Antigen-Stimulated PBMC Transcriptional Protective Signatures For Malaria Immunization

SCIENCE TRANSLATIONAL MEDICINE | RESEARCH ARTICLE
MALARIA Copyright © 2020

The Authors, some
Antigen-stimulated PBMC transcriptional protective rights reserved;
exclusive licensee
signatures for malaria immunization American Association
for the Advancement
of Science. No claim
Gemma Moncunill1,2*, Anja Scholzen3, Maximillian Mpina4,5,6, Augusto Nhabomba2, to original U.S.
Aurore Bouyoukou Hounkpatin7,8, Lourdes Osaba9, Raquel Valls10, Joseph J. Campo1,2, Government Works
Hèctor Sanz1, Chenjerai Jairoce2, Nana Aba Williams1, Erica M. Pasini11, David Arteta9,
Joan Maynou9, Lourdes Palacios9, Miquel Duran-Frigola12, John J. Aponte1,
Clemens H. M. Kocken11, Selidji Todagbe Agnandji7,8, José Manuel Mas10,
Benjamin Mordmüller8, Claudia Daubenberger5,6, Robert Sauerwein3, Carlota Dobaño1,2*
Identifying immune correlates of protection and mechanisms of immunity accelerates and streamlines the develop-
ment of vaccines. RTS,S/AS01E, the most clinically advanced malaria vaccine, has moderate efficacy in African
Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

children. In contrast, immunization with sporozoites under antimalarial chemoprophylaxis (CPS immunization)
can provide 100% sterile protection in naïve adults. We used systems biology approaches to identifying correlates
of vaccine-induced immunity based on transcriptomes of peripheral blood mononuclear cells from individuals
immunized with RTS,S/AS01E or chemoattenuated sporozoites stimulated with parasite antigens in vitro. Specifically,
we used samples of individuals from two age cohorts and three African countries participating in an RTS,S/AS01E
pediatric phase 3 trial and malaria-naïve individuals participating in a CPS trial. We identified both preimmunization
and postimmunization transcriptomic signatures correlating with protection. Signatures were validated in
independent children and infants from the RTS,S/AS01E phase 3 trial and individuals from an independent CPS
trial with high accuracies (>70%). Transcription modules revealed interferon, NF-B, Toll-like receptor (TLR), and
monocyte-related signatures associated with protection. Preimmunization signatures suggest that priming the
immune system before vaccination could potentially improve vaccine immunogenicity and efficacy. Last, signatures
of protection could be useful to determine efficacy in clinical trials, accelerating vaccine candidate testing.
Nevertheless, signatures should be tested more extensively across multiple cohorts and trials to demonstrate
their universal predictive capacity.
INTRODUCTION CVac) are promising candidates (4) that have shown up to 100%
Malaria remains a major public health problem, causing an estimated efficacy in controlled human malaria infection (CHMI) trials in
218 million cases and 405,000 deaths in 2018 (1). Although a malaria-naïve adults. However, efficacy under natural exposure or
vaccine is considered a crucial tool in combating this infectious in a pediatric population has not been evaluated extensively and
disease (2), decades of research have thus far only resulted in a may be lower compared to CHMI (5).
single subunit vaccine candidate, RTS,S/AS01E, which has been Major hurdles in the development of an effective malaria vaccine
recommended by the World Health Organization for pilot imple- are the absence of immune correlates of protection and understand-
mentation studies starting in 2019 in three African countries. The ing of the mechanisms of protective immunity among other diffi-
efficacy of RTS,S/AS01E against clinical malaria in a phase 3 trial for culties, including the complexity of the Plasmodium parasite’s cycle
licensure over 12 months of follow-up was moderate, about 31% in and its high polymorphism (4, 6). Efficacy testing of each vaccine
infants of 6 to 12 weeks and about 56% in children of 5 to 17 months candidate in humans requires conducting complex, long, and ex-
(3). Vaccines based on attenuated Plasmodium falciparum sporozoites pensive clinical trials, in which protection is assessed by subjecting
through irradiation (PfSPZ vaccine) and chemoprophylaxis (PfSPZ- vaccinated individuals to experimental (CHMI) or natural P. falciparum
challenge. Identification of in vitro surrogate markers of protection
1
would simplify clinical trial design and accelerate the development
ISGlobal, Hospital Clínic–Universitat de Barcelona, E-08036 Barcelona, Catalonia,
Spain. 2Centro de Investigação em Saúde de Manhiça (CISM), Rua 12, Cambeve, Vila
of new or improved vaccines. The best correlate thus far of RTS,S/
de Manhiça, CP 1929 Maputo, Mozambique. 3Department of Medical Microbiology, AS01E is the titer of immunoglobulin G (IgG) against the vaccine
Radboud University Medical Center, 6500 HB Nijmegen, Netherlands. 4Ifakara antigen (7), a region of the P. falciparum circumsporozoite protein
Health Institute, Bagamoyo Research and Training Centre. P.O. Box 74, Bagamoyo, (CSP). IgG against CSP targets the sporozoites and is thought to
Tanzania. 5Swiss Tropical and Public Health Institute, Socinstrasse 57, 4002 Basel,
Switzerland. 6University of Basel, Petersplatz 1, 4001 Basel, Switzerland. 7Centre de prevent liver-stage infection and subsequent infection of red blood
Recherches Médicales de Lambaréné (CERMEL), BP 242 Lambaréné, Gabon. 8Institute cells (RBCs). However, no IgG threshold for protection has been
of Tropical Medicine and German Center for Infection Research, University of identified. IgG to CSP is also associated with protection induced by
Tübingen, Wilhelmstraße 27, D-72074 Tübingen, Germany. 9Progenika Biopharma.
A Grifols Company, S.A., 48160 Derio, Vizcaya, Spain. 10Anaxomics Biotech, S.L,
irradiated PfSPZ in malaria-naïve adults (4) but not in preexposed
08008 Barcelona, Spain. 11Department of Parasitology, Biomedical Primate populations (5).
Research Centre, Rijswijk, Netherlands. 12Institute for Research in Biomedicine (IRB A powerful model to dissect immune correlates of sterile,
Barcelona). The Barcelona Institute of Science and Technology, 08028 Barcelona, preerythrocytic immunity to malaria is the chemoprophylaxis and
Catalonia, Spain.
*Corresponding author. Email: gemma.moncunill@isglobal.org (G.M.); carlota. sporozoite (CPS) immunization regimen (4, 8), which inspired the
dobano@isglobal.org (C.D.) PfSPZ-CVac. In PfSPZ-CVac, purified sporozoites previously
Moncunill et al., Sci. Transl. Med. 12, eaay8924 (2020) 13 May 2020 1 of 17
cryopreserved are given as direct venous inoculation to volunteers on a to detect acquired immune responses induced by immunization
prophylactic regimen of chloroquine. In CPS immunization, fresh spo- and mechanisms of protection. We stimulated PBMC in vitro using
rozoites are delivered by repeated bites of infected mosquitoes to a CSP peptide pool to assess preerythrocytic responses and whole
volunteers also on a prophylactic regimen of chloroquine (8). Opti- asexual blood-stage parasites [P. falciparum–infected RBCs (PfRBCs)].
mal dosing confers 100% protection in malaria-naïve adults (4, 8), This strategy was chosen to increase the chances to detect antigen-
similar to PfSPZ-CVac. Suboptimal dosing of CPS immunization specific responses to CPS (due to the large overlap between late liver-
results in different degrees of protection, ranging from sterile protec- stage and blood-stage antigens) and to evaluate naturally acquired
tion to delayed patency or no protection compared to malaria- immunity in African participants. CSP- and PfRBC-stimulated gene
naïve infection controls (9, 10). This heterogeneity and the controlled expression was background-corrected with the gene expression profile
conditions, in comparison with trials in endemic areas where vac- of PBMCs from the same participant and time point treated with
cinees are subject to natural exposure at unknown time points, make dimethyl sulfoxide (DMSO; present in the peptide pools) and unin-
the CPS regimen a useful tool to study malaria immunity. fected RBCs (uRBCs), respectively, to control for participant-specific
Currently, an emerging approach based on omics technologies differences. Thus, the differential gene expression measured (Fig. 1D)
and systems biology analyses is being used to decipher immune was specifically induced by the antigen stimulations.
signatures that predict vaccine immunogenicity (11) and has the

