythrocyte (Cowman and Crabb, 2006). This event typically occurs in post capillary venules where the flow rate of the blood is low thereby favoring the rapid invasion of erythrocytes. Free merozoites have a relatively short lifespan and thus it is thought that the invasion event occurs quickly (within 5 minutes). This poses a significant challenge for effector antibodies which have to engage the parasite during this brief period of vulnerability and to block the interactions between host cells and merozoites. Successful active or passive immunotherapy requires providing an ample supply of antigenspecific, high-affinity antibodies capable of blocking a 'tsunami' of parasites every 38-48

the leading pre-erythrocytic vaccine RTS,S have shown approximately 40% sterile protection of vaccinated US individuals (Kester et al., 2007). Protection in the field is restricted to reduced mortality and morbidity and requires vaccination of young children that have continued exposure to the parasite. Under these circumstances, partial protection (when defined as time to first clinical episode) was approx 30% (Sacarlal et al., 2009). The limited efficacy of pre-erythrocytic vaccines appears to be the result of the fact that even a few sporozoites which escape the vaccine-induced immune response are still capable of establishing full blown blood stage infection. Combining a pre-erythrocytic vaccine such as RTS,S with a blood stage vaccine would assure that individuals, who did not develop sterile immunity (sterile immunity has never been achieved in field trials) would have a reduced

This chapter will highlight the major readout methods that are currently used to gauge the efficacy of blood stage vaccines. In the absence of 'protection' models, the study of blood stage vaccines requires the development of *in vitro* assays that predict vaccine efficacy. In the absence of a definitive correlate of protection, the predictive value of these assays still awaits validation. Various target antigens for vaccine development will be highlighted with a special focus on the Merozoite Surface Protein (MSP)-1. Lastly, we will discuss modes of action of immune effector mechanisms against blood stage parasites and provide a preview

The ability to culture blood stage parasites of *P. falciparum* has greatly assisted in our understanding of events associated with maturation of the intra-erythrocytic parasite (Trager and Jensen, 1976; Haynes and Moch, 2002; Haynes et al., 2002). This is underscored by the lack of knowledge for other *Plasmodium* species that have not been successfully established in culture. Once the liver schizonts have reached maturity, the hepatocyte membrane dissolves and the merosome, *i.e*., the parasitophorous vacuole which resembles a bag containing infectious merozoites, is released from the liver destined for the lung capillaries where infection of erythrocytes occurs (Baer et al., 2007). The erythrocytic life cycle of *P. falciparum* blood stage parasites ranges from 38-48 hrs. During the first 24 hrs the parasite has a ring-like morphology (ring stage parasite) which subsequently develops into a trophozoite that is clearly distinguishable by the increasing amount of DNA and cytoplasm. Further maturation of the parasite into the schizont stage is marked by DNA segmentation resulting in individual nuclei. Additionally, schizonts have increasing amounts of digested hemoglobin (*aka*, malaria pigment or hemozoin). The schizonts rupture after first dissolving the parasitophorous vacuole and then the erythrocyte membrane, thus releasing 16-32 infectious merozoites per new infected erythrocyte (Cowman and Crabb, 2006). This event typically occurs in post capillary venules where the flow rate of the blood is low thereby favoring the rapid invasion of erythrocytes. Free merozoites have a relatively short lifespan and thus it is thought that the invasion event occurs quickly (within 5 minutes). This poses a significant challenge for effector antibodies which have to engage the parasite during this brief period of vulnerability and to block the interactions between host cells and merozoites. Successful active or passive immunotherapy requires providing an ample supply of antigenspecific, high-affinity antibodies capable of blocking a 'tsunami' of parasites every 38-48

**2. Methodological tools to study anti-parasite activities mediated by** 

risk of dying as a result of the added blood stage component.

of next-generation malaria vaccines.

*Plasmodium***-specific antibodies** 

hours. Whether this will be achievable to prevent morbidity or whether blood stage vaccineinduced antibodies will only be able to prevent severe disease and mortality is currently the focus of erythrocytic vaccines. Ideally, *in vitro* methods and/or preclinical models would assist in the down-selection of vaccine candidates. However, currently there is no preclinical model that adequately simulates human malaria. In contrast, there are several *in vitro* methods that have been used for analyzing functional antibodies against blood stage parasites *in vitro*. The following methods have been employed extensively in the evaluation of blood stage vaccines and we outline the advantages and drawbacks of the various methods as well as their predictive value.

