This case-control (2:3 ratio) study was carried out from March 2022 to May 2023. Study participants were pregnant women living in the urban and sub-urban areas of Yaoundé, Cameroon, who attended the health centers for delivery. A total of 50 peripheral, placental, and cord whole blood were collected from each participant (21 placental malaria positive and 29 negative) at delivery at the Marie Reine Health Center of Etoudi and at the Catholic Hospital of Marie des Anges of Koabang. The two healthcare centers are located in two outlying districts of Yaoundé where malaria transmission is perennial, with 2 wet and 2 dry seasons. Information on the mother’s age, mother’s health, gravidity, parity, gestational age, use of the intermittent preventive treatment with sulfadoxine-pyrimethamine during pregnancy (IPTp-SP), use of insecticide-treated bed nets (ITN), and baby’s weight at delivery were recorded in a standard questionnaire. The participants with pre-existing health conditions (e.g., preeclampsia, diabetes, toxoplasmosis, hepatitis, syphilis and HIV positive results) were excluded from the study. Birth weights lower than 2.5 kg were considered low as per the World Health Organization’s guidelines (29). Maternal venous, placental and cord blood samples were aseptically collected at delivery in BD Vacutainer heparinized tubes. A fragment of placental tissue was collected from the maternal side of the placenta immediately following delivery and used to diagnose placental malaria (PM). Blood samples were processed on the day of collection.

Thick blood smears from maternal, placental and cord blood samples, and impression smears from placental tissue were stained using Giemsa and examined by at least two skilled microscopists for the presence of malaria parasites. Plasmodium falciparum infection was further confirmed using the Commercialized One Step HRP-II and pLDH RDT rapid diagnostic test (SD Bioline malaria antigen P.f/Pan, Standard Diagnostics Inc, Kyonggi-do, Korea). Participants were considered placental malaria positive if infected erythrocytes were detected in impression smears of placenta tissue and/or placental or peripheral blood smears. Hemoglobin (Hb) levels in maternal and neonate cord blood were determined using a hemoglobinometer (ACON Laboratories, INC, San Diego, USA). Participants were considered anemic if Hb < 11 g/dl whereas neonates were considered anemic if Hb < 12,5 g/dl (30).

Isolated cells consist of Peripheral Blood Mononuclear Cells (PBMC) from the mother, Intervillous Blood Mononuclear Cells (IVBMC) from the placental and Cord Blood Mononuclear Cells (CBMC) from the cord. Briefly, approximately 15 ml of freshly collected peripheral, placental or cord blood samples were each diluted 1:1 in 1X PBS (without Ca²⁺ and Mg²⁺) supplemented with 2% FBS (Thermo Fisher Scientific, USA) and then layered onto 15ml Histopaque 1077 (Sigma-Aldrich, USA) in a 50 ml Falcon tube. Tubes were centrifuged for 30 minutes at 800g and 20°C without brake. The PBMCs, IVBMCs and CBMCs layers were recovered, washed twice with 1X PBS (without Ca²⁺ and Mg²⁺) + 2% FBS by centrifugation at 600 g and 20°C for 8 minutes. The viability of the recovered cells was determined by Trypan blue dye exclusion method. Only samples with cell viability greater than 95% were used in the assay. Approximately 1×10⁶ of isolated PBMCs, IVBMCs or CBMCs were re-suspended in 1mL of 1X PBS for surface staining.

The PBMCs, IVBMCs and CBMCs staining were done in 96-well Microplate V shaped bottom. The cells were washed and re-suspended in 50 µl of FACS buffer (1X PBS without Ca²⁺ and Mg²⁺, 5% FBS, 5mM EDTA) with human FcR Blocking Reagent (Miltenyi Biotech, USA) and incubated for 10 min at 4°C in the dark. After washing, the pellet was re-suspended in 50 µl of antibody staining mix and incubated for 30 min at 4°C in the dark. Cells were then washed with 200 µl of FACS buffer at 1800 rpm for 3 min and re-suspended in FACS buffer prior to acquisition. The following monoclonal antibodies were used for surface staining: Brilliant Violet (BV) 510 labelled anti-CD3 (clone UCHT1) (BD Biosciences, USA); Phycoerythrin (PE) labelled anti-TCRγδ PE (clone B1); Allophycocyanin-Cyanine 7 (APC-Cy7) labelled anti-TCR V delta 2 (clone B6), BV421 labelled anti-TIM-3 (clone F38–2E2), APC labelled anti-PD1 (clone EH12.2H7) (Biolegend, USA); PE-Cy7 labelled anti-TCR V delta 1 (clone TS8.2) and Fluorescein Isothiocyanate (FITC) labelled anti-HLA-DR (clone L243) (ThermoFisher Scientific, USA). Samples were acquired within 2 hours of staining on a BD FACS Canto II flow cytometer equipped with a BD FACSDiva software version 6.1.3 (Becton Dickinson, USA).

