- Email: [email protected]
Protection by α-thalassaemia against Plasmodium falciparum malaria: modified surface antigen expression rather than impaired growth or cytoadherence
Immunology Letters, 30 (1991) 233 - 240 Elsevier IMLET 01694
Protection by a-thalassaemia against Plasmodium falciparum malaria: modified surface antigen expression rather than impaired growth or cytoadherence G. A. Luzzi 1, A. H. Merry 2, C. I. Newbold 1, K. Marsh I and G. Pasvol 3 1Molecular Parasitology Group, Institute of Molecular Medicine, John Radcliffe Hospital Oxford, 2Oxford Glycosystems Ltd., Unit 12, Abingdon, and 3St Mary's Hospital Medical School Department of Infectious Diseases and Tropical Medicine, Lister Unit, Northwick Park Hospital Middlesex, U.K. (Received 18 June 1991; accepted 25 June 1991)
I. Summary We have attempted to determine the cellular mechanism by which a-thalassaemia may protect against Plasmodium falciparum malaria. Invasion and development of P. falciparum in the microcytic red cells of two-gene deletion forms of a-thalassaemia when measured morphologically or by [3H]hypoxanthine incorporation were normal compared to controls. Normal invasion rates were also observed following schizogony in thalassaemic red cells. Neither the addition of the oxidant menadione, 30% oxygen, nor modified medium, produced differential damage to parasites within thalassaemic cells. Furthermore, there were no significant differences in the binding of P. falciparum-parasitized o~-thalassaemic and normal cells to C32 melanoma cells in vitro. However, when neoantigen expression on the surface of infected thalassaemic cells was estimated using a quantitative radiometric antiglobulin assay, clear differences were observed. It was found that ot-thalassaemic cells bound higher levels of antibody from serum obtained from individuals living in a malaria endemic area than control normal red cells. The binding ratio for thalasKey words: Malaria; Plasmodium falciparum; Thalassaemia; Neoantigen
Correspondence to: Professor G. Pasvol, St Mary's Hospital Medical School, Department of Infectious Diseases and Tropical Medicine, Lister Unit, Northwick Park Hospital, Middlesex, HA1 3U J, U.K.
saemic compared with controls was 1.69 on a cellfor-cell basis, and 1.97 when related to surface area. The binding of antibody from immune serum increased exponentially during parasite maturation. We also found increased binding of naturally occurring antibody present in non-immune serum to parasitized thalassaemic red cells which also increased during parasite maturation. We conclude that the protection afforded by thalassaemia against malaria may not reside in the ability of parasites to enter, grow or cytoadhere to endothelium in such cells, but may be related to immune recognition and subsequent clearance of parasitized red cells. 2. Introduction In studying the mechanism by which inherited red cell variants provide resistance against malaria, most studies have examined the ability of parasites to invade and grow in a particular cell type under given conditions [1]. In this context the mechanism whereby a-thalassaemia protects against malaria has remained particularly elusive despite the compelling epidemiological evidence that exists to support such a claim [2]. The gene frequency for this haemoglobinopathy may be as high as 80°-/o in certain parts of the world where malaria is or was endemic [3, 4]. It has been suggested that the relative susceptibility of thalassaemic cells (especially in 3-thalassaemia) to oxidant stress is responsible for the failure of parasites to survive in them [5]. Such oxidant stress has been achieved in vitro either
0165 - 2478 / 91 / $ 3.50 © Elsevier Science Publishers B.V. All rights reserved.
233
by addition of an oxidant such as menadione or riboflavine, or by modification of the growth medium [6]. Whilst growth may be impaired in cells from patients with the more severe forms of o~-thalassaemia (e.g., haemoglobin H disease) [7], and infected c~-thalassaemia red cells may be more susceptible to phagocytosis [8], we have not observed significant growth impairment in the milder forms of o~-thalassaemia under standard growth conditions [9]. We therefore set out not only to re-examine the effects of oxidant stress on parasite growth in thalassaemic red cells, but also to determine the ability of thalassaemic cells to bind to amelanotic melanom a cells (an in vitro model of cytoadherence) and finally to measure the expression of neoantigen on their surface when compared with normal red cells using a quantitative radiometric antiglobulin assay.
cycle was determined in order to investigate whether parasites grown in the variant cells were capable of giving rise to a further brood of parasites. In a number of experiments the oxidant menadione (Sigma) was added to the culture medium and in some a mixture containing 30% oxygen as previously used was used to gas the mixture, whilst in a further set a modified culture medium ( " R E M " ) was used as previously described [6]. This combined medium contains nine volumes of minimal essential medium (Gibco) and one volume of RPMI-1640, and no added L-glutamine. It therefore has only one-tenth of the reduced glutathione of the standard medium (RPMI-1640). All media were carefully pre-gassed with the appropriate gas mixture before use in experiments.
3. Materials and Methods
The in vitro model for sequestration using the C32 amelanotic melanoma cells was used as previously described [12]. Red cell suspensions containing a-thalassaemic or control ceils were incubated in triplicate on melanoma monolayers in 35-ram Petri dishes for 40 rain at 37 °C. After careful washing, monolayers with adherent parasitized ceils were fixed with two percent glutaraldehyde and stained with Giemsa stain. The number of adherent parasitized ceils per 50 melanoma cells were counted.