#### **2.1 Functional assays capturing the biological activity of blood stage, parasitespecific antibodies**

Various methods have been developed that allow the measurement of the biological functionality displayed by either the antibodies themselves (growth inhibition assays (GIA) and invasion inhibition assays (IIA)) or antibodies in collaboration with immune cells expressing Fc-receptors (receptors that naturally bind certain immunoglobulin isotypes in order to eliminate antigens bound by antibodies). Antibodies bind their specific pathogen and form "immune complexes". Depending on the class and isotype of the antibody in the immune complex, this interaction leads to different outcomes: (1) the immune complex activates the complement system, *i.e.* complement fixation (which ultimately results in the lysis of the attached pathogen) or (2) the immune complex binds to Fc-receptors, *i.e.* opsonization (mediating uptake of the bound pathogen by the immune cells) or (3) antibody binds to the antigen and inactivates (neutralizes) the pathogen and thus prevents infection. The functional analysis of anti-blood stage antibodies, especially when obtained from individuals in endemic areas, is complicated by the presence of toxic factors or lipids in sera or plasma (such as anti-malarial drugs and anti-coagulants, or oxidized lipids due to long or inadequate sample storage). Eliminating nonspecific toxicities requires purification of the antibodies from serum or plasma using either bulk enrichment techniques (such as precipitation with ammonium sulfate and caprylic acid) or immunoglobulin purification methods (*e.g*., the use of Protein A, G , L-columns). It should be noted that most anti-malarial drugs can be removed by simply dialyzing the samples (Sy et al., 1990; Persson et al., 2006) which allows the testing of samples where only small volumes are available as is often the case for pediatric specimen. When comparing various methods in regards to yield, purity, integrity and retention of functional activity of recovered antibodies, we found that some methods lead to the selective enrichment of certain isotypes or in some cases fairly unstable immunoglobulin preparations (Bergmann-Leitner et al., 2008b). Also, not all purification methods perform equally well when purifying immunoglobulins from different species. The choice of purification method greatly influences the results obtained in the functional assay and can lead to significant artifacts.

#### **2.1.1 Microscopic analysis**

Microscopic evaluation of blood smears stained with Giemsa is the classic method for detecting parasitized erythrocytes (pRBC) and remains the gold standard in clinical trials. It is also used for studying the effect of antibodies on the morphology of the intra-erythrocytic parasite. To detect parasitemia in the blood, thick film-smears are prepared and 100

The Impact of Immune Responses on the Asexual

therefore, host cells do not introduce artifacts into this analysis.

culture is used to report the percentage of growth inhibition.

**2.1.3 Measurement of invasion inhibition by using transgenic parasites** 

In an attempt to identify the contribution of MSP-1 mediated invasion inhibition, O'Donnell *et al*. (O'Donnell et al., 2001) developed a *P. falciparum* D10 clone (MAD 20 allele) transgenic in the MSP-1p19 portion where the wildtype sequence was replaced by the p19 of *P. chabaudi*. Testing the inhibitory activity of antibodies against the parental D10 compared to the transgenic parasites identifies the contribution of *P.falciparum* MSP-1 mediated parasite inhibition. Other transgenic parasites have been developed that express the green fluorescence protein to facilitate the detection of the parasites within erythrocytes (O'Donnell et al., 2000). This approach eliminates the need for intracellular DNA staining of pRBCs. Using these parasites requires maintaining an additional parasite strain in the laboratory, which adds cost and labor, but it greatly facilitates the analysis because no additional manipulation of the cultures, *i.e*., DNA staining, is required to perform the

Erythrocytic Stages of *Plasmodium* and the Implication for Vaccine Development 205

(2) Use of viability DNA stains such as hydroethidine (HE) (van der Heyde et al., 1995). HEstaining depends on the intracellular conversion of HE into ethidium by parasitic NADPH oxidase and has been described in various protozoan systems including malaria to be a reliable metabolic indicator of parasite viability (Wyatt et al., 1991; van der Heyde et al., 1995). The host erythrocytes' enzymatic activity is not sufficient to convert the dye and,

Fig. 1. DNA content of blood cultures infected with *P. falciparum* in the presence of control (malaria-naïve, not growth inhibitory) serum (Panel A) or immune serum (Panel B) determined by flow cytometric analysis. X-axis = Mean fluorescence intensity of Syto-16 DNA staining, Y-axis = arbitrary unit for cell number. Brackets under X-axis indicate the DNA content of uninfected erythrocytes (RBC), ring stage parasites (RS), multiply infected ring stage parasites (MI-RS) and trophozoites (Tz) and schizont stage parasites (Sz). Marker M1 captures young pRBC and M2 mature pRBC and the changes of the proportion have been used to calculate growth inhibition. Panel A serves as reference for healthy, unimpaired parasite growth. Panel B indicates growth retardation as there is a higher prevalence of ring and trophozoite stage parasites compared to the control serum. The difference between the area under the curve for M1 of the test culture and the control

microscopic fields are screened for the presence of parasites at a 1000x magnification. For quantitation, 200 leukocytes and all pRBC in the respective fields of a thin film-smear are counted. If nine or less pRBC per 200 leukocytes are detected the quantitation is extended to 500 leukocytes. The equipment is readily available in all clinical labs and is fairly cheap. However, since the analysis is done by humans, it is essential that it is performed in a double-blinded fashion to avoid introduction of bias and subjectivity. Moreover, the microscopists have to be highly trained and experienced in slide reading in order to reliably identify malaria parasites in thick and thin film blood smears. The expertise of microscopists has long been recognized as the key issue for evaluating vaccine and malarial drug trials as quantifying parasitemia of malaria infected individuals commonly defines a study's primary endpoint. Furthermore, manual analysis, *i.e*. individual slide reading, is also rather slow and therefore not suitable for high-throughput screening. The microscopic analysis permits the quantification of parasitemia, but cannot objectively characterize changes in morphology which would indicate developmental growth inhibition or retardation.

