We showed earlier in Papua that trading Plasmodium falciparum Severe malaria is caused at a younger age than uncomplicated malaria1. Thompson Locke et al.2 He later claimed that in our and other datasets, prevalence of parasites in the blood is inversely correlated with the estimate of parasite age and that this is due to Plasmodium falciparum that causes severe malaria express more cell-coherent PfEMP1s, resulting in early parasite sequestration in the microvasculature and reduced spleen clearance. Here, we show that in our data set, disseminated parasitemia and the proportion of total disseminated parasites are not associated with disseminated parasite age, and therefore our data does not support their hypothesis.
Thompson Locke et al. Our discovery confirmed that rolling Plasmodium falciparum Parasites in severe malaria are smaller than those that cause uncomplicated malaria1 They suggested that this difference confounded comparisons between versions of the parasites that cause severe and uncomplicated malaria. This is precisely the reason why we developed and applied mathematical approaches to control parasite stage variability prior to identifying upregulated genes in severe malaria by differential gene expression (DGE) analysis. It appears that Thomson-Luque et al. These identified details were missed in our methods and results. Thus, they may have misinterpreted the set of differentially expressed genes in their reanalysis. Thompson Locke et al. They themselves showed that the genes we identified as upstream in severe malaria were not expressed before the upregulated genes in uncomplicated malaria (Thomson-Luque et al. Fig. 3a,b). Thus, our analysis of parasite stage change that did not confound the set of differentially expressed genes was successfully controlled for.
Thomson-Luque showed that the genes upregulated in severe malaria in our differentially expressed genes were not expressed before the genes upregulated in uncomplicated malaria. They claimed that this similarity in timing of expression was due to the lack of parasite difference between severe malaria and uncomplicated malaria patients. In our paper we analyzed 44 samples of patients for us I was de novo gene assemblies and comparison of these parasite densities, but only 35 of them were used for DGE due to pre-acceptance drug therapy or insufficient sequence coverage1. Parasitological incidences were significantly higher in the 16 severe malaria cases used in DGE (median 2.071%, IQR 0.422-12.83) than in the 19 cases of uncomplicated malaria used in DGE (median 0.59% IQI 0.092-1.18) s= 0.0136 Mann-Whitney two-sided test yo = 78. Thus, because we did not provide individual parasite densities by Thomson-Luque et al. Understandably but incorrectly assumed that the samples used for the DGE did not also differ in parasitemia.
Thompson Locke et al. Our admixture model was used to estimate parasite stage in multiple datasets and showed inverse associations between these estimates of parasite age and prevalence. They concluded that the separate parasite load is associated with disseminated parasitism and thus inversely correlated with the age of the disseminated parasite (Thomson-Luque et al., Figure 5). However, our data does not support these associations. We did not introduce parasitism to individual patients in our publication but reanalyzed our data and compared the proportions of parasite stages in our samples with parasitism. There is no association in our data set between blood parasites and the proportion of ring stage parasites (Spearman .). s = 0.2033 95% CI −0.1494 equals 0.5101 s = 0.2415) or asexual asexual parasites estimated by our mixture model (Spearman s= −0.2088 95% CI −0.5143 equals 0.1438 s = 0.2288) (Fig. 1). Thus, in our samples, the parasitemia in the blood is not related to the age of the parasite.
In Figure 1c, Thompson-Lock et al. Rank 41 of our patient samples by returning RNAseq numbers of a single glycine tRNA ligase gene PF3D7_1420400 rather than individual parasites. This gene is described as a housekeeping gene and its expression level is inferred to represent blood parasites. However, in 41 samples with sufficient sequence coverage for expression analysis from Tonkin-Hill et al.1 No association between normal reading levels of PF3D7_1420400 and parasitism (Spearman) s = −0.1908 95% CI −0.4687 equals 0.1213, s = 0.2147). There is a relationship between the levels of the measured readouts of PF3D7_1420400 and the ratios of the phases of the ring (Spearman s = 0.4934, 95% confidence interval 0.1432 to 0.6634, s = 0.004) or the ratio of asexual asexual phases (Spearman s = −0.4608 95% CI −0.6782 to 0.1693 s = 0.0024). That the normal glycine tRNA ligase ligase reading has a positive correlation with the ratio of loop stages and a negative relationship with the ratio of asexual non-cyclic phases indicates that glycine tRNA ligase transcript levels better reflect the youth of the parasite than the parasite, consistent with the majority of RNAseq data sets were prepared using the same approach. Like our own and available on plasmodb3And the4And the5And the67. Thus, the inferred relationship between raw glycine tRNA ligase reads and circulating parasitism in Tonkin-Hill et al. It was incorrect and the observed trends in gene expression by Tonkin Hill et al. Samples in Figure 1c are from Thomson-Luque et al. It is not associated with an increase in parasites in the blood.