3.1. Parasite culture Strains of Plasmodium falciparum were maintained in continuous culture by conventional methods [10]. Parasite growth was regularly synchronised by sorbitol lysis and gelatin flotation or Percoll gradient purification. Whole blood was collected into acid citrate-dextrose (ACD) (1:10, v/v) stored at 4 °C and used within 48 h of collection. Samples were taken from individuals with two-gene deletion forms of a-thalassaemia (heterozygous oPthalassaemia ( - - / c w 0 , or homozygous o~+-tha lassaemia ( - c ~ / - a ) ) , and appropriate controls. Sickle cell trait and G 6 P D deficiency were excluded in all samples. 3.2. Invasion and growth assay These were carried out in triplicate in 96-well microtitre plates as previously described [9]. Cultures were incubated at 37 °C in sealed gas jars containing either a low oxygen gas mixture (1% 0 2, 5% CO2, 94°70 N2) or as indicated. After invasion had occurred, smears were prepared to assess invasion rates and a further incubation was carried out to determine maturation by morphological assessment or by [3H]hypoxanthine uptake [11]. In some experiments the rate of invasion in the following 234
3.3.
Cytoadherence assay
3.4. Quantitative estimation of lgG binding to red cells This was carried out by a method adapted from that described by Merry et al. [13]. Briefly, cultures of P. falciparum were established in thalassaemic and control cells. A fraction rich in trophozoites and schizonts (layer) was obtained by gelatin flotation [14] and the sedimented cells (pellet) were retained as a control. Such a pellet contained virtually no parasites. Both the layer and pellet were washed three times in wash buffer (WB) (PBS with 0.5% BSA; Sigma). For each sample, equal suspensions of infected and uninfected red cells were incubated with endemic serum (ES) obtained from healthy individuals living in an area highly endemic for malaria (The Gambia) or WB, washed, and then incubated with 100/zl of a 1:5 dilution of sheep or rabbit
serum and 100 t~l 125I-anti-human IgG (20 #g/ml). These were incubated for a further 30min at 37 °C and then washed three times in PBS. The red cells were resuspended in 120/~1 of wash buffer and a small aliquot removed to count the red cells in a Coulter counter. 100/~1 of the red cell suspension was spun through oil (4:1 di-n-butyl/dinonylphthalate; Sigma). The red cell pellets were cut off and counted in a gamma counter. The following formula was used to calculate the IgG bound (molecules per cell): IgG bound = (cpm/sra) x ( 6 z 103/n); involving a factor to convert/~g IgG to molecules of IgG/107 cells, where cpm equals counts per minute/100 ml; sra = specific radioactivity of 125I-IgG in counts per minute//~g IgG; and n = cell count/ 100/~lx 10 -7. To estimate IgG bound to parasitized cells, values for specific IgG bound (molecules/cell) for non-parasitized cells (pellet E s - pellet wB) were subtracted from the respective parasitized fraction results (layer Es - layerWB), and the result was divided by parasitaemia ( x 10-2). The value for non-parasitized cells was then added. Thus the number of IgG molecules bound per parasitized cell could be calculated. A further correction was made in order to relate the numbers of IgG bound to the surface area of the cells. This was carried out using the formula SA = 44.8 + 1.17 x MCV; where SA = the surface area #m2; and MCV = mean cell volume. 4. Results
4.1. Invasion and development in ~-thalassaemic cells Rates of invasion of P. falciparum (Uganda Palo Alto strain, UPA and a Thai cloned line, T9/96) into thalassaemic cells were the same as in controls. Moreover parasite development assessed both morphologically and by [3H]hypoxanthine incorporation indicated no difference between control and thalassaemic red ceils (Table 1). The invasion rates into the next cycle following schizogony within thalassaemic cells was also found to be completely normal (Table 1). When invasion and development of parasites was carried out in the presence of 0, 1, 2, 5 or 10ttmol of menadione progressive growth inhibition was observed. However, there was no difference between the normal and ct-thalassaemic
cells (Fig. 1). When these experiments were repeated using conditions of 30°70 oxygen, a further reduction in parasite incorporation of [3H]hypoxanthine was seen (Fig. 2). The invasion and growth of parasites in ot-thalassaemic and normal red cells were not different. Parasite development in the modified medium (REM) was slightly inhibited (+88070 of [3H]hypoxanthine uptake in standard medium) but the extent of this inhibition was no greater in thalassaemic cells (data not shown). 4.2. Cytoadherence assays No difference was observed in the ability of parasitized thalassaemic or normal cells to bind to amelanotic melanoma cells using the ITO (Brazil) parasite strain (Fig. 3). 4.3. Quantitative binding of IgG to infected tha-
lassaemic and normal red cells In these experiments using strain IT (Brasil), parasitized a-thalassaemic red cells consistently bound more IgG than controls (Fig. 4). The mean binding ratios of thalassaemic versus normal cells was 1.69 on a cell-for-cell basis and 1.97 when calculated per unit surface area. When infected cells were incubated in WB rather than ES, even though the amount of antibody was considerably less than that observed with ES, parasitized thalassaemic cells still appeared to bind more antibody when compared to controls (Fig. 5). The mean binding ratio of thalassaemic versus normal red ceils was 1.29 on a cellfor-cell basis, and 1.58 when calculated per unit red cell surface area. When the number of molecules of IgG bound to uninfected cells were determined, no such difference in binding between the thalassaemic and control ceils was observed (Table 2). Finally, the binding of IgG molecules during maturation of parasites within normal red cells incubated in either ES or WB buffer was determined. Results showed a steady increase from 12 h into the parasite life cycle until well into schizont maturation (Fig. 6). This increase applied to cells incubated in both ES and WB.
235
TABLE 1 Invasion, growth and reinvasion of P. falciparum in ~-thalassaemic compared to normal red cells. Genotype
--
- - / O t ~
--
--
Invasion (% control)
Growth (% control)
102.5 _+ 4.1 107.9 +_ 4.4 100.8 4-_ 1.7
/OZO(
O r / - - O~
(Control): (Range)
Invasion: Growth: Reinvasion:
Reinvasion (% control)
Morphology
3H uptake
101.7 _+ 4.5 100.2 + 4.0 101 _+ 2.0
104.3 _+ 1.0 96.9 +_ 1.0 106.5 _+ 2.0
101.8 +_ 4.6 117.8 _+ 7.0 102.2 _+ 10.7
2 - 17 Rings/100 cells 6 6 - 86 Schizonts per 100 singly infected red cells 1.54 to 4.85 × 103 counts per minute 2 - 13 Rings per 100 red cells
6.0 1"3 0
== o (.)