#### **2.1.2 Flow cytometric analysis**

Flow cytometry is an ideal methodology for the automated and objective analysis of large numbers of cells. It provides information about the number of positively stained cells as well as the intensity of staining (Shapiro, 2004). Staining of mature human erythrocytes with DNA dyes allows for the detection of *Plasmodium* infection as these host cells are devoid of DNA unless parasites are present. This analysis reveals both the percentage parasitemia in the culture (% DNA containing cells) as well as the intracellular DNA content which is indicative of the parasite's maturation stages within the infected cells. Various DNA dyes have been reported to be useful for the detection and quantification of pRBC (reviewed in (Grimberg, 2011)). We have compared several DNA dyes side-by-side (*i.e.,* Syto-16, SybrGreen and Hydroethidine) in order to determine their usefulness in measuring invasion and growth inhibition of anti-blood stage antibodies. Using standard DNA binding dyes such as Syto-16 (Brand et al., 2003) and SybrGreen allows only the quantification of pRBC in culture and does not assess the viability of the intra-erythrocytic parasite (Green et al., 1981; Pang et al., 1999; Tebo et al., 2001; Haynes and Moch, 2002). If invasion occurs but the parasite fails to thrive, then the infected cell is detectable. Thus this approach measures only invasion inhibition since antibodies that block invasion result in either low or no parasitemia. In order to assess actual developmental growth inhibition, a different strategy has to be employed:

(1) Measurement of the DNA content of the pRBC as a correlate of the maturation stage of the parasite (Figure 1). A prerequisite for this approach is the tight synchronization of the parasites – either through repeated Percoll/Sorbitol purification or growth in temperaturecycling incubators - prior to setting up the experiment as the analysis requires setting strict cut-off values (see Fig. 1) for the different maturation stages of the parasite (ring/trophozoite/schizont). Such cut-off values have to be established for every parasite clone and isolate and have to be revised whenever culture conditions (new batch of serum/ culture media) change. Failure to control for these changes and the different growth kinetics of isolates and clones results in inaccurate estimates of parasite growth and yields uninterpretable results.

microscopic fields are screened for the presence of parasites at a 1000x magnification. For quantitation, 200 leukocytes and all pRBC in the respective fields of a thin film-smear are counted. If nine or less pRBC per 200 leukocytes are detected the quantitation is extended to 500 leukocytes. The equipment is readily available in all clinical labs and is fairly cheap. However, since the analysis is done by humans, it is essential that it is performed in a double-blinded fashion to avoid introduction of bias and subjectivity. Moreover, the microscopists have to be highly trained and experienced in slide reading in order to reliably identify malaria parasites in thick and thin film blood smears. The expertise of microscopists has long been recognized as the key issue for evaluating vaccine and malarial drug trials as quantifying parasitemia of malaria infected individuals commonly defines a study's primary endpoint. Furthermore, manual analysis, *i.e*. individual slide reading, is also rather slow and therefore not suitable for high-throughput screening. The microscopic analysis permits the quantification of parasitemia, but cannot objectively characterize changes in morphology which would indicate developmental growth

Flow cytometry is an ideal methodology for the automated and objective analysis of large numbers of cells. It provides information about the number of positively stained cells as well as the intensity of staining (Shapiro, 2004). Staining of mature human erythrocytes with DNA dyes allows for the detection of *Plasmodium* infection as these host cells are devoid of DNA unless parasites are present. This analysis reveals both the percentage parasitemia in the culture (% DNA containing cells) as well as the intracellular DNA content which is indicative of the parasite's maturation stages within the infected cells. Various DNA dyes have been reported to be useful for the detection and quantification of pRBC (reviewed in (Grimberg, 2011)). We have compared several DNA dyes side-by-side (*i.e.,* Syto-16, SybrGreen and Hydroethidine) in order to determine their usefulness in measuring invasion and growth inhibition of anti-blood stage antibodies. Using standard DNA binding dyes such as Syto-16 (Brand et al., 2003) and SybrGreen allows only the quantification of pRBC in culture and does not assess the viability of the intra-erythrocytic parasite (Green et al., 1981; Pang et al., 1999; Tebo et al., 2001; Haynes and Moch, 2002). If invasion occurs but the parasite fails to thrive, then the infected cell is detectable. Thus this approach measures only invasion inhibition since antibodies that block invasion result in either low or no parasitemia. In order to assess actual developmental growth inhibition, a different strategy

(1) Measurement of the DNA content of the pRBC as a correlate of the maturation stage of the parasite (Figure 1). A prerequisite for this approach is the tight synchronization of the parasites – either through repeated Percoll/Sorbitol purification or growth in temperaturecycling incubators - prior to setting up the experiment as the analysis requires setting strict cut-off values (see Fig. 1) for the different maturation stages of the parasite (ring/trophozoite/schizont). Such cut-off values have to be established for every parasite clone and isolate and have to be revised whenever culture conditions (new batch of serum/ culture media) change. Failure to control for these changes and the different growth kinetics of isolates and clones results in inaccurate estimates of parasite growth and yields

inhibition or retardation.

has to be employed:

uninterpretable results.