potential to also identify signatures of protection (12) and provide Differential gene expression associated with CPS
insights into mechanisms relevant for vaccine efficacy. Successful immunogenicity
studies have identified transcriptional signatures induced early after We found 60 genes significantly up-regulated [q < 0.1 and |fold
immunization that correlate with and predict the later adaptive im- change (FC)| > 1.5] and 18 down-regulated after CPS immunization
mune responses in humans for vaccines targeting yellow fever virus compared to preimmunization upon PfRBC recall but no signifi-
(13), influenza (14, 15), meningococcus (16), HIV (17), and even for cant differences upon CSP stimulation (data file S1). Many of the
RTS,S (12). However, only a few studies have investigated signa- changes corresponded to genes involved in immune responses.
tures of protection, and these were performed precisely in RTS,S Among the 10 top up-regulated genes were IFNG (interferon gamma)
and CPS CHMI studies in malaria-naïve adults (12, 18–20). In this (FC = 4.68 and q < 0.0001) and GZMB (granzyme B) (FC = 2.39 and
study, we used peripheral blood mononuclear cell (PBMC) samples q < 0.0001), which have been previously associated with CPS immuno-
from volunteers (malaria-naïve adults) immunized by CPS (9, 10) genicity (9), IL-22 (FC = 6.64 and q < 0.0001), CISH (FC = 2.23 and
and from naturally exposed African children and infants who par- q < 0.0001), CCL4 (FC = 2.14 and q < 0.0001) and LTA (lymphotoxin
ticipated in the RTS,S/AS01E phase 3 trial (3), combined with a sys- alpha; FC = 1.98 and q < 1.71 × 10−6). No differential gene expression
tems biology approach to identifying gene signatures of protective was identified comparing protected with unprotected volunteers
immunity against malaria. either before or after immunization (data file S1).
Consistently, principal components analysis (PCA) using FC
expression of CSP/DMSO and PfRBC/uRBC stimulations did not
RESULTS reveal any differences by immunization, dose, sex, or protection in
CPS and RTS,S cohorts for the identification of the CPS cohort. Samples showed some clustering only by stimulation
candidate signatures condition (PfRBC/uRBC versus CSP/DMSO) (fig. S4). Similarly,
In the CPS trial (9), 24 malaria-naïve adults were immunized with unsupervised hierarchical clustering and heatmaps did not reveal
suboptimal doses of P. falciparum–infected mosquitoes under a any clustering by study conditions besides stimulation (fig. S5).
chloroquine prophylaxis. After 5 months, they were challenged CPS immunization often leads to detectable parasitemia after
with parasites through the bites of infected mosquitoes (Fig. 1A and the immunizing mosquito bites (9), especially after the first dose,
fig. S1). Sterile protection was observed in 17 out of 24 volunteers. In and an antibody and cellular recall response to PfRBCs (21, 22) may
the African phase 3 trial, we selected a subset of 255 infants and be related to this initial blood-stage exposure. We assessed the
children who received three doses of RTS,S/AS01E or a comparator correlation of PfRBC/uRBC FC gene expression in CPS-immunized
vaccine (rabies vaccine for children and meningococcal C conjugate volunteers with cumulative parasitemia during the immunization
vaccine for infants) in three sites (Bagamoyo in Tanzania, Lambaréné period to identify genes associated with blood-stage parasite expo-
in Gabon, and Manhiça in Mozambique). On the basis of a case-control sure and immunogenicity. Many genes had | values| > 0.4 and P <
design, we included all children who had clinical malaria episodes 0.05, but none of them were significant after adjusting for multiple
during a follow-up period of 12 months after the third primary testing (data file S2). In addition, some nonsterilely protected indi-
vaccination (n = 50; Fig. 1B and fig. S1B) and selected up to four viduals showed partial protection (longer prepatency period) (9). We
nonmalaria controls for each malaria case. After sample processing, found only 10 and 4 significant genes after adjusting for multiple
we ended up analyzing all 24 CPS-immunized volunteers (fig. S2) testing (q < 0.1), correlating with time to positive quantitative real-time
and 178 phase 3 trial vaccinees: 127 RTS,S individuals (24 nonpro- polymerase chain reaction (qRT-PCR) and thick blood smear
tected and 70 protected) and 51 comparator individuals (7 nonpro- prepatency, respectively (data file S2). Annotation of genes for both
tected and 32 protected) (fig. S3). cumulative parasitemia and prepatency analyses revealed immune
We analyzed the transcriptomic profile of in vitro P. falciparum responses that are likely to be antigen-specific (data file S2).
recall responses for CPS volunteers both before immunization and
just before challenge with infected mosquitoes and for African Lack of differential gene expression associated
volunteers before immunization (only children) and 1 month after with RTS,S/AS01E
RTS,S immunization. This approach (Fig. 1C) was selected instead Gene expression of RTS,S-vaccinated individuals was analyzed in
of ex vivo transcriptomic responses because it maximized the chances relation to comparator vaccinees and not contrasting to the same
A Sporozoite immunization under CQ prophylaxis (CPS) B RTS,S/AS01E multicenter phase 3 African vaccine trial
N = 24 N = 255
Adults 17/24 protected Clinical follow-up 50/205 non protected
4/5 in the 3 × 15 group 35/176 RTS,S
0 3 8 8/9 in the 3 × 10 group 0 3 12 15/79 comparator
Months Month
5/10 in the 3 × 5 group
5–17months
Immunization doses with Pf mosquitoes: 5–17months Vaccine doses with RTS,S (n = 176)
5 (n = 10), 10 (n = 9) or 15 (n = 5) 6–12weeks or comparator (n = 79)
5–17months
Blood samples 6–12weeks Blood samples
Pf challenge
C Experimental design D Computational analysis (detection of biomarkers)

Selection of ~70
Dif. gene expr. Gene set enrichment
PBMC stimulations Gene expression genes for validation
microarrays involved in the
CSP and signatures of
PBMCs 24 hours
DMSO protection