5.0 4.0
- - ~ '
3.0 2.0
--/c'.o. []
--IChor.
•
control
•
control
1.0 0.0
'
0.0
"
'
I
2.0
•
'
'
I
'
'
4.0
I
"
'
6.0
'
I
8.0
=
'
'
I
'
10.0
)uM Menadlone Fig. 1. The effect of menadione on the growth o f P . falciparum in c~-thalassaemic or control red cells as measured by [3H]hypoxanthine incorporation. Error bars show S.E. for triplicate values. 18 16 ~o
14
x
12
5%
g
•
: .::-.::::: 8
[]
30%
-{:d-o. conffol
. . . . . . . . . "[3" .......
---OJ-~..z
......... •~ .......
control
6 4 2 0
|
0.0
5.0
i
10.0
pM m e n a d l o n e Fig. 2. The effect of 30% oxygen on the growth of P. falciparum in ePthalassaemic or control red cells in the presence of increasing concentrations of menadione as measured by [3H]hypoxanthine incorporation. Error bars show S.E. for triplicate counts. 236
120 " o
o=
100 =
[]
80
[]
'~
--/~a
(I)
--/c~a (21
'~
60
[]
control (1)
.e.~
40
[]
control (2)
.2
-
.2
L
2o
0
1
5 10 Parasites added (xl0 ~) Fig. 3. Bindingto C32 melanoma cells of increasing numbers of P. fal¢iparum-infected a-thalassaemic and control red cells. Error bars show S.E. for 7 experiments.
o~ "0 e-
-1
40000]
30000-
0
20000 o
[]
control
[]
O. -thai
O
E (9
10000
0' 1
2
3
4
5
Expt. Fig. 4. Antibody detected on P. falciparum-infected c~-thalassaemic and control cells (molecules/100/zm 2) after incubation in endemic
sera (ES). Error bars show S.E. for triplicate values.
5. Discussion Whilst for m a n y of the haemoglobinopathies the epidemiological evidence for protection by them against fatal P. falciparum malaria is convincing, the molecular mechanisms by which these are brought about remain unclear. Moreover there is increasing evidence that these mechanisms might differ from condition to condition [1]. However, the way in which thalassaemic red cells might protect against malaria has remained most elusive. Whilst for sickle haemoglobin the failure of para-
sites to develop in such cells when maintained under reduced oxidant tension provides a rational basis for the way in which sickle haemoglobin might protect against malaria [15], for thalassaemia the susceptibility of parasite to oxidant stress has remained the favoured mechanism for protection in this disorder. Reduced growth of parasites within thalassaemic cells when exposed to menadione, increased oxygen tensions [5], and modified medium containing reduced amounts of antioxidants [6], would all support this mechanism of protection. However, in rigorous experiments using these vari237
6000 0
-e
5000
o
..~
4000 []
--
[]
3000
control cc -thai
o
o E
2000
0
1000 o 1
2
3
4
5
Expt. Fig. 5. Antibody detected on P. falciparum-infected c~-thalassaemic and control cells (molecules/100 ~m 2) after incubation in wash buffer (WB). Error bars show S.E. for triplicate values. TABLE 2 IgG detected on uninfected c~-thalassaemic and control cells after incubation in endemic serum or wash buffer. Control
c~-Thalassaemia
P
516 ± 85 345 ± 50
306 _+ 85 287 ± 39
0.26 NS
310 ± 31 207 ___ 20
220 _+ 21 172 _+ 16
NS NS
Endemic serum
Molecules per cell Molecules per 100/tin 2 Wash buffer
Molecules per cell Molecules per 100 # m 2 1ooooo
o
ES
-o
o
100(30
-~ E
looo
r=0.92
0 •
100 6
12
18
hours
, , ,
24 Into
30
36
42
48
life-cycle
Fig. 6. Antibody detected on P. falciparum-infected normal red cells 6-hourly during a complete asexual cycle in vitro using tightly synchronized parasites incubated in endemic serum (ES) or wash buffer (WB). Points represent mean values for triplicate samples (SE values within 10% for each point), r, correlation coefficient for exponential fit.