**2.1.2 Flow cytometric analysis** 

(2) Use of viability DNA stains such as hydroethidine (HE) (van der Heyde et al., 1995). HEstaining depends on the intracellular conversion of HE into ethidium by parasitic NADPH oxidase and has been described in various protozoan systems including malaria to be a reliable metabolic indicator of parasite viability (Wyatt et al., 1991; van der Heyde et al., 1995). The host erythrocytes' enzymatic activity is not sufficient to convert the dye and, therefore, host cells do not introduce artifacts into this analysis.

Fig. 1. DNA content of blood cultures infected with *P. falciparum* in the presence of control (malaria-naïve, not growth inhibitory) serum (Panel A) or immune serum (Panel B) determined by flow cytometric analysis. X-axis = Mean fluorescence intensity of Syto-16 DNA staining, Y-axis = arbitrary unit for cell number. Brackets under X-axis indicate the DNA content of uninfected erythrocytes (RBC), ring stage parasites (RS), multiply infected ring stage parasites (MI-RS) and trophozoites (Tz) and schizont stage parasites (Sz). Marker M1 captures young pRBC and M2 mature pRBC and the changes of the proportion have been used to calculate growth inhibition. Panel A serves as reference for healthy, unimpaired parasite growth. Panel B indicates growth retardation as there is a higher prevalence of ring and trophozoite stage parasites compared to the control serum. The difference between the area under the curve for M1 of the test culture and the control culture is used to report the percentage of growth inhibition.

## **2.1.3 Measurement of invasion inhibition by using transgenic parasites**

In an attempt to identify the contribution of MSP-1 mediated invasion inhibition, O'Donnell *et al*. (O'Donnell et al., 2001) developed a *P. falciparum* D10 clone (MAD 20 allele) transgenic in the MSP-1p19 portion where the wildtype sequence was replaced by the p19 of *P. chabaudi*. Testing the inhibitory activity of antibodies against the parental D10 compared to the transgenic parasites identifies the contribution of *P.falciparum* MSP-1 mediated parasite inhibition. Other transgenic parasites have been developed that express the green fluorescence protein to facilitate the detection of the parasites within erythrocytes (O'Donnell et al., 2000). This approach eliminates the need for intracellular DNA staining of pRBCs. Using these parasites requires maintaining an additional parasite strain in the laboratory, which adds cost and labor, but it greatly facilitates the analysis because no additional manipulation of the cultures, *i.e*., DNA staining, is required to perform the

The Impact of Immune Responses on the Asexual

Erythrocytic Stages of *Plasmodium* and the Implication for Vaccine Development 207

However, antibodies directed against MSP-1p42 preferentially inhibited invasion or inhibited either parasite growth and development, depending on the parasite test strain. MSP-1 specific antibodies acted on the FVO parasite clone mainly by inhibiting invasion. In contrast, the same antibodies mediated invasion- and growth inhibition in the case of 3D7 parasites (the exact mechanisms involved in this anti-parasite activity will be discussed in section 5). This study demonstrated that readout methods, which can distinguish between invasion inhibition and growth inhibition of the intra-erythrocytic parasite must be

In an effort to increase the sensitivity of the readout methods used to assess the anti-parasite activity through DNA dyes and flow cytometric analysis, multi-cycle experiments have been evaluated (Haynes et al., 2002; Persson et al., 2006), (Bergmann-Leitner, unpublished observations). To this end, cultures were not limited to the length of a single life cycle (40-48 hours duration), but allowed to continue for at least another cycle (72-96 hours duration). Several caveats are associated with this experimental setup: (1) The starting parasitemia of the culture has to be adjusted for each isolate/clone to assure that the parasites will not overgrow after the completion of the first cycle. This would result in the depletion of nutrients and the exhausted culture conditions would result in an overestimation of growth inhibition. One remedy has been to feed the cultures once or repeatedly. This can, however, pose yet another challenge as the exchange of media dilutes the antibodies and without knowing at which stage they will take effect, this manipulation could reduce their biological activity. Replenishing the antibodies when changing the media may lead to an overestimation of the biological effect in cases where the antibodies have already bound to the surface of the pRBC or entered the parasitophorous vacuole. Therefore, adding more antibody artificially increases the total antibody concentration; (2) Outgasing of cultures (*i.e.,* change in the ratio of CO2 and O2 in the atmosphere due to exposure to regular air) during the feeding process, can result in reduced invasion or slowed growth; (3) Interrupting invasion events when feeding occurs very closely to the time of schizont rupture. Estimating a safe time frame may be challenging because antibodies may slow the growth of the parasite, but not inhibit it and the growth inhibitory effect would be amplified; and (4) Controlling for proper starting parasitemia and modifying feeding times still leads to reduced multiplication rates in static cultures. Using suspension cultures for larger culture vessels such as 24-well plates and culture flasks yielded similar multiplication rates of the parasites during the second cycle compared to the first cycle (when feeding was also performed). It should be noted, however, that we have been unable to replicate the beneficial effect of suspension cultures when scaling down the assay format to 96 well and

employed in order to comprehensively define the antibodies' mechanism of action.

48 well plates (Bergmann-Leitner, unpublished observation).