Microarrays
data
PfRBCs and Protein network
16 hours Data science Proposed signatures
uRBCs (TPMS) of protection
(3–5 protein
combinations)
>70% accuracies
0 5 10
Fig. 1. Study design for the identification of signatures of protection. (A) Samples from a chemoprophylaxis and sporozoite (CPS) immunization clinical trial (9) and
(B) the RTS,S/AS01E vaccine phase 3 clinical trial (3) were used. The CPS trial involved immunization of malaria-naïve adult volunteers by bites from P. falciparum
(Pf)–infected mosquitoes during chloroquine (CQ) chemoprophylaxis. Three groups of volunteers received three different doses of bites from infected mosquitoes. After
infectious challenge, 17 of 24 volunteers were protected from infection. Blood samples were collected at baseline (before immunization) and 5 months after the last dose
(after immunization). Children and infants from three different African countries in the RTS,S/AS01 phase 3 trial received three doses of the RTS,S vaccine or a comparator
vaccine 1 month apart and were followed up for detection of clinical malaria episodes. Samples from 50 volunteers who had malaria (nonprotected) and 205 volunteers
who did not have any malaria episode (protected) were selected. Blood samples were collected at baseline for children and 1 month after third vaccine dose for children
and infants. (C) Previously cryopreserved PBMCs were stimulated in vitro with circumsporozoite protein (CSP) peptide pool or P. falciparum–infected red blood cells
(PfRBCs) and their respective background controls, dimethyl sulfoxide (DMSO), and uninfected red blood cells (uRBCs). Gene expression was measured by microarrays.
(D) Differential gene expression analysis using linear regression models, and subsequently, GSEA was performed. Microarray data were also used for protein network–
based models. Proteins behaving differently in the models were down-selected using data science methods, leading to identification of pre- and postimmunization
signatures of protection, consisting of three to five proteins, with accuracies of >70%. TPMS, Therapeutic Performance Mapping System.
individuals at preimmunization because these transcriptional data analyses, we used the lists of all genes regardless of statistical signif-
were available only for a subset of children and the comparator icance (data file S1) ranked according to their average expression
group allowed controlling for malaria exposure and age. There were FC in volunteers after CPS immunization relative to preimmunization
no significantly up- or down-regulated genes upon CSP or PfRBC for each recall response (considering its stimulation background control).
stimulation in RTS,S vaccinees in relation to comparators (data file A summary of all gene set results can be found in the data file S4.
S3). Similarly, no differential gene expression was detected in CPS immunization induced some common BTM activity upon
protected versus nonprotected individuals who received the RTS,S CSP and PfRBC stimulation including dendritic cell (DC) activa-
or the comparator vaccine (data file S3). PCA and unsupervised tion, T cells, natural killer (NK) cells and IFN/antiviral sensing
hierarchical clustering and heatmaps using FC expression of CSP/ (Fig. 2A). Two of the genes found in “signaling in T cells (M35.0
DMSO and PfRBC/uRBC stimulations did not reveal any differ- and M35.1)” BTMs were IFNG and GZMB, which were among the
ences between vaccine groups, time points, protection, site, age, and top up-regulated genes upon immunization and have been previ-
sex either (figs. S6 and S7), and samples were only clustered by ously associated with CPS (9). In addition, upon CSP stimulation,
stimulation condition (PfRBC/uRBC versus CSP/DMSO). BTMs related to B cells, inflammatory/Toll-like receptor (TLR)/
chemokines and signal transduction were up-regulated. These BTMs
Gene sets associated with CPS immunization and indicate both innate and acquired responses, probably related to
sterile protection antigen-specific responses, subsequent signaling amplification, and
We next performed gene set enrichment analysis (GSEA), which bystander activation. Curiously, for the PfRBC recall response
allows us to investigate sets of genes when the individual gene asso- after CPS relative to preimmunization, there was a repression of
ciations are not strong. As gene sets, we used blood transcriptional BTMs of antigen presentation, DC surface receptors, monocytes
modules (BTMs) (16) that have been successfully applied in previ- and some BTMs of inflammatory/TLR/chemokines, and signal
ous systems vaccinology studies (12), to enable functional inter transduction. The simultaneous positive and negative enrichment
pretation of the transcriptional responses. We considered only the of related BTMs and the positively enriched module “immune
BTMs that were significantly enriched also using two alternative regulation–monocytes, T and B cells (M57)” may indicate immuno-
enrichment analyses, correlation-adjusted mean rank (CAMERA) modulatory and inhibitory processes in response to PfRBC stimulation
and Tmod, to increase the confidence in the findings. For all these in CPS-immunized volunteers.
A Immunogenicity B CSP PfRBC Protection negatively associated with protection

PfRBC
were found positively enriched at postim-

CSP
Post
Post
Pre
Pre
Enriched in antigen presentation (I) (M71)
Enriched in antigen presentation (III) (M95.1)
' Enriched in antigen presentation (III) (M95.1)
Enriched in antigen presentation (II) (M95.0)
munization compared to preimmunization;
Enriched in antigen presentation (II) (M95.0)
Regulation of antigen presentation and immune response (M5.0)
Enriched in antigen presentation (I) (M71)
Regulation of antigen presentation and immune response (M5.0)
Fig. 2B). Analysis before immunization
Enriched in activated dendritic cells / monocytes (M64)
Activated dendritic cells (M67)
Enriched in activated dendritic cells (II) (M165)
Activated dendritic cells (M67) revealed not only negative enrichment
Enriched in activated dendritic cells (II) (M165)
Chemokines and inflammatory molecules in myeloid cells (M86.0)
Myeloid, dendritic cell activation via NF-κB (I) (M43.0)
Proinflammatory dendritic cell, myeloid cell response (M86.1) of some of the same BTMs related to
Myeloid, dendritic cell activation via NF-κB (II) (M43.1) Myeloid, dendritic cell activation via NF-κB (II) (M43.1)
Myeloid, dendritic cell activation via NF-κB (I) (M43.0)
Proinflammatory dendritic cell, myeloid cell response (M86.1)
Complement and other receptors in dendritic cells (M40)
Enriched in dendritic cells (M168)
DC activation, chemokine clusters, and
Complement and other receptors in dendritic cells (M40)
Enriched in dendritic cells (M168)
Enriched in activated dendritic cells (I) (M119)
Chemokines and inflammatory molecules in myeloid cells (M86.0)
viral sensing in protected relative to
B cell surface signature (S2)
Enriched in B cells (I) (M47.0)
Enriched in activated dendritic cells/monocytes (M64)
Naïve B cell surface signature (S8)
unprotected volunteers but also positive
Enriched in B cells (II) (M47.1)
Naive B cell surface signature (S8)
Enriched in B cells (I) (M47.0)
BCR signaling (M54) enrichment of some other BTMs related
BCR signaling (M54)
T cell activation (II) (M7.3)
Enriched in B cells (III) (M47.2)
Plasma cells, immunoglobulins (M156.1) to T cells and NK cells.
In contrast, in CPS-immunized in-
Signaling in T cells (II) (M35.1) Plasma cells and B cells, immunoglobulins (M156.0)
Signaling in T cell (I) (M35.0) Signaling in T cells (I) (M35.0)
CD28 costimulation (M12) T cell activation (II) (M7.3)
CD4+ T cell surface signature T 2-stimulated (S7)
H
T cell activation and signaling (M5.1)
T cell activation (IV) (M52)
Signaling in T cells (II) (M35.1)
dividuals, a more robust response was
Cell cycle, mitotic phase (M230)
Mismatch repair (I) (M22.0)
Enriched in T cells (II) (M223)
T cell surface, activation (M36)
induced upon PfRBC recall compared
+
Mitotic cell cycle in stimulated CD4 T cells (M4.5) T cell differentiation (M14)
to CSP stimulation. Only positively en-