238
ous methods of exposing cells to oxidant stress, we have been unable to confirm such findings for o~thalassaemic cells. Parasites were certainly capable of invading, developing and multiplying in thalassaemic cells equally well as in normal red cells. It could be hypothesized that by exhibiting reduced cytoadherence, thalassaemic red cells could protect against malaria. Using a well recognized in vitro model of cytoadherence, we could establish no difference in the ability of infected thalassaemic or normal red cells to bind to C32 amelanotic melanoma cells. Finally, in preliminary work using a semiquantitative microagglutination assay, we demonstrated a greater degree of agglutination of parasitized thalassaemic cells than control [16]. We therefore proceeded to quantitate the binding of IgG from ES to the surface of thalassaemic and normal red ceils. This appeared to be particularly relevant as previous work had shown that antibodies to antigens exposed on the surface of infected cells were the most likely to correlate with any degree of protective immunity [17]. We found that in the mild form of ec-thalassaemia, infected ceils bound more antibody than controls which, on a surface area for area basis, was approximately between one and a half and twice normal. There is some evidence that homozygous/3-thalassaemic cells show enhanced phagocytosis by mouse macrophages when compared to normal red cells [18]. Furthermore there appears to be a signifi-
cant amount of IgG bound to the red cells of patients with homozygous thalassaemia which appears to be due to a naturally-occurring IgG antibody with anti-ot-galactosyl specificity [19]. Preliminary evidence has suggested that there is increased phagocytosis of oL-thalassaemic red ceils infected with P. falciparum [61. Our results would indicate that there is increased binding of antibody obtained both from immune and normal serum to the surface of the thalassaemic infected cell as compared to control, and that this continues to increase with parasite maturation well beyond the stage at which agglutination and surface immunofluorescence appear to plateau [20]. Whilst our experiments indicate increased expression of neoantigen on the surface of thalassaemic red cells, the identity of this molecule is as yet undetermined. The nature of parasite proteins on the surface of the infected cell are poorly defined and only one, P f EMP 1, has been characterized to any degree [20]. Our evidence has also suggested that a determinant is generated on the surface of the infected cells which is recognized by non-immune serum. This may be directed towards a modified host membrane component, for example modified band 3 [21], for which antibodies are said to exist in normal serum [22]. Previous work has suggested that a terminal o~-galactosyl residue may be involved in the immune recognition of thalassaemic red cells [19]. Our studies would also suggest that non-immune serum contains antibodies that recognize determinants on the infected red cell surface which are exposed or synthesized as the parasite matures. Overall our data would indicate that the mechanism whereby thalassaemia protects against malaria does not lie in reduced invasion or growth within these variant cells, nor would they support the hypothesis that parasites within these cells are more susceibtible to oxidant stress. Moreover parasitized thalassaemic and normal red-'cells are equally capable of binding to amelanotic melanoma cells. Instead we have found that there is a quantitative difference in the ability of infected o~-thalassaemic red cells to bind antibody from both immune and non-immune serum when compared to controls. This might form the basis for the selective advantage of thalassaemia in areas endemic for P. faiciparum malaria. Thus protection against malaria in
thalassaemia might well involve immune mechanisms, a suggestion which was recently made for sickle cell trait [23].
Acknowledgements This work was supported by the Wellcome Trust and the Medical Research Council of Great Britain. We thank Dr Y. Richards (Manchester Blood Transfusion Service) for providing the radiolabelled anti-IgG antibodies and Dr. D. Higgs for the et-thalassaemia genotyping. We would also like to thank Miss Dina Shah for typing of the manuscript and to our many colleagues at Oxford for their helpful comments and support during the period of this work. G.A.L., C.I.N., K.M. and G.P. were supported by the Wellcome Trust.
References [1] Nagel, R. L. (1989) Blood 74, 1213. [2] Flint, J., Hill, A.V.S., Bowden, D.K., Oppenheimer, S.J., Sill, P.R., Serjeantson, S.W., Bana-Koiri, J., Bhatia, J., Alpers, M. P., Boyce,A. J., Weatherall, D.J. and Clegg, J. G. (1986) Nature 321,744. [3] Oppenheimer, S. J., Higgs, D. R., Weatherall, D. J., Barker, J. and Spark, R. A. 0984) Lancet i, 424. [4] Modiano, G., Morpurgo, G., Terrenato, L., Novelletto, A., Di Rienzo, A., Colombo, B., Purpura, M., Mariani, M., Santachiara-Benerecetti, S., Brega, A., Dixit, K.A., Shresha, S. L., Lania, A., Wanachiwanawin, W. and Luzzatto, L. (1991) Am. J. Hum. Genet. 48, 390. [5] Friedman, M. J. (1979) Nature 280, 245. [6] Brockelman, C.R., Wongsattayanont, B., Tan-ariya, P. and Fucharoen, S. (1987) J. Clin. Microbiol. 25, 56. [7] Ifediba, T.C., Stern, A., Ibrahim, A. and Rieder, R.F. (1985) Blood 65, 452. [8] Yuthavong,Y., Butthep, P., Bunyaratvei, A., Fucharoen, S. and Khusmith, S. 0988) Am. J. Clin. Pathol. 89, 521. [9] Luzzi, G. A., Torii, M., Aikawa, M. and Pasvol, B. (1990) Br. J. Haematol. 74, 519. [10] Trager, W. and Jensen, J. B. (1976) Science 193,673. [11] Chulay, J. D., Haynes, J. D. and Diggs, C. L. (1983) Exp. Parasitol. 55, 138. [12] Marsh, K., Marsh, V.M., Brown, J., Whittle, H.C. and -- Greenwood,~vMv-(1988)Exp. Parasitol. 65,202. [13] Merry, A. H., Thomson, E. E., Rawlinson, V. I. and Stratton, F. (1982) Clin. Lab. Haematol. 4, 393. [14] Pasvol, B., Wilson, R. J. M., Smalley, M.E. and Brown, J. (1978) Ann. Trop. Med. Parasitol. 72, 87. [15] Pasvol, G., Weatherall, D. J. and Wilson, R. J. M. (1978) Nature 274, 701. [16] Luzzi, G.A., Merry, A. H., Newbold, C.I., Marsh, K., Pasvol, G. and Weatherall, D. J. (1991) J. Exp. Med. 173, 785. 239
[17] Marsh, K., Otoo, L., Hayes, R.J., Carson, D.C. and Greenwood, B.M. (1989) Trans. R. Soc. Trop. Med. 83, 293. [18] Knyszynski, A., Danon, D., Kahane, I. and Rachmilowitz, E.A. (1983) Br. J. Haematol. 43,251. [19] Galili, U., Korkesh, A., Kahane, A. E. and Rachmilowitz, E. A. (1983) Blood 61, 1258.
240
[20] Howard, R. J. (1988) Prog. Allergy 41, 98. [21] Sherman, I. W. and Winograd, E. (1990) Parasitol. Today 6, 317. [22] Kay, M. M. B., Flowers, N., Goodman, J. and Bosman, G. (1989) Proc. Natl. Acad. Sci. 86, 5834. [23] Bayoumi, R.A. (1987) Med. Hypoth. 22, 287.