Thompson Locke et al. used staging of disseminated parasite copies to infer relationships between parasite stage and disseminated parasite density and suggested that early isolation occurs in hyperparasitaemia/severe malaria. However, we present data indicating that in our samples this was not the case using the estimated total biomass of the parasites (Pto) calculated from HRP2 levels. PtoBoth sequestered and disseminated parasites are estimated8 Thus it directly measures the removal of parasites from the circulation by cellular loading rather than inferring cellular tolerance from differences in circulating parasitemia. PtoIt can also be influenced by variation in parasite reproduction rate, duration of infection, magnitude of distribution, and interindividual variation in the half-life of PfHRP2, but it is nonetheless a univariate association with clinical outcome and prognostic indicators of severity than for peripheral parasitism.8. PtoCan be calculated for 21 uncomplicated and 16 severe malaria samples1. √Ptot was higher in severe malaria (mean ± SEM 2,132,578 ± 399,093) than in uncomplicated malaria (mean ± SEM 1,090,126 ± 133,654) in these samples (two-sided without pairs). R-a test s= 0.0094, R= 2.749, df = 35, the raw Ptot was skewed to the right, so the data were converted to square root and then normality was confirmed by D’Agostino & Pearson’s normal test). However, the circulating parasitemia/Ptot ratio, that is, the relative measure of the proportion of circulating parasites that make up the total parasite load in the body, did not match the proportion of circulating parasites that were spearman rings. s= −0.0767 95% CI −0.3994 equals 0.2629, s= 0.6520) or asexual acyclic stages (Spearman s= 0.0923 95% CI −0.2483 equals 0.4125, s= 0.5871) (Fig. 2). This highlights that in our data set, the proportion of total parasites isolated had no association with the stage of disseminated parasites. These results do not support the model proposed by Thomson-Luque of elevated circulating parasitemia leading to severe disease due to earlier isolation and therefore younger disseminated parasites, which we would expect to appear as a lower proportion of total circulating parasites in patients with smaller circulating parasites.
Our data support part of the hypothesis of Thomson-Luque et al. , in particular the established role of PfEMP1 sequestration in severe disease. While this was the main focus of our study, Thomson-Luque et al. Our manuscript has been interpreted as reports I wasGene expression was reduced in severe cases when we actually reported that “there is no difference between severe malaria and uncomplicated malaria in the whole gene expression of var”1. Thompson Locke et al. He appears to have misinterpreted our statement that “I wasGene expression has been modified,” which indicates the histone methyltransferase involved in it I wasGene silencing and reduced switching in severe malaria cases. Thus, we were referring to the potential change I wasAltering or silencing genes, which may affect I wasGenes were expressed but not the total level of I wasgene expression.
In summary, we have previously reported that the parasites prevalent in severe malaria were younger than those in uncomplicated malaria.1. Our interpretation of these data differs significantly from that of Thomson-Luque et al. Because in our data there is no relationship between parasitaemia (Fig. 1) nor direct evidence for the proportion of the total parasites that were infested (Fig. 2) with the age of the disseminated parasite. While the data of Thomson-Luque et al. Of potential importance in understanding the pathogenesis of malaria, our data do not support the hypothesis of Thomson-Luque et al. , which was based on the indirect inference of parasite-carrying separated from circulating blood parasites. Both our study and that of Thomson-Luque et al. Bulk RNAseq data were used to estimate disseminated parasite life cycle stage distributions. This allows general conclusions that parasite lifespans vary but full accuracy of the life cycle stage distribution will require single-cell RNAseq analyses, the expression of which can also be improved I wasgenetic associations. Identification of the expressed PfEMP1 surface will require the development of detection reagents for the PfEMP1 variant.
More information about research design is available in the Nature Research Reporting Summary linked to this article.