Cell cycle and transcription (M4.0) T cell activation (I) (M7.1)
Enriched in NK cells (II) (M61.0)
Enriched in NK cells (I) (M7.2)
Enriched in T cells (I) (M7.0)
Cell cycle and growth arrest (M31) riched BTMs in the protected individu-
NK cell surface signature (S1)
Enriched in NK cells (KIR cluster) (M61.1)
Cell cycle and transcription (M4.0)
Enriched in NK cells (I) (M7.2) als were found relative to unprotected
Enriched in NK cells (KIR Cluster) (M61.1)
volunteers (Fig. 2B). These BTMs were
Extracellular matrix (II) (M2.1)
MHC-TLR7-TLR8 cluster (M146) Enriched in NK cells (III) (M157)
Chemokine cluster (II) (M27.1) NK cell surface signature (S1)
Activated (LPS) dendritic cell surface signature (S11)
Chemokine cluster (I) (M27.0)
Enriched in NK cells (receptor activation) (M61.2)
Enriched in NK cells (II) (M61.0)
related to antigen presentation, DC
DC surface signature (S5)
TLR and inflammatory signaling (M16)
Extracellular matrix (II) (M2.1)
Activated (LPS) dendritic cell surface signature (S11)
activation, B cells, T cells, cell cycle,
Immune activation–generic cluster (M37.0)
Inflammatory response (M33)
MHC-TLR7-TLR8 Cluster (M146)
Chemokine cluster (I) (M27.0) extracellular matrix and migration,
Myeloid cell–enriched receptors and transporters (M4.3)
Resting dendritic cell surface signature (S10)
Chemokine cluster (II) (M27.1)
DC surface signature (S5) inflammatory/TLR/chemokines, signal
transduction, and monocytes. Several
Viral sensing and immunity/IRF2 targets network (I) (M111.0) Resting dendritic cell surface signature (S10)
Innate antiviral response (M150) Myeloid cell–enriched receptors and transporters (M4.3)
Innate activation by cytosolic DNA sensing (M13) TLR and inflammatory signaling (M16)
Type I interferon response (M127)
Antiviral IFN signature(M75)
Immune activation - generic cluster (M37.0)
Antiviral IFN signature (M75)
postimmunization BTMs that correlated
Putative targets of PAX3 (M89.0)
Regulation of signal transduction (M3)
Innate antiviral response (M150)
with protection were also correlated with
Enriched in monocytes (surface) (M118.1)
Enriched in monocytes (II) (M11.0)
Innate activation by cytosolic DNA sensing (M13)
Viral sensing and immunity/IRF2 targets network (I) (M111.0)
immunogenicity (Fig. 2B), although many
Enriched in monocytes (III) (M73)
Enriched in monocytes (IV) (M118.0)
Putative targets of PAX3 (M89.0)
Regulation of signal transduction (M3) of these were negatively enriched by CPS
Enriched in myeloid cells and monocytes (M81)
Monocyte surface signature (S4)
IL-2, IL-7, TCR network (M65)
Enriched in monocytes (SURFACE) (M118.1) immunization, which seems contradic-
Enriched in myeloid cells and monocytes (M81)
tory. Some of the BTMs associated with
Golgi membrane (II) (M237)
Immune regulation–monocytes, T and B cells (M57) Enriched in monocytes (IV) (M118.0)
Cell activation (IL-15, IL-23, TNF) (M24) Monocytes surface signature (S4)
AP-1 transcription factor network (M20)
Leukocyte differentiation (M160)
Enriched in monocytes (III) (M73)
protection in postimmunization sam-
Chaperonin-mediated protein folding (I) (M204.0)
Collagen, TGF-β family et al (M77)
Enriched in neutrophils (I) (M37.1)
Cell activation (IL-15, IL-23, TNF) (M24)
ples were already positively associated with
Lysosomal/endosomal proteins (M139)
Proteosome (M226)
Immune regulation—monocytes, T and B cells (M57)
Leukocyte differentiation (M160) protection in preimmunization samples
Purine nucleotide biosynthesis (M212)
Transcription regulation in cell development (M49)
AP-1 transcription factor network (M20)
Transcription regulation in cell development (M49) (Fig. 2B) such as the ones related to DC
activation, monocytes or inflammation,
Transmembrane transport (I) (M87)
Blood coagulation (M11.1)
Collagen, TGF-β family et al. (M77)
Endoplasmatic reticulum (M37.2)
Lysosomal/endosomal proteins (M139)
and TLR signaling. When looking at the
−3 −2 −1 0 1 2 3
FC in the expression of genes upon
Antigen presentation Cell cycle Interferon/antiviral sensing
CSP immunogenicity, opposite direction
PfRBC immunogenicity, same direction
DC activation
B cells
NK cells
ECM and migration
Signal transduction
Monocytes
PfRBC stimulation relative to uRBC
PfRBC immunogenicity, opposite direction T cells Inflammatory/TLR/chemokines Neutrophils background, we found that although
individuals had many genes more down-
Fig. 2. Transcriptional responses associated with CPS immunogenicity and protection. Each square represents
a blood transcription module (BTM). The color shading indicates normalized enrichment scores obtained by GSEA
regulated after immunization than be-
analysis for BTMs. Assignment of a BTM to a high-level annotation group is illustrated by a colored sidebar. GSEA, fore immunization, protected immunized
CAMERA, and Tmod were run with genes ranked by (A) the expression of postimmunization (Post) relative to pre- volunteers displayed consistently less down-
immunization (Pre) and (B) protected relative to nonprotected individuals for CSP and PfRBC recall stimulations after regulation than unprotected immunized
and before immunization. Modules that did not reach the significance cutoff of an FDR q value of 0.1 in all three volunteers (Fig. 3).
enrichment methods or a minimum of 10 matched genes were eliminated. Modules without annotation are not We also performed the gene set
shown. Modules that represent common associations of both immunogenicity and protection at postimmunization analyses using Kyoto Encyclopedia of
are highlighted with a circle for CSP recall responses and triangle for PfRBC recall responses, and filled symbols Genes and Genomes (KEGG) (23),
indicate that enrichment had the same direction, whereas empty symbols indicate that enrichment had the opposite
BioCarta, Reactome (24), and another set
direction. AP-1, activator protein-1; MHC, major histocompatibility complex; ECM, extracellular matrix.
of modules defined by Chaussabel et al.
(25). Results (data file S4) are also con-
Unexpectedly, gene set analysis comparing protected with non- sistent with the identified BTM responses (Fig. 2). For instance,
protected CPS-immunized individuals (Fig. 2B) upon CSP recall for CSP stimulations, gene sets positively enriched with immu-
showed several negatively enriched BTMs related to innate responses nization included IFN responses (Reactome IFN- signaling, IFN-
(antigen presentation, DC activation, inflammatory/TLR/chemokines,  signaling, IFN signaling, or regulation of IFN- signaling;
and IFN/antiviral sensing among others) and also some BTMs Chaussabel’s modules M1.2, M3.4, and M5.12) or responses related
related to acquired immune responses (B and T cells). There was a to T cells [BioCarta cytotoxic T cells, IL-22BP (interleukin-22BP),
large overlap between BTMs associated with protection as assessed TH1TH2, or IL-2RB pathways]. These gene sets were negatively en-
postimmunization on the one hand, and immunogenicity on the riched in protected compared to unprotected CPS-immunized vol-
other hand, although in the opposite direction (28 of the 34 BTMs unteers. For PfRBC stimulation, there were also many enriched
A Myeloid cell-enriched receptors and transporters (M4.3) B Regulation of antigen presentation and immune response (M5)
RNASE6
MNDA
CLEC7A FCGR1B
SIGLEC7 CHST15
0 IRAK3 CTSH
1
−0.5
HCK 0.5 LAT2
CD14 SLC7A7
−1
0
SYK TRAT1
−1.5
1.5
−0.5
−
−2
LILRA1 CD86 CD3G −1
−1 TLR1