Protection by a-thalassaemia against Plasmodium falciparum malaria: modified surface antigen expression rather than impaired growth or cytoadherence G. A. Luzzi 1, A. H. Merry 2, C. I. Newbold 1, K. Marsh I and G. Pasvol 3 1Molecular Parasitology Group, Institute of Molecular Medicine, John Radcliffe Hospital Oxford, 2Oxford Glycosystems Ltd., Unit 12, Abingdon, and 3St Mary's Hospital Medical School Department of Infectious Diseases and Tropical Medicine, Lister Unit, Northwick Park Hospital Middlesex, U.K. (Received 18 June 1991; accepted 25 June 1991)
I. Summary We have attempted to determine the cellular mechanism by which a-thalassaemia may protect against Plasmodium falciparum malaria. Invasion and development of P. falciparum in the microcytic red cells of two-gene deletion forms of a-thalassaemia when measured morphologically or by [3H]hypoxanthine incorporation were normal compared to controls. Normal invasion rates were also observed following schizogony in thalassaemic red cells. Neither the addition of the oxidant menadione, 30% oxygen, nor modified medium, produced differential damage to parasites within thalassaemic cells. Furthermore, there were no significant differences in the binding of P. falciparum-parasitized o~-thalassaemic and normal cells to C32 melanoma cells in vitro. However, when neoantigen expression on the surface of infected thalassaemic cells was estimated using a quantitative radiometric antiglobulin assay, clear differences were observed. It was found that ot-thalassaemic cells bound higher levels of antibody from serum obtained from individuals living in a malaria endemic area than control normal red cells. The binding ratio for thalasKey words: Malaria; Plasmodium falciparum; Thalassaemia; Neoantigen
Correspondence to: Professor G. Pasvol, St Mary's Hospital Medical School, Department of Infectious Diseases and Tropical Medicine, Lister Unit, Northwick Park Hospital, Middlesex, HA1 3U J, U.K.
saemic compared with controls was 1.69 on a cellfor-cell basis, and 1.97 when related to surface area. The binding of antibody from immune serum increased exponentially during parasite maturation. We also found increased binding of naturally occurring antibody present in non-immune serum to parasitized thalassaemic red cells which also increased during parasite maturation. We conclude that the protection afforded by thalassaemia against malaria may not reside in the ability of parasites to enter, grow or cytoadhere to endothelium in such cells, but may be related to immune recognition and subsequent clearance of parasitized red cells. 2. Introduction In studying the mechanism by which inherited red cell variants provide resistance against malaria, most studies have examined the ability of parasites to invade and grow in a particular cell type under given conditions [1]. In this context the mechanism whereby a-thalassaemia protects against malaria has remained particularly elusive despite the compelling epidemiological evidence that exists to support such a claim [2]. The gene frequency for this haemoglobinopathy may be as high as 80°-/o in certain parts of the world where malaria is or was endemic [3, 4]. It has been suggested that the relative susceptibility of thalassaemic cells (especially in 3-thalassaemia) to oxidant stress is responsible for the failure of parasites to survive in them [5]. Such oxidant stress has been achieved in vitro either
0165 - 2478 / 91 / $ 3.50 © Elsevier Science Publishers B.V. All rights reserved.
233
by addition of an oxidant such as menadione or riboflavine, or by modification of the growth medium [6]. Whilst growth may be impaired in cells from patients with the more severe forms of o~-thalassaemia (e.g., haemoglobin H disease) [7], and infected c~-thalassaemia red cells may be more susceptible to phagocytosis [8], we have not observed significant growth impairment in the milder forms of o~-thalassaemia under standard growth conditions [9]. We therefore set out not only to re-examine the effects of oxidant stress on parasite growth in thalassaemic red cells, but also to determine the ability of thalassaemic cells to bind to amelanotic melanom a cells (an in vitro model of cytoadherence) and finally to measure the expression of neoantigen on their surface when compared with normal red cells using a quantitative radiometric antiglobulin assay.
cycle was determined in order to investigate whether parasites grown in the variant cells were capable of giving rise to a further brood of parasites. In a number of experiments the oxidant menadione (Sigma) was added to the culture medium and in some a mixture containing 30% oxygen as previously used was used to gas the mixture, whilst in a further set a modified culture medium ( " R E M " ) was used as previously described [6]. This combined medium contains nine volumes of minimal essential medium (Gibco) and one volume of RPMI-1640, and no added L-glutamine. It therefore has only one-tenth of the reduced glutathione of the standard medium (RPMI-1640). All media were carefully pre-gassed with the appropriate gas mixture before use in experiments.
3. Materials and Methods
The in vitro model for sequestration using the C32 amelanotic melanoma cells was used as previously described [12]. Red cell suspensions containing a-thalassaemic or control ceils were incubated in triplicate on melanoma monolayers in 35-ram Petri dishes for 40 rain at 37 °C. After careful washing, monolayers with adherent parasitized ceils were fixed with two percent glutaraldehyde and stained with Giemsa stain. The number of adherent parasitized ceils per 50 melanoma cells were counted.
3.1. Parasite culture Strains of Plasmodium falciparum were maintained in continuous culture by conventional methods [10]. Parasite growth was regularly synchronised by sorbitol lysis and gelatin flotation or Percoll gradient purification. Whole blood was collected into acid citrate-dextrose (ACD) (1:10, v/v) stored at 4 °C and used within 48 h of collection. Samples were taken from individuals with two-gene deletion forms of a-thalassaemia (heterozygous oPthalassaemia ( - - / c w 0 , or homozygous o~+-tha lassaemia ( - c ~ / - a ) ) , and appropriate controls. Sickle cell trait and G 6 P D deficiency were excluded in all samples. 3.2. Invasion and growth assay These were carried out in triplicate in 96-well microtitre plates as previously described [9]. Cultures were incubated at 37 °C in sealed gas jars containing either a low oxygen gas mixture (1% 0 2, 5% CO2, 94°70 N2) or as indicated. After invasion had occurred, smears were prepared to assess invasion rates and a further incubation was carried out to determine maturation by morphological assessment or by [3H]hypoxanthine uptake [11]. In some experiments the rate of invasion in the following 234
3.3.