Flow cytometry data were analyzed using FlowJo software version 10.8.1 (Tree Star, Inc) and OMIQ. Color compensation was performed using beads stained for each fluorochrome. Only samples from which at least 100 000 CD3⁺ single cells were acquired were included in the analysis. A lymphocyte gate was set based on SSC-A/FSC-A, followed by single cell gating based on FSC-H/FSC-A, and CD3⁺ events within this gate analyzed. Unsupervised flow cytometry analysis of TCRγδ⁺ cells in PBMC, IVBMC and CBMC samples were done using Uniform Manifold Approximation Projection (UMAP) along with the FlowSOM automated clustering tool of OMIQ. 5000 TCRγδ⁺ cells for each study were concatenated and dimensionality reduction was assessed using the UMAP (15 nearest neighbors, 0.4 minimum distance). For clustering visualization, we used FlowSOM (6 metaclusters). Markers considered in the analyses included TCR Vδ1, TCR Vδ2, HLA-DR, PD1 and TIM-3.

Statistical analyses were performed using GraphPad Prism version 9.4.1. Results were reported as means with standard deviation or medians with interquartile ranges. Comparisons between two groups were assessed using Mann-Whitney rank sum test. Spearman rank order correlation coefficient (r_s) was used to evaluate associations. Proportions were compared using Fisher’s exact test. P values <0.05 were considered statistically significant.

Samples of 29 women without placental malaria (PM-) and 21 women with placental malaria (PM+) were used in the present study. The basic characteristics of the study population are summarized in Table 1 . Mean age, median parity and median gravidity of PM- women were not significantly different from those of PM+ women (p=0.343; p=0.086 and p=0.535 respectively). Median maternal hemoglobin level was significantly lower (p=0.002) in PM+ women (10.67 g/dL) compared to PM- women (12.67 g/dL). Moreover, the prevalence of maternal anemia was significantly higher (p=0.003) in PM+ women (52.38%) compared to PM- women (10.71%). Median gestational age was not significantly different between PM- and PM+ women (p= 0.212). No significant difference (p=0.186) was observed in fetal hemoglobin level between PM+ women (15.66 g/dL) and PM- women (14.50 g/dL). A similar trend was observed with the prevalence of fetal anemia: 50% in PM- women and 38.10%) in PM+ women (p=0.563). Baby weight at delivery was significantly lower for PM+ women (3,045g) compared to PM- women (3,353g) (p=0.014). In terms of proportion of women who took IPTp-SP or used ITNs during pregnancy, no significant difference (p>0.999 and p=0.683, respectively) was observed between PM- women (88.89% and 82.14), and PM+ women (85.71% and 90.48%).