BTK MEF2C
CD1D KCTD12
LIME1 FCER1G
NCKAP1L LYN
LRRC25 MS4A14
FYB FGR
SLC8A1 MPEG1 RPS6KA1 CD3E
CD86
C Enriched in monocytes (M11.0) D TLR and inflammatory signaling (M16)

FCGR1A
FCGR1B
CRISP
PLXDC2
SIGLE
FCN1
KCNJ15
STAB1
WDFY3 FCGR1A
CD1 1
FPR2
LRP1
TN CD9
HPS 2
LD2
NO P2
CS FBI
C7
FBP R
FA 3
1 SIGLEC9 FCGR1B
TG T1 4
63
2
E
D
F1
I
SO D1 2
CH L
CT T15 2
S
R
SH
C H
SL
IRAK3 FPR2
AL
M 0 C 1
LIL NDA
AT
D
LIL 7A
RA 7
TIM RA1 −1 LY 6
P2 FG N BST1
0
FES
EPB AIF1 LILR R
41L −2
− 2 FCG A5
3 −1
−1
PYG R2A
TBXAS L LILRB3
1 P2RY13 LILRA6
SIRPB1 −3
3 LILRA2
−2
CLEC7A
SERPINA1
HNMT
CD86
SIGLEC9 ALOX5 TLR1
IFI30
FCGRT
GCA
AQP9
CD1D
M1 KYNU
PRA 3 D12 TLR8 FGR
K
IRA FP
KCT AX
C 1 ITG AR
T
LS K FC 31
HC 2
B
RA GK1 2 ANPEP LILRB3
B
XN P2
S N
PL DA VN
4 7
EC 46
P2 ST 7
A
H E
C 4A
CL NF
RY 3
MA K3
C5AR1 NCF4
EG 3
S
3R B
Z
MP 1
S 1
M
ANP A1
F
ALO YK
P
X5
B
2
SPI1
E
9
1
GRN
NCF
DMXL2
CYB
RNF130
FCAR TLR4
MEGF
SLC11A
IL1
ITGAX AQP9
Before immunization Unprotected after immunization Protected after immunization
Fig. 3. Fold-change gene expression in CPS volunteers for genes from BTMs negatively enriched upon immunization and positively enriched in protected
compared to unprotected immunized volunteers. Charts show the log2 fold change expression upon PfRBC stimulation relative to uRBC background for genes that
are found in the leading edge of enrichment in the GSEA analysis in each represented module: (A) BTM “myeloid cell–enriched receptors and transporters (M4.3),”
(B) BTM “regulation of antigen presentation and immune response (M5),” (C) BTM “enriched in monocytes (M11.0),” and (D) BTM “TLR and inflammatory signaling
(M16).” Different colored lines represent volunteers before immunization (green) and CPS immunized protected (pink) and unprotected (blue) volunteers.
gene sets similar to the BTM identified, e.g., those related to in- RTS,S vaccinees (Fig. 5B). Thus, this vaccine-induced down-regulation
flammatory responses that were positively associated with protection may be negative for protection. These BTMs were associated with
or Chaussabel’s monocyte module M4.14, which was negatively en- protection regardless whether children or infants had been vaccinated
riched upon CPS immunization but positively enriched in protected with RTS,S or the comparator vaccine, and we had found similar
versus unprotected immunized individuals. associations with protection for PfRBC recall responses in CPS
We also performed a weighted gene correlation network analysis volunteers (both before and after immunization). Therefore, these
(WGCNA) (26) to identify groups of coexpressed genes and applied responses seem vaccine nonspecific.
GSEA to ranked genes according to the differential correlation. Gene set analyses using alternative data files to BTMs resulted in
The results were consistent with the GSEA analysis applied to dif- the identification of similar responses (data file S4). For instance,
ferentially expressed genes, particularly for CPS immunogenicity for CSP stimulations, RTS,S vaccinees had enriched gene sets related
upon CSP stimulation (data file S5). to IFN responses (e.g., Reactome “IFN- signaling” or “IL-2 signaling”
and Chaussabel’s modules M1.2, M5.12, or M3.4), T cells (BioCarta
Gene sets associated with RTS,S/AS01E immunization “cytotoxic T cells” or “TH1TH2 responses”), or cell cycle (BioCarta,
and clinical protection Reactome, and KEGG “cell cycle”) that reflect antigen-specific
As expected, gene set analysis of RTS,S immunogenicity revealed responses. Some of Chaussabel’s modules were negatively enriched,

significantly enriched BTMs in response to CSP (q ≤ 0.1; Fig. 4A) mainly related to inflammatory responses (M3.2 and M4.13) and
but almost no BTMs in response to PfRBC. CSP stimulation in monocytes (M4.14), which were also consistent with the results
RTS,S vaccinees relative to comparators induced innate and obtained with BTMs. Curiously, only six gene sets were positively
acquired responses, including DC activation, NK cells, inflammatory/ enriched in protected versus unprotected RTS,S vaccinees for CSP
TLR/chemokines, IFN/antiviral sensing, T cells, and cell cycle stimulations, but these included the Chaussabel’s IFN modules, as
BTMs. However, it also induced the repression of BTMs mainly well as the Reactome IFN- signaling and IFN- signaling sets,
related to inflammatory/TLR/chemokines, signal transduction, and reinforcing the importance of these responses on protection.
monocytes. Similar BTMs were also repressed in PfRBC recall Similarly, when we performed GSEA of WGCNA, we found four
responses in CPS immunization compared to preimmunization. BTMs significantly (q ≤ 0.1) enriched in RTS,S versus comparator
For CSP stimulation in RTS,S vaccinees, BTMs related to DC vaccinees for CSP stimulations, all related to IFN/antiviral responses
activation, cell cycle, complement, inflammatory/TLR/chemokines, (data file S5). Regarding RTS,S protection, for CSP stimulation, two
IFN/antiviral sensing, monocytes, and neutrophils were positively gene sets were identified and were related to recognition of antigen
associated with protection from clinical malaria. Of these BTMs, by B cells (BioCarta “BCR signaling pathway” and Reactome “antigen
the ones related to DC activation, cell cycle, and IFN/antiviral sens- activates B cell receptor, leading to generation of second messengers”),
ing were associated with RTS,S immunogenicity and were not which is interesting given the role of IgG antibodies to CSP in
found in comparator vaccinees (Fig. 4B), suggesting vaccine (antigen) RTS,S-induced protection (7). This could indicate that there are
specificity in these responses. Curiously, BTM related to inflammatory/ CSP-specific memory B cells in samples from RTS,S vaccinees.
TLR/chemokine responses, monocytes, and neutrophils that were
associated with protection in RTS,S vaccinees were found to be Changes in frequencies of leukocyte populations
inversely associated with RTS,S immunogenicity. We performed leukocyte phenotyping (fig. S8A) to identify changes
In comparator vaccinees, additional BTMs were associated with in the frequencies of cell subsets due to CPS or RTS,S immunization
protection, suggesting innate and naturally acquired responses in and between protected and nonprotected volunteers, because these
these children. Positively enriched BTMs were related to DC activa- changes may drive, in part, the observed differences in module re-
tion, cell cycle, signal transduction, and monocytes. Three different sponses. Consistent with previous observations (9), we detected an
T cell BTMs were negatively associated with protection. expansion of  T cells after CPS immunization (adjusted P < 0.001)
Despite the scarce significant associations of BTMs with RTS,S and higher frequencies of B cells in protected than in nonprotected
immunogenicity for PfRBC recall responses, some BTMs were CPS-immunized volunteers (adjusted P = 0.031) (fig. S8B). In the
negatively associated with protection in RTS,S vaccinees (Fig. 4C). RTS,S study, there were no statistically significant changes between
These BTMs were mainly related to cell cycle and inflammatory/ RTS,S and comparators or between protected and nonprotected
TLR/chemokine response and included “proinflammatory cytokines volunteers (fig. S8C).
and chemokines (M29),” which was also inversely correlated with
vaccination. These responses could be related to the decrease of Predictive transcriptional signatures of
PfRBC exposure due to the protection induced by the vaccine in immunization-induced protection
RTS,S vaccinees. Instead, comparator vaccinees had some BTMs We aimed to identify transcriptional signatures that predict protec-
positively correlated with protection (Fig. 4C) that could be related tion in each immunization strategy separately using a previously
to naturally acquired responses. developed systems biology approach (27, 28). This approach con-
As expected, most of the genes in the leading edge of the BTMs sisted of a network-based analysis, which overlays the gene differen-
associated with protection and RTS,S immunogenicity were more tial expression values on its corresponding protein in a human
up-regulated in protected compared to unprotected RTS,S vaccinees protein functional network. This analysis increases the signal-to-
(Fig. 5A). When looking at the FC of genes in BTMs negatively noise ratio and takes into consideration the interactions among
enriched in RTS,S vaccinees upon CSP stimulations but positively multiple components to obtain a more robust and biologically rele-
associated with protection, such as “enriched in monocytes (M11.0),” vant output. In addition, the network collects previously published
we found that despite vaccination-induced down-regulation of genes, biological data on the proteins in relation to the immune response
protected RTS,S vaccinees had less down-regulation than unprotected to malaria, reducing the solution space and focusing the analysis on
A Immunogenicity B Protection (CSP stimulation)