Cytoadherence assay
3.4. Quantitative estimation of lgG binding to red cells This was carried out by a method adapted from that described by Merry et al. [13]. Briefly, cultures of P. falciparum were established in thalassaemic and control cells. A fraction rich in trophozoites and schizonts (layer) was obtained by gelatin flotation [14] and the sedimented cells (pellet) were retained as a control. Such a pellet contained virtually no parasites. Both the layer and pellet were washed three times in wash buffer (WB) (PBS with 0.5% BSA; Sigma). For each sample, equal suspensions of infected and uninfected red cells were incubated with endemic serum (ES) obtained from healthy individuals living in an area highly endemic for malaria (The Gambia) or WB, washed, and then incubated with 100/zl of a 1:5 dilution of sheep or rabbit
serum and 100 t~l 125I-anti-human IgG (20 #g/ml). These were incubated for a further 30min at 37 °C and then washed three times in PBS. The red cells were resuspended in 120/~1 of wash buffer and a small aliquot removed to count the red cells in a Coulter counter. 100/~1 of the red cell suspension was spun through oil (4:1 di-n-butyl/dinonylphthalate; Sigma). The red cell pellets were cut off and counted in a gamma counter. The following formula was used to calculate the IgG bound (molecules per cell): IgG bound = (cpm/sra) x ( 6 z 103/n); involving a factor to convert/~g IgG to molecules of IgG/107 cells, where cpm equals counts per minute/100 ml; sra = specific radioactivity of 125I-IgG in counts per minute//~g IgG; and n = cell count/ 100/~lx 10 -7. To estimate IgG bound to parasitized cells, values for specific IgG bound (molecules/cell) for non-parasitized cells (pellet E s - pellet wB) were subtracted from the respective parasitized fraction results (layer Es - layerWB), and the result was divided by parasitaemia ( x 10-2). The value for non-parasitized cells was then added. Thus the number of IgG molecules bound per parasitized cell could be calculated. A further correction was made in order to relate the numbers of IgG bound to the surface area of the cells. This was carried out using the formula SA = 44.8 + 1.17 x MCV; where SA = the surface area #m2; and MCV = mean cell volume. 4. Results
4.1. Invasion and development in ~-thalassaemic cells Rates of invasion of P. falciparum (Uganda Palo Alto strain, UPA and a Thai cloned line, T9/96) into thalassaemic cells were the same as in controls. Moreover parasite development assessed both morphologically and by [3H]hypoxanthine incorporation indicated no difference between control and thalassaemic red ceils (Table 1). The invasion rates into the next cycle following schizogony within thalassaemic cells was also found to be completely normal (Table 1). When invasion and development of parasites was carried out in the presence of 0, 1, 2, 5 or 10ttmol of menadione progressive growth inhibition was observed. However, there was no difference between the normal and ct-thalassaemic
cells (Fig. 1). When these experiments were repeated using conditions of 30°70 oxygen, a further reduction in parasite incorporation of [3H]hypoxanthine was seen (Fig. 2). The invasion and growth of parasites in ot-thalassaemic and normal red cells were not different. Parasite development in the modified medium (REM) was slightly inhibited (+88070 of [3H]hypoxanthine uptake in standard medium) but the extent of this inhibition was no greater in thalassaemic cells (data not shown). 4.2. Cytoadherence assays No difference was observed in the ability of parasitized thalassaemic or normal cells to bind to amelanotic melanoma cells using the ITO (Brazil) parasite strain (Fig. 3). 4.3. Quantitative binding of IgG to infected tha-
lassaemic and normal red cells In these experiments using strain IT (Brasil), parasitized a-thalassaemic red cells consistently bound more IgG than controls (Fig. 4). The mean binding ratios of thalassaemic versus normal cells was 1.69 on a cell-for-cell basis and 1.97 when calculated per unit surface area. When infected cells were incubated in WB rather than ES, even though the amount of antibody was considerably less than that observed with ES, parasitized thalassaemic cells still appeared to bind more antibody when compared to controls (Fig. 5). The mean binding ratio of thalassaemic versus normal red ceils was 1.29 on a cellfor-cell basis, and 1.58 when calculated per unit red cell surface area. When the number of molecules of IgG bound to uninfected cells were determined, no such difference in binding between the thalassaemic and control ceils was observed (Table 2). Finally, the binding of IgG molecules during maturation of parasites within normal red cells incubated in either ES or WB buffer was determined. Results showed a steady increase from 12 h into the parasite life cycle until well into schizont maturation (Fig. 6). This increase applied to cells incubated in both ES and WB.
235
TABLE 1 Invasion, growth and reinvasion of P. falciparum in ~-thalassaemic compared to normal red cells. Genotype
--
- - / O t ~
--
--
Invasion (% control)
Growth (% control)
102.5 _+ 4.1 107.9 +_ 4.4 100.8 4-_ 1.7
/OZO(
O r / - - O~
(Control): (Range)
Invasion: Growth: Reinvasion:
Reinvasion (% control)
Morphology
3H uptake
101.7 _+ 4.5 100.2 + 4.0 101 _+ 2.0
104.3 _+ 1.0 96.9 +_ 1.0 106.5 _+ 2.0
101.8 +_ 4.6 117.8 _+ 7.0 102.2 _+ 10.7
2 - 17 Rings/100 cells 6 6 - 86 Schizonts per 100 singly infected red cells 1.54 to 4.85 × 103 counts per minute 2 - 13 Rings per 100 red cells
6.0 1"3 0
== o (.)