γδ T cells and its subsets were analyzed using the gating strategy shown in Figure 1 and Supplementary Figure 1 . The frequency of total γδ T cells, Vδ1⁺, Vδ2⁺ and Vδ1^-Vδ2^- subsets in PBMC, IVBMC, and CBMC between PM- and PM+ women are presented in ( Figure 2 ) and the correlations in ( Figure 3 ; Tables 2 – 4 ). We observed no significant difference in proportions of cells expressing each marker type, comparing the infected and uninfected samples, reason for which only gMFIs data were presented in the manuscript ( Tables 2 – 4 ). The frequency of total γδ T cells in IVBMC and of the Vδ1⁺ subset in PBMC and IVBMC were significantly higher in PM+ women compared to PM- women (p=0.018, p=0.030 and p=0.022, respectively) ( Figures 2A, B ). Although the frequency of total γδ T cells in PBMC was high in PM+ women compared to PM- women, the difference was not significant (p=0.495). Furthermore, the frequency of total γδ T cells in IVBMC correlated positively with parasitemia from placenta tissue impression smear and placental blood parasitemia [(r_s =0.366; p=0.011) and (r_s =0.330; p=0.023, respectively)] ( Figure 3 and Table 3 ). Indeed, the frequency of the Vδ1⁺ subset correlated positively with parasitemia from placenta tissue impression smears in PBMC (r_s =0.300; p=0.036) and with parasitemia from placenta tissue impression smears, peripheral and placental blood parasitemia in IVBMC (0.313 ≤ r_s ≤ 0.347; 0.017 ≤ p ≤ 0.034) ( Figure 3 ; Tables 2 , 3 ). In contrast, the frequency of the Vδ2⁺ subset in PBMC and IVBMC were significantly lower in PM+ compared to PM- women (p=0.020 and p=0.021, respectively) ( Figure 2C ). In addition, the frequency of the Vδ2⁺ subset in PBMC, IVBMC correlated negatively with parasitemia from placenta tissue impression smear and peripheral blood parasitemia (-0.404 ≤ r_s ≤ -0.313; 0.005 ≤ p ≤ 0.028) ( Figure 3 ; Tables 2 , 3 )., suggesting that P. falciparum placental infection leads to a selective loss of Vδ2⁺ subset. A similar trend in the Vδ1^-Vδ2^- subset in PBMC was observed between the two groups of women although not significant (p=0.672) ( Figure 2D ). Regarding the frequency of total γδ T cells, Vδ1⁺, Vδ2⁺ and Vδ1^-Vδ2^- subsets in CBMC, no significant difference was observed between PM+ and PM- women (p=0.353; p=0.644; p=0.792 and p= 0.897, respectively) ( Figures 2A–D , 3 , Table 4 ). Together, these data indicate that they are significant differences in the expansion of Vδ1⁺ and Vδ2⁺ subsets in PBMC and IVBMC during P. falciparum placental infection.

The HLA-DR marker is often used as an activation and/or antigenic presentation phenotype of γδ T cells. In the present study, HLA-DR expression within total γδ T cells, Vδ1⁺ and Vδ1^-Vδ2^- subsets in CBMC were significantly higher in PM+ compared to PM- women (p=0.009, p=0.005 and p=0.017 respectively) ( Figures 4A, B, D ; Supplementary Figure 2 ). Positive correlations were observed between parasitemia from placenta tissue impression smear and HLA-DR expression within total γδ T cells in CBMC (r_s=0.332; p=0.023), within Vδ1⁺ subset in IVBMC and CBMC (r_s=0.344, p=0.019 and r_s=0.386, p=0.008, respectively), and within Vδ1^-Vδ2^- in CBMC (r_s=0.353; p=0.016) ( Figure 3 ; Tables 3 , 4 ).

These data suggest that the absence of direct contact with the parasites lead to expression of HLA-DR within total γδ T cells, Vδ1⁺ and Vδ1^-Vδ2^- subsets.

Similar but non-significant trends in frequencies between PM+ and PM- women were observed for HLA-DR expression within total γδ T cells and its subsets in PBMC and IVBMC (0.060 ≤ p ≤ 0.763) ( Figures 4A, D ). In general, the levels of HLA-DR expression within total γδ T cells and its subsets in PBMC, IVBMC positively correlated with parasitemia from placenta tissue impression smear, placental and peripheral blood ( Figure 3 ; Tables 2 , 3 ).

Additionally, HLA-DR expression within total γδ T cells, Vδ2⁺ and Vδ1^-Vδ2^- subsets in PBMC correlated positively with parasitemia from placenta tissue impression smear, placental blood and peripheral blood parasitemia (0.299 ≤ r_s≤ 0.377; 0.008 ≤ p ≤ 0.037 for total γδ T cells, (0.322 ≤ r_s ≤ 0.327, 0.022 ≤ p ≤ 0.024 for Vδ2⁺ and 0.318 ≤ r_s ≤ 0.362, 0.011 ≤ p ≤ 0.026 for Vδ1^-Vδ2). Meanwhile, in the Vδ1⁺ subset, HLA-DR expression correlated with parasitemia only from the placental tissue impression smear and placental blood (0.295 ≤ r_s≤ 0.330; 0.021 ≤ p ≤ 0.039) ( Figure 3 and Table 2 ). Furthermore, HLA-DR expression within Vδ1⁺ subset in IVBMC correlated positively with placental blood parasitemia (r_s=0.318; p=0.031). These data are consistent with the hypothesis that the parasites induced T cell activation and/or antigenic peptide presentation by γδ T cells.