**2.3 Functional assays to evaluate antibody dependent cellular cytotoxicity (ADCC)**  Early studies have shown that antibodies *per se* may not be sufficient to block the blood stage parasites (Bouharoun-Tayoun et al., 1990; Ouevray et al., 1994) and a cellular component is involved in the antibody-mediated anti-parasitic effect. Compared to classic antibody-dependent cellular cytotoxicity (ADCC), the postulated antibody-dependent cellular inhibition (ADCI) is mediated only by blood monocytes (not macrophages, polymorphonuclear neutrophils (PMN), lymphocytes, platelets) and primarily by cytophilic immunoglobulin classes (particularly IgG3) that target merozoite surface antigens

analysis. The analysis would yield similar results as staining with a general DNA dye such as SybrGreen, which allows accurate detection of invasion inhibition while the quantitation of growth inhibition is challenging as outlined above.

### **2.1.4 Measurement of the metabolic activity of the intra-erythrocytic parasite**

An alternative to quantifying either parasitized cells or the DNA content of infected erythrocytes is the measurement of enzymatic activity. The following approaches have been previously described: (1) the quantification of 3H-hypoxanthine incorporation into newly synthesized DNA (Bungener and Nielsen, 1968; Rahman, 1997), (2) the conversion of hydroethidine to ethidium (Wyatt et al., 1991; van der Heyde et al., 1995), and (3) detection/measurement of the parasite-derived lactate dehydrogenase (pLDH) (Makler and Hinrichs, 1993; Prudhomme and Sherman, 1999). Measuring the metabolic activity allows for the detection of parasitized erythrocytes that are viable and metabolically active thus enabling the quantitation of growth inhibition. It should be pointed out that multiple LDHsubstrates have been described in the literature for *Plasmodium ssp*. There are, however, qualitative differences between these substrates that will affect the sensitivity and the signalto-noise ratio. The protocol developed at NIAID/NIH (Kennedy et al., 2002) has demonstrated robustness, reproducibility and specificity (Corran PH, manuscript in preparation). This basic protocol was optimized in our laboratory to accommodate small sample volumes (< 50 µl) and high throughput screening utilizing 384 well plates and assay volumes of ≤ 20 µl (Bergmann-Leitner et al., 2008a).

#### **2.2 The quest for the most sensitive method for identifying anti-parasite activity of anti-malarial antibodies**

We have compared four methods in an attempt to identify the most sensitive method capable of measuring an array of functional activities displayed by immune antibodies, namely: (1) GIEMSA-staining of blood smears (with visual detection); (2) flow cytometric analysis using either a standard DNA binding dye (Syto-16) or (3) hydroethidine; and (4) pLDH detection using a substrate specific for the parasitic enzyme LDH (Bergmann-Leitner et al., 2006). The model system to evaluate anti-parasite activities was based on the use of immune sera specific for either AMA-1 (Kennedy et al., 2002) or MSP-1p42 (Angov et al., 2003; Darko et al., 2005) and previously established *P. falciparum* parasite clones (namely 3D7 and FVO). While staining with Giemsa revealed antibody-induced morphological changes in parasite development after exposure to immune serum, directly correlating these changes with parasite viability and thus efficacy of the antibodies is limited due to the subjective nature of the readout method. The DNA dye Syto16 readily permeates membranes of both viable and non-viable cells and thus cannot be used to determine the parasites' viability since any erythrocyte with DNA content will equally be identified as an "infected cell", thus under-estimating the inhibition. Growth inhibition (viability), however, can be determined by either flow cytometric analysis of parasites whose DNA was stained with HE or by measuring the enzymatic activity of pLDH. The conclusion from this comparative study was that the mode of action of antibodies directed against malaria blood stage antigens depends on not only the target antigen but also on the parasite strain. In cases where antibodies primarily mediated invasion inhibition, such as anti-AMA-1 specific antibodies (Triglia et al., 2000; Healer et al., 2004), all four methods yielded similar results.

analysis. The analysis would yield similar results as staining with a general DNA dye such as SybrGreen, which allows accurate detection of invasion inhibition while the quantitation

An alternative to quantifying either parasitized cells or the DNA content of infected erythrocytes is the measurement of enzymatic activity. The following approaches have been previously described: (1) the quantification of 3H-hypoxanthine incorporation into newly synthesized DNA (Bungener and Nielsen, 1968; Rahman, 1997), (2) the conversion of hydroethidine to ethidium (Wyatt et al., 1991; van der Heyde et al., 1995), and (3) detection/measurement of the parasite-derived lactate dehydrogenase (pLDH) (Makler and Hinrichs, 1993; Prudhomme and Sherman, 1999). Measuring the metabolic activity allows for the detection of parasitized erythrocytes that are viable and metabolically active thus enabling the quantitation of growth inhibition. It should be pointed out that multiple LDHsubstrates have been described in the literature for *Plasmodium ssp*. There are, however, qualitative differences between these substrates that will affect the sensitivity and the signalto-noise ratio. The protocol developed at NIAID/NIH (Kennedy et al., 2002) has demonstrated robustness, reproducibility and specificity (Corran PH, manuscript in preparation). This basic protocol was optimized in our laboratory to accommodate small sample volumes (< 50 µl) and high throughput screening utilizing 384 well plates and assay