5.0 4.0
- - ~ '
3.0 2.0
--/c'.o. []
--IChor.
•
control
•
control
1.0 0.0
'
0.0
"
'
I
2.0
•
'
'
I
'
'
4.0
I
"
'
6.0
'
I
8.0
=
'
'
I
'
10.0
)uM Menadlone Fig. 1. The effect of menadione on the growth o f P . falciparum in c~-thalassaemic or control red cells as measured by [3H]hypoxanthine incorporation. Error bars show S.E. for triplicate values. 18 16 ~o
14
x
12
5%
g
•
: .::-.::::: 8
[]
30%
-{:d-o. conffol
. . . . . . . . . "[3" .......
---OJ-~..z
......... •~ .......
control
6 4 2 0
|
0.0
5.0
i
10.0
pM m e n a d l o n e Fig. 2. The effect of 30% oxygen on the growth of P. falciparum in ePthalassaemic or control red cells in the presence of increasing concentrations of menadione as measured by [3H]hypoxanthine incorporation. Error bars show S.E. for triplicate counts. 236
120 " o
o=
100 =
[]
80
[]
'~
--/~a
(I)
--/c~a (21
'~
60
[]
control (1)
.e.~
40
[]
control (2)
.2
-
.2
L
2o
0
1
5 10 Parasites added (xl0 ~) Fig. 3. Bindingto C32 melanoma cells of increasing numbers of P. fal¢iparum-infected a-thalassaemic and control red cells. Error bars show S.E. for 7 experiments.
o~ "0 e-
-1
40000]
30000-
0
20000 o
[]
control
[]
O. -thai
O
E (9
10000
0' 1
2
3
4
5
Expt. Fig. 4. Antibody detected on P. falciparum-infected c~-thalassaemic and control cells (molecules/100/zm 2) after incubation in endemic
sera (ES). Error bars show S.E. for triplicate values.
5. Discussion Whilst for m a n y of the haemoglobinopathies the epidemiological evidence for protection by them against fatal P. falciparum malaria is convincing, the molecular mechanisms by which these are brought about remain unclear. Moreover there is increasing evidence that these mechanisms might differ from condition to condition [1]. However, the way in which thalassaemic red cells might protect against malaria has remained most elusive. Whilst for sickle haemoglobin the failure of para-
sites to develop in such cells when maintained under reduced oxidant tension provides a rational basis for the way in which sickle haemoglobin might protect against malaria [15], for thalassaemia the susceptibility of parasite to oxidant stress has remained the favoured mechanism for protection in this disorder. Reduced growth of parasites within thalassaemic cells when exposed to menadione, increased oxygen tensions [5], and modified medium containing reduced amounts of antioxidants [6], would all support this mechanism of protection. However, in rigorous experiments using these vari237
6000 0
-e
5000
o
..~
4000 []
--
[]
3000
control cc -thai
o
o E
2000
0
1000 o 1
2
3
4
5
Expt. Fig. 5. Antibody detected on P. falciparum-infected c~-thalassaemic and control cells (molecules/100 ~m 2) after incubation in wash buffer (WB). Error bars show S.E. for triplicate values. TABLE 2 IgG detected on uninfected c~-thalassaemic and control cells after incubation in endemic serum or wash buffer. Control
c~-Thalassaemia
P
516 ± 85 345 ± 50
306 _+ 85 287 ± 39
0.26 NS
310 ± 31 207 ___ 20
220 _+ 21 172 _+ 16
NS NS
Endemic serum
Molecules per cell Molecules per 100/tin 2 Wash buffer
Molecules per cell Molecules per 100 # m 2 1ooooo
o
ES
-o
o
100(30
-~ E
looo
r=0.92
0 •
100 6
12
18
hours
, , ,
24 Into
30
36
42
48
life-cycle
Fig. 6. Antibody detected on P. falciparum-infected normal red cells 6-hourly during a complete asexual cycle in vitro using tightly synchronized parasites incubated in endemic serum (ES) or wash buffer (WB). Points represent mean values for triplicate samples (SE values within 10% for each point), r, correlation coefficient for exponential fit.