To further characterize the phenotype of γδ T cells in the context of placental malaria, expression of the exhaustion marker PD1 was compared between PM- and PM+ women ( Figure 5 ; Supplementary Figure 3 ). PD1 expression within total γδ T cells and Vδ1^-Vδ2^- subset in PBMC as well as that of Vδ2⁺ subset in PBMC and IVBMC was significantly higher in PM+ women compared to PM- women (p=0.035 and 0.005) and p=0.002 and 0.003, respectively) ( Figures 5AC, D ). Although not significant, PD1 expression within total γδ T cells and Vδ1^-Vδ2^- subset in IVBMC and within Vδ1⁺ subset in PBMC and IVBMC was higher in PM+ compared to PM- women (p=0.094 and 0.088, and p=0.276 and 0.302), respectively) ( Figures 5A, B, D ). No significant difference in PD1 expression between PM+ and PM- women was observed for total γδ T cells, Vδ1⁺, Vδ2⁺ and Vδ1^-Vδ2^- subsets in CBMC (0.392 ≤ p ≤ 0.601) ( Figures 5A–D ). The levels of PD1 expression within total γδ T cells and its subsets in PBMC, IVBMC associated with parasitemia from placenta tissue impression smear, placental and/or peripheral blood ( Figure 3 ; Tables 2 , 3 ). Indeed, PD1 expression within total γδ T cells in PBMC correlated positively with parasitemia from placenta tissue impression smear and placental blood parasitemia (r_s= 0.305; p=0.031 and r_s= 0.283; p=0.046, respectively) and borderline significance with parasitemia from a placental tissue impression smear in IVBMC (r_s=0.283; p=0.054) ( Figure 3 ; Tables 2 , 3 ). PD1 expression within Vδ2⁺ subset in PBMC and IVBMC correlated positively with parasitemia from placenta tissue impression smear, placental blood and peripheral blood (0.441 ≤ r_s ≤ 0.560, 0.00002 ≤ p ≤0.001 and 0.329 ≤ r_s ≤ 0.437, 0.002 ≤ p ≤0.026, respectively). Regarding PD1 expression within Vδ1^-Vδ2^- subset, results showed a positive correlation with parasitemia from placenta tissue impression smear in PBMC and IVBMC (r_s= 0.387, p=0.006 and r_s= 0.305, p=0.037, respectively). A similar trend was observed with placental blood parasitemia in PBMC (r_s=0.319, p=0.024), and in CBMC (r_s=0.314, p=0.030) ( Figure 3 ; Tables 2 – 4 ). These data suggest that placental malaria leads to up-regulation of PD1 within total γδ T cells and its subsets.

The activation and function of γδ T cells can be regulated by the expression of immunoregulatory markers. In this study, expression of the exhaustion marker TIM-3 within total γδ T cells, Vδ1⁺, Vδ2⁺ and Vδ1^-Vδ2^- subsets in PBMC, IVBMC and CBMC was significantly higher in PM+ compared to PM- women (0.0001 ≤ p ≤ 0.016) ( Figures 6A–D ; Supplementary Figure 4 ). As presented in Figure 3 and Tables 2 – 4 , TIM-3 expression within total γδ T cells in PBMC, IVBMC, and CBMC positively correlated with different parasitemia from placenta tissue impression smear, placental blood and peripheral blood parasitemia (0.408 ≤ r_s ≤ 0.522, 0.0002 ≤ p ≤ 0.007). Similar trends were observed for its two subsets (Vδ1⁺, Vδ2⁺) in IVBMC (0.318 ≤ r_s ≤ 0.347, 0.017 ≤ p ≤ 0.034 for Vδ1⁺ subset and 0.395 ≤ r_s ≤ 0.541, 0.0001 ≤ p ≤ 0.007 for Vδ2⁺ subset), and for Vδ1^-Vδ2^- subset in PBMC and IVBMC (0.435 ≤ r_s ≤ 0.561, 0.00004 ≤ p ≤ 0.003). TIM-3 expression on the two subsets (Vδ1⁺ and Vδ2) also correlated positively with parasitemia from placenta tissue impression smear and placental blood parasitemia in PBMC and CBMC (0.332 ≤ r_s ≤ 0.394, 0.005 ≤ p ≤ 0.023) and in CBMC for Vδ1^-Vδ2^- subset (r_s= 0.395, p= 0.006 and r_s= 0.404; p= 0.005; respectively) ( Figure 3 and Tables 2 – 4 ). Together, these data indicate that placental malaria leads to increased expression of the TIM-3 immunoregulatory markers within total γδ T cells.