**2.2 The quest for the most sensitive method for identifying anti-parasite activity of** 

We have compared four methods in an attempt to identify the most sensitive method capable of measuring an array of functional activities displayed by immune antibodies, namely: (1) GIEMSA-staining of blood smears (with visual detection); (2) flow cytometric analysis using either a standard DNA binding dye (Syto-16) or (3) hydroethidine; and (4) pLDH detection using a substrate specific for the parasitic enzyme LDH (Bergmann-Leitner et al., 2006). The model system to evaluate anti-parasite activities was based on the use of immune sera specific for either AMA-1 (Kennedy et al., 2002) or MSP-1p42 (Angov et al., 2003; Darko et al., 2005) and previously established *P. falciparum* parasite clones (namely 3D7 and FVO). While staining with Giemsa revealed antibody-induced morphological changes in parasite development after exposure to immune serum, directly correlating these changes with parasite viability and thus efficacy of the antibodies is limited due to the subjective nature of the readout method. The DNA dye Syto16 readily permeates membranes of both viable and non-viable cells and thus cannot be used to determine the parasites' viability since any erythrocyte with DNA content will equally be identified as an "infected cell", thus under-estimating the inhibition. Growth inhibition (viability), however, can be determined by either flow cytometric analysis of parasites whose DNA was stained with HE or by measuring the enzymatic activity of pLDH. The conclusion from this comparative study was that the mode of action of antibodies directed against malaria blood stage antigens depends on not only the target antigen but also on the parasite strain. In cases where antibodies primarily mediated invasion inhibition, such as anti-AMA-1 specific antibodies (Triglia et al., 2000; Healer et al., 2004), all four methods yielded similar results.

**2.1.4 Measurement of the metabolic activity of the intra-erythrocytic parasite** 

of growth inhibition is challenging as outlined above.

volumes of ≤ 20 µl (Bergmann-Leitner et al., 2008a).

**anti-malarial antibodies** 

However, antibodies directed against MSP-1p42 preferentially inhibited invasion or inhibited either parasite growth and development, depending on the parasite test strain. MSP-1 specific antibodies acted on the FVO parasite clone mainly by inhibiting invasion. In contrast, the same antibodies mediated invasion- and growth inhibition in the case of 3D7 parasites (the exact mechanisms involved in this anti-parasite activity will be discussed in section 5). This study demonstrated that readout methods, which can distinguish between invasion inhibition and growth inhibition of the intra-erythrocytic parasite must be employed in order to comprehensively define the antibodies' mechanism of action.

In an effort to increase the sensitivity of the readout methods used to assess the anti-parasite activity through DNA dyes and flow cytometric analysis, multi-cycle experiments have been evaluated (Haynes et al., 2002; Persson et al., 2006), (Bergmann-Leitner, unpublished observations). To this end, cultures were not limited to the length of a single life cycle (40-48 hours duration), but allowed to continue for at least another cycle (72-96 hours duration). Several caveats are associated with this experimental setup: (1) The starting parasitemia of the culture has to be adjusted for each isolate/clone to assure that the parasites will not overgrow after the completion of the first cycle. This would result in the depletion of nutrients and the exhausted culture conditions would result in an overestimation of growth inhibition. One remedy has been to feed the cultures once or repeatedly. This can, however, pose yet another challenge as the exchange of media dilutes the antibodies and without knowing at which stage they will take effect, this manipulation could reduce their biological activity. Replenishing the antibodies when changing the media may lead to an overestimation of the biological effect in cases where the antibodies have already bound to the surface of the pRBC or entered the parasitophorous vacuole. Therefore, adding more antibody artificially increases the total antibody concentration; (2) Outgasing of cultures (*i.e.,* change in the ratio of CO2 and O2 in the atmosphere due to exposure to regular air) during the feeding process, can result in reduced invasion or slowed growth; (3) Interrupting invasion events when feeding occurs very closely to the time of schizont rupture. Estimating a safe time frame may be challenging because antibodies may slow the growth of the parasite, but not inhibit it and the growth inhibitory effect would be amplified; and (4) Controlling for proper starting parasitemia and modifying feeding times still leads to reduced multiplication rates in static cultures. Using suspension cultures for larger culture vessels such as 24-well plates and culture flasks yielded similar multiplication rates of the parasites during the second cycle compared to the first cycle (when feeding was also performed). It should be noted, however, that we have been unable to replicate the beneficial effect of suspension cultures when scaling down the assay format to 96 well and 48 well plates (Bergmann-Leitner, unpublished observation).

#### **2.3 Functional assays to evaluate antibody dependent cellular cytotoxicity (ADCC)**

Early studies have shown that antibodies *per se* may not be sufficient to block the blood stage parasites (Bouharoun-Tayoun et al., 1990; Ouevray et al., 1994) and a cellular component is involved in the antibody-mediated anti-parasitic effect. Compared to classic antibody-dependent cellular cytotoxicity (ADCC), the postulated antibody-dependent cellular inhibition (ADCI) is mediated only by blood monocytes (not macrophages, polymorphonuclear neutrophils (PMN), lymphocytes, platelets) and primarily by cytophilic immunoglobulin classes (particularly IgG3) that target merozoite surface antigens

The Impact of Immune Responses on the Asexual

Serum Serum & monocytes

section 5).