238
ous methods of exposing cells to oxidant stress, we have been unable to confirm such findings for o~thalassaemic cells. Parasites were certainly capable of invading, developing and multiplying in thalassaemic cells equally well as in normal red cells. It could be hypothesized that by exhibiting reduced cytoadherence, thalassaemic red cells could protect against malaria. Using a well recognized in vitro model of cytoadherence, we could establish no difference in the ability of infected thalassaemic or normal red cells to bind to C32 amelanotic melanoma cells. Finally, in preliminary work using a semiquantitative microagglutination assay, we demonstrated a greater degree of agglutination of parasitized thalassaemic cells than control [16]. We therefore proceeded to quantitate the binding of IgG from ES to the surface of thalassaemic and normal red ceils. This appeared to be particularly relevant as previous work had shown that antibodies to antigens exposed on the surface of infected cells were the most likely to correlate with any degree of protective immunity [17]. We found that in the mild form of ec-thalassaemia, infected ceils bound more antibody than controls which, on a surface area for area basis, was approximately between one and a half and twice normal. There is some evidence that homozygous/3-thalassaemic cells show enhanced phagocytosis by mouse macrophages when compared to normal red cells [18]. Furthermore there appears to be a signifi-
cant amount of IgG bound to the red cells of patients with homozygous thalassaemia which appears to be due to a naturally-occurring IgG antibody with anti-ot-galactosyl specificity [19]. Preliminary evidence has suggested that there is increased phagocytosis of oL-thalassaemic red ceils infected with P. falciparum [61. Our results would indicate that there is increased binding of antibody obtained both from immune and normal serum to the surface of the thalassaemic infected cell as compared to control, and that this continues to increase with parasite maturation well beyond the stage at which agglutination and surface immunofluorescence appear to plateau [20]. Whilst our experiments indicate increased expression of neoantigen on the surface of thalassaemic red cells, the identity of this molecule is as yet undetermined. The nature of parasite proteins on the surface of the infected cell are poorly defined and only one, P f EMP 1, has been characterized to any degree [20]. Our evidence has also suggested that a determinant is generated on the surface of the infected cells which is recognized by non-immune serum. This may be directed towards a modified host membrane component, for example modified band 3 [21], for which antibodies are said to exist in normal serum [22]. Previous work has suggested that a terminal o~-galactosyl residue may be involved in the immune recognition of thalassaemic red cells [19]. Our studies would also suggest that non-immune serum contains antibodies that recognize determinants on the infected red cell surface which are exposed or synthesized as the parasite matures. Overall our data would indicate that the mechanism whereby thalassaemia protects against malaria does not lie in reduced invasion or growth within these variant cells, nor would they support the hypothesis that parasites within these cells are more susceibtible to oxidant stress. Moreover parasitized thalassaemic and normal red-'cells are equally capable of binding to amelanotic melanoma cells. Instead we have found that there is a quantitative difference in the ability of infected o~-thalassaemic red cells to bind antibody from both immune and non-immune serum when compared to controls. This might form the basis for the selective advantage of thalassaemia in areas endemic for P. faiciparum malaria. Thus protection against malaria in
thalassaemia might well involve immune mechanisms, a suggestion which was recently made for sickle cell trait [23].
Acknowledgements This work was supported by the Wellcome Trust and the Medical Research Council of Great Britain. We thank Dr Y. Richards (Manchester Blood Transfusion Service) for providing the radiolabelled anti-IgG antibodies and Dr. D. Higgs for the et-thalassaemia genotyping. We would also like to thank Miss Dina Shah for typing of the manuscript and to our many colleagues at Oxford for their helpful comments and support during the period of this work. G.A.L., C.I.N., K.M. and G.P. were supported by the Wellcome Trust.
References [1] Nagel, R. L. (1989) Blood 74, 1213. [2] Flint, J., Hill, A.V.S., Bowden, D.K., Oppenheimer, S.J., Sill, P.R., Serjeantson, S.W., Bana-Koiri, J., Bhatia, J., Alpers, M. P., Boyce,A. J., Weatherall, D.J. and Clegg, J. G. (1986) Nature 321,744. [3] Oppenheimer, S. J., Higgs, D. R., Weatherall, D. J., Barker, J. and Spark, R. A. 0984) Lancet i, 424. [4] Modiano, G., Morpurgo, G., Terrenato, L., Novelletto, A., Di Rienzo, A., Colombo, B., Purpura, M., Mariani, M., Santachiara-Benerecetti, S., Brega, A., Dixit, K.A., Shresha, S. L., Lania, A., Wanachiwanawin, W. and Luzzatto, L. (1991) Am. J. Hum. Genet. 48, 390. [5] Friedman, M. J. (1979) Nature 280, 245. [6] Brockelman, C.R., Wongsattayanont, B., Tan-ariya, P. and Fucharoen, S. (1987) J. Clin. Microbiol. 25, 56. [7] Ifediba, T.C., Stern, A., Ibrahim, A. and Rieder, R.F. (1985) Blood 65, 452. [8] Yuthavong,Y., Butthep, P., Bunyaratvei, A., Fucharoen, S. and Khusmith, S. 0988) Am. J. Clin. Pathol. 89, 521. [9] Luzzi, G. A., Torii, M., Aikawa, M. and Pasvol, B. (1990) Br. J. Haematol. 74, 519. [10] Trager, W. and Jensen, J. B. (1976) Science 193,673. [11] Chulay, J. D., Haynes, J. D. and Diggs, C. L. (1983) Exp. Parasitol. 55, 138. [12] Marsh, K., Marsh, V.M., Brown, J., Whittle, H.C. and -- Greenwood,~vMv-(1988)Exp. Parasitol. 65,202. [13] Merry, A. H., Thomson, E. E., Rawlinson, V. I. and Stratton, F. (1982) Clin. Lab. Haematol. 4, 393. [14] Pasvol, B., Wilson, R. J. M., Smalley, M.E. and Brown, J. (1978) Ann. Trop. Med. Parasitol. 72, 87. [15] Pasvol, G., Weatherall, D. J. and Wilson, R. J. M. (1978) Nature 274, 701. [16] Luzzi, G.A., Merry, A. H., Newbold, C.I., Marsh, K., Pasvol, G. and Weatherall, D. J. (1991) J. Exp. Med. 173, 785. 239
[17] Marsh, K., Otoo, L., Hayes, R.J., Carson, D.C. and Greenwood, B.M. (1989) Trans. R. Soc. Trop. Med. 83, 293. [18] Knyszynski, A., Danon, D., Kahane, I. and Rachmilowitz, E.A. (1983) Br. J. Haematol. 43,251. [19] Galili, U., Korkesh, A., Kahane, A. E. and Rachmilowitz, E. A. (1983) Blood 61, 1258.
240
[20] Howard, R. J. (1988) Prog. Allergy 41, 98. [21] Sherman, I. W. and Winograd, E. (1990) Parasitol. Today 6, 317. [22] Kay, M. M. B., Flowers, N., Goodman, J. and Bosman, G. (1989) Proc. Natl. Acad. Sci. 86, 5834. [23] Bayoumi, R.A. (1987) Med. Hypoth. 22, 287.