In the present study, we investigated the relationship between pregnancy outcomes and the γδ T cells phenotype in PBMC, IVBMC and CBMC ( Figure 7 ; Supplementary Tables 1 – 3 ). In PM+ women, maternal hemoglobin levels correlated negatively with the frequency of total γδ T cells (r_s=-0.468; p=0.032), and with expression levels of the exhaustion marker TIM-3 within total γδ T cells, Vδ1⁺, Vδ2⁺, Vδ1^-Vδ2^- subsets in IVBMC (-0.601 ≤ r_s ≤ -0.452, 0.004 ≤ p ≤ 0.040). In addition, maternal hemoglobin levels correlated negatively with expression of the activation marker HLA-DR within total γδ T cells and Vδ2⁺ subset (r_s=-0.473, p=0.035 and r_s=-0.550, p=0.012, respectively) ( Figure 7 ; Supplementary Table 2 ). In contrast, maternal hemoglobin levels correlated positively with the frequency of Vδ1^-Vδ2^- subsets in PBMC (r_s=0.560, p=0.008). Fetal hemoglobin levels correlated positively with PD1 expression within total γδ T cells in PBMC and within Vδ2⁺ subset in IVBMC (r_s=0.507, p=0.019) ( Figure 7 ; Supplementary Tables 1 , 2 ). Regarding birth weight, a negative correlation was observed with the frequency of total γδ T cells in PBMC (r_s=-0.606; p=0.006) and the PD1 expression within Vδ2⁺ subset in CBMC (r_s=-0.501, p=0.034). On the other hand, HLA-DR expression within Vδ2⁺ subset in IVBMC correlated positively with birth weight (r_s=0.489, p=0.040) ( Figure 7 ; Supplementary Tables 1 , 3 ). Together, these results suggest that in placental malaria, the expansion of total γδ T cells as well as its subsets and the expression of the activation and exhaustion markers may be associated with poor pregnancy outcomes or newborn protection.

Cell profile is often associated with maternal age, parity and gestational age. In this study, we examined the relationship between these parameters and the γδ T cells phenotype in PBMC, IVBMC and CBMC ( Figure 8 ; Supplementary Tables 4 – 6 ). In PM+, maternal age correlated negatively with TIM-3 expression within total γδ T cells and Vδ1^-Vδ2^- subset in CBMC (r_s=-0.454, p=0.045 and r_s=-0.477, p=0.034, respectively]. HLA-DR expression within total γδ T cells, Vδ1⁺, and Vδ1^-Vδ2^- subsets in CBMC correlated negatively with gestational age (r_s=-0.513, p=0.025, r_s=-0.541, p=0.017 and r_s=-0.470, p=0.042, respectively) ( Figure 8 ; Supplementary Table 6 ).

Regarding parity, the frequency of Vδ1⁺ subset in IVBMC correlated negatively with parity (r_s=-0.603, p=0.005). PD1 expression within Vδ2⁺ subset in PBMC and IVBMC as well as within total γδ T cells and Vδ1^-Vδ2^- subsets in IVBMC correlated negatively with parity (-0.598 ≤ r_s ≤ -0.498, 0.004 ≤ p ≤ 0.022). In addition, TIM-3 expression within total γδ T cells and Vδ1^-Vδ2^- subset correlated negatively with parity (r_s=-0.497, p=0.026 and r_s=-0.513, p=0.021, respectively). The data suggest that expansion of γδ T cells may be affected by some of the above studied parameters.