PfRBC
RTS,S
Comp
CSP
Activated dendritic cells (M67) Activated dendritic cells (M67)

Enriched in activated dendritic cells (II) (M165) Enriched in activated dendritic cells (II) (M165)
Complement and other receptors in DCs (M40)
Enriched in activated dendritic cells/monocytes (M64)
Chemokines and inflammatory molecules in myeloid cells (M86.0) Enriched in T cells (I) (M7.0)
Mitotic cell cycle in stimulated CD4+ T cells (M4.9) T cell differentiation (M14)
Signaling in T cells (I) (M35.0) T cell surface signature (S0)
Signaling in T cells (II) (M35.1) Cell division (stimulated CD4+ T cells) (M46)
T cell activation (II) (M7.3) Cell cycle (I) (M4.1)
Cell cycle (I) (M4.1) Cell division in stimulated CD4+ T cells (M4.6)
Cell cycle (II) (M4.10) Mismatch repair (I) (M22.0)
Cell cycle (III) (M103) Mitotic cell cycle–DNA replication (M4.4)
Cell division (stimulated CD4+ T cells) (M46) Cell cycle and transcription (M4.0)

Cell division CD4+ T cells (M4.6) Complement activation (I) (M112.0)
E2F transcription factor network (M8) Chemokine cluster (I) (M27.0)
Mismatch repair (I) (M22.0) DC surface signature (S5)
Mitotic cell cycle–DNA replication (M4.4) Immune activation–generic cluster (M37.0)
Mitotic cell cycle (M4.7) Myeloid cell-enriched receptors and transporters (M4.3)
Mitotic cell cyclein stimulated CD4+ T cells (M4.5) Resting dendritic cell surface signature (S10)
Mitotic cell cycle (M6) TLR and inflammatory signaling (M16)
Enriched in NK cells (I) (M7.2) Proinflammatory cytokines and chemokines (M29)
NK cell surface signature (S1) Antiviral IFN signature (M75)
Enriched in NK cells (KIR cluster) (M61.1) Innate antiviral response (M150)
Chemokine cluster (I) (M27.0) Type I interferon response (M127)
Chemokine cluster (II) (M27.1) Viral sensing and immunity/IRF2 targets network (II) (M111.1)
DC surface signature (S5) Suppression of MAPK signaling (M56)
Immune activation–generic cluster (M37.0) Enriched in myeloid cells and monocytes (M81)
Myeloid cell-enriched receptors and transporters (M4.3) Enriched in monocytes (surface) (M118.1)
Resting dendritic cell surface signature (S10) Enriched in monocytes (II) (M11.0)
TLR and inflammatory signaling (M16) Enriched in monocytes (IV) (M118.0)
Proinflammatory cytokines and chemokines (M29) Monocytes surface signature (S4)
Antiviral IFN signature (M75) Formyl peptide receptor mediated neutrophil response (M11.2)
Innate activation by cytosolic DNA sensing (M13) Enriched in neutrophils (I) (M37.1)
Innate antiviral response (M150) Blood coagulation (M11.1)
Viral sensing and immunity/IRF2 targets network (II) (M111.1)
PLK1 signaling events (M4.2)
Regulation of signal transduction (M3) C Protection (PfRBC stimulation)
RTS,S
Comp
Suppression of MAPK signaling (M56)

Enriched in monocytes (I) (M4.15)
Enriched in monocytes (III) (M73) Chemokines and inflammatory molecules in myeloid cells (M86.0)
Enriched in monocytes(IV) (M118.0) Enriched in activated dendritic cells (II) (M165)
Enriched in myeloid cell and monocytes (M81) Enriched in dendritic cells (M168)
Monocyte surface signature (S4) Enriched in B cells (IV) (M47.3)
Enriched in neutrophils (I) (M37.1) E2F transcription factor network (M8)
E2F1 targets (Q3) (M10.0) Cell cycle (I) (M4.1)
Lysosomal/endosomal proteins (M139) Cell cycle (II) (M4.10)
Transmembrane transport (I) (M87) Chemokine cluster (I) (M27.0)
Proinflammatory cytokines and chemokines (M29)
Chemokine cluster (II) (M27.1)
CSP immunogenicity, same direction PLK1 signaling events (M4.2)
0 CSP immunogenicity, opposite direction Immuneregulation–monocytes, T and B cells (M57)
−3 −2 −1 0 1 2 3 PfRBC immunogenicity, same direction Transmembrane transport (I) (M87)
DC activation NK cells
B cells ECM and migration Signal transduction
T cells Inflammatory/TLR/chemokines Monocytes
Cell cycle Interferon/antiviral sensing Neutrophils
Fig. 4. Transcriptional responses associated with RTS,S/AS01 immunogenicity and protection in RTS,S/AS01 and comparator vaccinees. Each square represents
a BTM. The color shading indicates normalized enrichment scores obtained by GSEA analysis for BTMs. Assignment of a BTM to a high-level annotation group is illustrated
by a colored sidebar. GSEA, CAMERA, and Tmod were run with genes ranked by the expression of (A) RTS,S/AS01E vaccinees (after immunization) relative to comparator
vaccinees and (B) protected relative to nonprotected individuals in RTS,S/AS01E and comparator vaccinees for CSP recall stimulations and (C) for PfRBC recall stimulations.
Modules that did not reach the significance cutoff of FDR q value of 0.1 in all three enrichment methods or a minimum of 10 matched genes were eliminated. Modules
without annotation are not shown. Modules that represent common associations of both immunogenicity and protection are highlighted with a circle for CSP recall
responses and a triangle for PfRBC recall responses, filled symbols when enrichment had the same direction, and empty symbols when enrichment had the opposite
direction.
A IFN signatures (M67, M75, M111.1, M127, and M150) B Enriched in monocytes (M11.0)
MS4A4A
CD163
RAS
LILR F
FBP1
QPC
RNA R 1
RBP7
C1QB
CXCL10
M
C1QC
SF4
Y
A
SE2
FG N
0.5
FP 1
O
6
NR AM
IFITM1 0.5
SL
STAT1
L2
A IF 1
IT
C1 B32
LG
G
1A
AL
R
0.25
C5
A
EIF2AK2 0.25 OAS3
S2
AR FG
CD 1 R CD
36 0 86
0 SL NC
DDX58 SERPING1 C7 F1C
A7
S10 −0.2
−0.25
−0
0.2
0.
0.25
0 2
25 LILR
−0.25
0 5 0A9 A5
ALDH
1A1
PLBD1
PLSCR1 IFIT2 −0.5
5
−0.5 BLVRB CCR1
LILRB2
FPR2
IRF7 PARP9 IMPA2 CLEC4A