**% Growth Inhibition**

**0**

**areas** 

**10**

**20**

**30**

**40**

**50**

**60**

**3D7**

Erythrocytic Stages of *Plasmodium* and the Implication for Vaccine Development 209

What role does the parasite isolate or clone play when measuring ADCI activity? Using a sample from an individual from a malaria-endemic area (Lyon et al., 1997) we were able to detect a differential contribution of ADCI in the growth-inhibition of two *P. falciparum* clones: various amounts of serum in the assay in the presence or absence of monocytes were tested against the 3D7 and FVO parasite clones (Figure 2). The main difference in the clones is the fact that they represent the two distinct MSP-1 alleles (the MAD20 and the Wellcome K1 allele, respectively) and they have also been described as invading either through sialic acid independent (3D7) or dependent (FVO) pathways. The inhibitory activity of serum against 3D7 parasites was not increased by the presence of monocytes while the inhibition of FVO parasites could be amplified by adding monocytes (data not published). This synergistic effect of antibodies and monocytes became apparent when limiting the amount of serum in the assay. More studies are needed to determine the underlying mechanism for this differential effect as this may be a matter of antibody fine specificity (discussed in

**Serum Concentration (vol%)**

Fig. 2. Effect of inhibitory antibodies and antibody-dependent cellular inhibition on growth of 3D7 (left Panel) and FVO (right Panel) parasite clones. X-axis indicates percentage of human malaria-immune serum in the culture; y-axis shows the % growth inhibition measured by quantifying pLDH. White bars = growth inhibition induced by serum only, black bars = growth inhibition induced by serum and CD14+ IFN-γ activated monocytes.

**3. Anti-parasite activities in sera from individuals living in malaria-endemic** 

Protection against malaria mediated by antibodies recognizing the erythrocytic parasite or the parasitized erythrocyte is mediated by several distinct mechanisms: (1) Binding of the antibody to the surface of merozoites can interfere with the invasion of new erythrocytes and opsonize the merozoite, which results in complement activation and/or phagocytosis; (2) Binding to the merozoite may not suffice to block invasion, but antibodies carried into

Data expressed as mean % growth inhibition, error bars indicate SD.

**5 2.5 1.25 0 5 2.5 1.25 0**

**0**

**10**

**20**

**30**

**40**

**50**

**60**

**% Growth Inhibition**

**FVO**

(Bouharoun-Tayoun et al., 1995; Tebo et al., 2001). The activation of monocytes through FcγRII triggers the release of soluble TNF-α that in turn blocks the development of adjacent intra-erythrocytic parasites (Bouharoun-Tayoun et al., 1995). Furthermore, a role of FcγRI in contributing to clinical malaria has been suggested based on the functional properties of recombinant human antibodies derived from immune Gambian adults (McIntosh et al., 2007). These antibodies were tested in mice transgenic for human Fc-receptors challenged with transgenic rodent malaria parasites (expressing *Pf*MSP-1p19) in an effort to evaluate the role of antibodies in protection. Another study demonstrated that the activation of monocytic cells requires two distinct Fcγ receptors (FcγRII and FcγRIII) simultaneously engaged by a least two cytophilic IgG molecules, which are part of the same immune complexes (Jafarshad et al., 2007). Finally, the concept of the contribution of Fc receptors in natural immunity is supported by epidemiological studies (Shi et al., 2001; Israelsson et al., 2008; Leoratti et al., 2008).

We conclude from these studies that ADCI as an anti-parasite defense mechanism depends on the target antigen, a concept with significant implications for vaccine development (revisited below in section 4). Experiments measuring the ADCI activity of anti-blood stage antibodies are set up in principle as described for growth/invasion inhibition assays (above): parasite cultures are established at a defined percentage of parasitemia in the presence or absence of immune antibodies and – in addition - in the presence of monocytic cells. Readout methods used are the same as for the standard "growth inhibition" assays as described earlier (our lab prefers the pLDH assay for this purpose). Note that the source and quality of the monocytic cells will greatly influence the results obtained in these experiments: (1) freshly isolated PBMC (plastic adherence *vs*. CD14 MACS separated cells *vs*. monocytic tumor cell lines); (2) resting *vs*. cytokine activated cells; or (3) inter-donor variations. When optimizing the ADCI protocol, it is important that the *in vitro* conditions including the cell types of phagocytes used simulate the *in vivo* setting as closely as possible in order to reproduce the conditions in the malaria infected host. Using cancer cells such as the monocytic tumor cell line THP-1 may not adequately mimic the effector population found in the patient. This is underscored by recent findings that infection and acute malaria can drastically change the phenotype of monocytic subpopulations (Chimma et al., 2009). This includes changes in chemokine receptors as well as functional properties. This was evidenced as monocytic cells derived from patients during an acute malaria infection were superior over the monocytes derived from healthy donor individuals. However, other studies have described significant impairments of monocyte functions due to malaria infection, such as their ability to phagocytize (Leitner and Krzych, 1997). This underscores that there is more to monocytic cells than just the expression of Fc-receptors and that these observations warrant the characterization of effector cells used for *in vitro* assays in an effort to minimize *in vitro* artifacts.