G LILRB3
FCER1

FCN1
3
15A
SLC
TNFSF13B RSAD2 467
F6
ZNF IGS P
C7 B
LE RO 3
SIG CFP
TY T
OASL CS 86
PML RN LY
G
2
PE A
CP ULF
ST
RX D1
P
DDX60
S
IFIH1
A
C2
PE
R
1
BTK
AN
RB
PA
XD
FC
X5
8B
MNDA
FCGR1B
PLXNB2
IFIT1 BST2
L
R
PL
PDG
ALO
LI
I
FAM19
S
HERC5 OAS1
Comparators RTS,S unprotected RTS,S protected

Fig. 5. Higher fold change gene expression in protected than unprotected RTS,S vaccinees for genes from BTM related to IFN signatures and monocytes. Charts
show the log2 fold change of genes upon CSP recall stimulation related to DMSO background for genes that are found in the leading edge of enrichment in the GSEA
analysis in the BTMs related to (A) IFN signatures (M67, M75, M111.1, M127, and M150) and (B) the genes contributing to enrichment of the BTM “enriched in monocytes
(II) (M11.0).” Different colored lines represent comparator vaccinees (green), protected (pink), and unprotected (blue) RTS,S/AS01E vaccinees.
biologically meaningful regions. The network-based analysis pro- member 11B)], cell signaling/signal transduction [F2, KL (Klotho),
duced a list of proteins that behaved differently in protected and G protein subunit alpha 11 (GNA11), protein tyrosine kinase 2 (PTK2),
unprotected individuals, and signatures of protection, consisting in calcium voltage-gated channel auxiliary subunit alpha2delta 2
combinations of three to five of these proteins, were identified (CACNA2D2), CACNA1F, rho GDP dissociation inhibitor beta
through data science analysis. (ARHGDIB), potassium voltage-gated channel subfamily B member 1
We identified several signatures of protection not only after (KCNB1), and adrenoceptor alpha 2B (ADRA2B)], an inhibitor of
immunization but also before immunization, when using CSP or nuclear factor B (NF-B) transcription factor [NFBIE (NF-B
PfRBC stimulations as recall response and for both immunization inhibitor epsilon)], chaperones [heat shock protein family A
strategies. Table 1 shows the best signature for each experimental member 8 (HSPA8) and calreticulin (CALR)], and cell adhesion
condition for the CPS and RTS,S immunization studies based on molecules (ITGA2 and ITGB7). Some of these genes were found
generalization capabilities and predictive accuracies calculated in BTMs associated with CPS-i nduced protection: CXCL10 in
through leave-one-out (LOO) cross-validation. Accuracy of the “chemokine cluster (I) and (II) (M27.0 and M27.1),” “antiviral IFN
signatures corresponds to the proportion of samples correctly clas- signature (M75),” and “innate activation by cytosolic DNA sensing
sified, whereas generalization capability corresponds to the proba- (M13)”; CXCL16 in “DC surface signature (S5)” and “antiviral
bility to correctly predict a sample that was not used in the training sensing and immunity/IRF2 targets network (I) (M111.0)”; IL3RA in
sets of LOO cross-validation. For the CPS top signatures shown, “activated (LPS) DC surface signature (S11)”; and PTK2 in “immune
generalization capabilities ranged from 71 to 90% and accuracies activation-generic cluster (M37.0),” although PTK2 was not found
from 71 to 100%. For the RTS,S top signatures shown, generaliza- at the leading edge of the GSEA enrichment. The other genes did not
tion capabilities ranged from 79 to 83%, and accuracies ranged from appear in the identified BTMs. In the top RTS,S signatures, there
84 to 100%. Many of the genes from these signatures encode were genes encoding for cytokines (IL-18, IL-3, and IL-12B), cell
proteins involved in immune response with different functions. receptors [PRLR (prolactin receptor), TLR4, FCGR2A, and ciliary
CPS protection signatures included genes encoding for cytokines neurotrophic factor receptor (CNTFR)], apoptosis [caspase 6
[CXCL10, CXCL16, colony-stimulating factor 2 (CSF2) or granulocyte- (CASP6) and tumor necrosis factor receptor superfamily member 10A
macrophage colony-stimulating factor (GM-CSF), transforming (TR10A)], cell cycle [cyclin E2 (CCNE2)], cell signaling/signal transduc-
growth factor–3 (TGF-3), and IL-7], receptors [F2RL2, IL-3RA, tion {Erb-B2 receptor tyrosine kinase 2 (ERBB2), FAK2 (focal adhe-
and TNFRSF11B (tumor necrosis factor receptor superfamily sion kinase 2), GNAT3, serum/glucocorticoid regulated kinase 1 (SGK1),
mitogen-activated protein kinase kinase kinase 7 (M3K7), Cbl Prospective validation of signatures of protection
proto-oncogene C (CBLC), JAK2 (janus kinase 2), GNA11, P85B Each of the experimental conditions and immunization strategies
(PIK3R2, Phosphoinositide-3-Kinase Regulatory Subunit 2), SH2B2 provided a list of signatures with good accuracies and generaliza-
(SH2B Adaptor Protein 2), WASL (WASP Like Actin Nucleation tion capabilities, not only the ones shown in Table 1. We applied a
Promoting Factor), MAP2K7 [mitogen-activated protein kinas (MAPK) prioritization algorithm to select a reduced number of proteins,
kinase 7], and IKKB [inhibitor of NF-B (IB) kinase B]}, and cell with the aim of enhancing the probability of identifying signatures
adhesion molecules [ITB1 (integrin beta 1), NCAM1 (neural cell that predict protection in the validation and covering most of the
adhesion molecule–1), and VCAM1 (vascular cell adhesion molecule–1)]. conditions analyzed. We chose 70 genes from the CPS study and
Some of these genes were found in BTMs associated with RTS,S- 74 genes from the RTS,S study (data file S6), 13 of which were in both
induced protection: SGK1 in “immune activation–generic cluster lists (MACROD2, SGK1, MEN1, SMAD6, TYROBP, F2, CXCL4,
(M37.0)” and “enriched in monocytes (II) (M11.0)”; TLR4 in “TLR CXCL10, JAM2, YOD1, CD276, RMDN2, and KL). The list of selected
and inflammatory signaling (M16)” and “enriched in myeloid cells genes was used for validation through qRT-PCR (Fig. 6A) on an
and monocytes (M81)”; and IL-18 in “enriched in activated DC additional set of 15 volunteers (120 samples) from a second human
(M165),” although only the latter was found in the leading edge of CPS immunization trial and an independent set of 23 children and
the enrichment. 10 infants from the RTS,S phase 3 trial (41 samples, only for CSP