To this end, we tested various methods for generating activated monocytic cells *in vitro* and found that the purification of monocytes with magnetic bead labeled anti-CD14 monoclonal antibody resulted in higher and more reproducible ADCI-activities as compared to cells that were isolated by plastic adherence. Moreover, pre-activating the isolated cells overnight with 300 U/ml IFN-γ reduced the inter-assay variability. Pre-activation also decreased the variability observed between different blood donors consistent with the concept that the stimulation "normalizes" the activation state of the cells (data not published).

(Bouharoun-Tayoun et al., 1995; Tebo et al., 2001). The activation of monocytes through FcγRII triggers the release of soluble TNF-α that in turn blocks the development of adjacent intra-erythrocytic parasites (Bouharoun-Tayoun et al., 1995). Furthermore, a role of FcγRI in contributing to clinical malaria has been suggested based on the functional properties of recombinant human antibodies derived from immune Gambian adults (McIntosh et al., 2007). These antibodies were tested in mice transgenic for human Fc-receptors challenged with transgenic rodent malaria parasites (expressing *Pf*MSP-1p19) in an effort to evaluate the role of antibodies in protection. Another study demonstrated that the activation of monocytic cells requires two distinct Fcγ receptors (FcγRII and FcγRIII) simultaneously engaged by a least two cytophilic IgG molecules, which are part of the same immune complexes (Jafarshad et al., 2007). Finally, the concept of the contribution of Fc receptors in natural immunity is supported by epidemiological studies (Shi et al., 2001; Israelsson et al.,

We conclude from these studies that ADCI as an anti-parasite defense mechanism depends on the target antigen, a concept with significant implications for vaccine development (revisited below in section 4). Experiments measuring the ADCI activity of anti-blood stage antibodies are set up in principle as described for growth/invasion inhibition assays (above): parasite cultures are established at a defined percentage of parasitemia in the presence or absence of immune antibodies and – in addition - in the presence of monocytic cells. Readout methods used are the same as for the standard "growth inhibition" assays as described earlier (our lab prefers the pLDH assay for this purpose). Note that the source and quality of the monocytic cells will greatly influence the results obtained in these experiments: (1) freshly isolated PBMC (plastic adherence *vs*. CD14 MACS separated cells *vs*. monocytic tumor cell lines); (2) resting *vs*. cytokine activated cells; or (3) inter-donor variations. When optimizing the ADCI protocol, it is important that the *in vitro* conditions including the cell types of phagocytes used simulate the *in vivo* setting as closely as possible in order to reproduce the conditions in the malaria infected host. Using cancer cells such as the monocytic tumor cell line THP-1 may not adequately mimic the effector population found in the patient. This is underscored by recent findings that infection and acute malaria can drastically change the phenotype of monocytic subpopulations (Chimma et al., 2009). This includes changes in chemokine receptors as well as functional properties. This was evidenced as monocytic cells derived from patients during an acute malaria infection were superior over the monocytes derived from healthy donor individuals. However, other studies have described significant impairments of monocyte functions due to malaria infection, such as their ability to phagocytize (Leitner and Krzych, 1997). This underscores that there is more to monocytic cells than just the expression of Fc-receptors and that these observations warrant the characterization of effector cells used for *in vitro* assays in an effort

To this end, we tested various methods for generating activated monocytic cells *in vitro* and found that the purification of monocytes with magnetic bead labeled anti-CD14 monoclonal antibody resulted in higher and more reproducible ADCI-activities as compared to cells that were isolated by plastic adherence. Moreover, pre-activating the isolated cells overnight with 300 U/ml IFN-γ reduced the inter-assay variability. Pre-activation also decreased the variability observed between different blood donors consistent with the concept that the

stimulation "normalizes" the activation state of the cells (data not published).

2008; Leoratti et al., 2008).

to minimize *in vitro* artifacts.

What role does the parasite isolate or clone play when measuring ADCI activity? Using a sample from an individual from a malaria-endemic area (Lyon et al., 1997) we were able to detect a differential contribution of ADCI in the growth-inhibition of two *P. falciparum* clones: various amounts of serum in the assay in the presence or absence of monocytes were tested against the 3D7 and FVO parasite clones (Figure 2). The main difference in the clones is the fact that they represent the two distinct MSP-1 alleles (the MAD20 and the Wellcome K1 allele, respectively) and they have also been described as invading either through sialic acid independent (3D7) or dependent (FVO) pathways. The inhibitory activity of serum against 3D7 parasites was not increased by the presence of monocytes while the inhibition of FVO parasites could be amplified by adding monocytes (data not published). This synergistic effect of antibodies and monocytes became apparent when limiting the amount of serum in the assay. More studies are needed to determine the underlying mechanism for this differential effect as this may be a matter of antibody fine specificity (discussed in section 5).

Fig. 2. Effect of inhibitory antibodies and antibody-dependent cellular inhibition on growth of 3D7 (left Panel) and FVO (right Panel) parasite clones. X-axis indicates percentage of human malaria-immune serum in the culture; y-axis shows the % growth inhibition measured by quantifying pLDH. White bars = growth inhibition induced by serum only, black bars = growth inhibition induced by serum and CD14+ IFN-γ activated monocytes. Data expressed as mean % growth inhibition, error bars indicate SD.