Table 1. Signatures of protection induced by CPS and RTS,S/AS01E identified from mathematical network models based on the transcriptional data.
Signature of Generalization
Time point Age group Antigen Accuracy
protection capability
CPS
Adults CSP/DMSO F2, CXCL10, and KL 87.50% 87.50%
HSPA8, CALR, and
Adults PfRBC/uRBC 70.83% 70.83%
Postimmunization CXCL16
F2RL2, IL-3RA, and
Adults CSP/DMSO and PfRBC/uRBC 81.25% 93.75%
GNA11
CSF2, NFkBIE,
Adults CSP/DMSO TNFRSF11B, and 87.50% 100%
ITGA2
Preimmunization Adults PfRBC/uRBC PTK2, TGF-3, and 70.83% 70.83%

HSPA8
CACNA2D2,
Adults CSP/DMSO and PfRBC/uRBC CACNA1F, ITGB7, 77.08% 89.58%
ARHGDIB, and KCNB1
Pre- and IL-7, ADRA2B, and
Adults CSP/DMSO 89.58% 93.75%
postimmunization F2RL2
RTS,S/AS01E
GNAT 3, SGK1, and
Infants CSP/DMSO 80% 86.67%
TLR4
TLR4, P85B, and
Infants PfRBC/uRBC 82.69% 88.46%
SH2B2
ERBB2, PRLR, and
Children CSP/DMSO 80.49% 92.68%
FAK2
JAK2, NCAM1,
Postimmunization
Children PfRBC/uRBC VCAM1, CNTFR, and 80.65% 100%
GNA11
Infants and children CSP/DMSO ITB1, CASP6, and IL-18 79.29% 83.57%
M3K7, IL-3, FCG2A,
Infants and children PfRBC/uRBC 80% 90%
and CBLC
GNAT3, SGK1, and
Children CSP/DMSO and PfRBC/uRBC 81.94% 84.72%
TLR4
TR10A, CCNE2,
Children PfRBC/uRBC ADA2A, SRC, and 79.49% 100%
IL-12B
Preimmunization
Children CSP/DMSO TLR4, WASL, and P85B 81.13% 86.79%
TLR4, P85B, SH2B2,
Children CSP/DMSO and PfRBC/uRBC 78.79% 87.88%
MAP2K7, and IKKB
A B C CPS pre vaccination signature

Sporozoite immunization under CQ/MQ prophylaxis (CPS) ITGA2 NFKBIE CSF2 TR11B
(CSF2, NFKBIE, TNFRSF11B, and ITGA2)
N = 15
Adults 10/15 protected MK11, MK12, IKK IL-2RB RANK
True-positive rate (1-specificity)

0 3 8 3/5 CPS-CQ group MK13,MK14
Months 7/10 CPS-MQ group
NFkBIA
Immunization doses with 8 Pf mosquitoes

:
NF- B NF- B
FOXO1,FOXO3,
CQ (n = 5), MQ (n = 10) NFAC1, NFAC2, RelA/p50 STAT5B RelB/p52
Blood samples NFAC3
Pf challenge
IFN- IL-12A, IL-2RA
TNFA
RTS,S/AS01E multicenter phase 3 African vaccine trial
N = 33 Increased vaccine take
Clinical follow-up False-positive rate (1-specificity)
11/33 non protected Protective malaria responses
0 3 12
Months
CPS post vaccination signature
F2 KL CXCL10 (F2, CXCL10, KL)
Vaccine doses (RTS,S)

5–17 months Blood samples VWF

5–17 months
6–12 weeks
Platelets
NF- B
RelA/p50 CXCR3
Experimental design VEGFA, VEGFB,
VEGFC, VEGFD
PBMC stimulations Gene expression
qRT-PCR IFN- IL-12A,
24/12 CSP and GM-CSF TNFA
hours DMSO
PfRBCs and ~70 genes involved Protective malaria responses False-postivie rate (1-specificity)
16 hours
uRBCs in the proposed
signatures of
RTS,S post vaccination signature
protection
ERBB2 PRLR FAK2 (ERBB2, PRLR, and FAK2)
RAC1,
JAK3 JAK2 SRC RAC2,

Computational analysis: Validation of biomarkers and RAC3
generation of predictive models
EGFR CTNB1 MK01, MK03
Proposed signatures Data science Signatures
of protection of protection NF- B
~70 genes measured
IgM.Pf.M0
IgG3.HBsAg.M0
IgG2.C term.M3
IgG.C term.M0
IgM.NANP.M0
IgG4.C term.M3
IgG3.NANP.M0
IgG4.C term.M0
IgG2.NANP.M3
IgG.HBsAg.M3
>70% accuracies STAT5B STAT3 RelA/p50
IgG2.C term.M0
by qRT-PCR
IgG4.HBsAg.M0
-
IgG3.HBsAg.M3
IgG4.NANP.M3
IgG.C term.M3
IgG1.NANP.M0
IgG1.HBsAg.M3
IgG2.NANP.M0
IgG1.NANP.M3
IgM.C term.M0
IgG1.C term.M3
IgG.NANP.M0
IgG4.NANP.M0
IgG1.C term.M0
IgG.NANP.M3
IgG3.C term.M0
IgG.HBsAg.M0
IgM.C term.M3
IgG4.HBsAg.M3
IgM.HBsAg.M3
IgG3.NANP.M3
IgG3.C term.M3
FOXO1, SMAD2,
IgG2.HBsAg.M0
IgG1.HBsAg.M0
IFN- IL-12A,
IgG2.HBsAg.M3
IgM.HBsAg.M0
IgM.NANP.M3
IL-2RA IL-6
0 5 10
Selected signatures
Mean Decrease Accuracy
TNFA FOXO3 SMAD3

Predictive models
(generated by
ANN, validated
Protective malaria responses False-positive rate (1-specificity)
by LOOCV)
Fig. 6. Validation of signatures of protection. (A) Samples for validation were obtained from a CPS trial in malaria-naïve adults that used chloroquine (CQ) and
mefloquine (MQ) and a separate group of children and infants from two different African countries from the RTS,S phase 3 trial. PBMCs were similarly stimulated in vitro
with CSP peptide pool and DMSO, or in only CPS samples, PfRBC, and uRBC. Gene expression was measured by qRT-PCR from about 70 genes involved in the signatures
of protection of each immunization strategy and that were selected on the basis of Greedy algorithms. Normalized qRT-PCR data and data science methods were used to
validate 32 and 37 previously identified signatures for CPS and RTS,S, respectively. Predictive mathematical models were developed using artificial neural network (ANN)
and LOO cross-validation (LOOCV) for three selected signatures. (B) Potential mechanisms by which the genes of the three selected validated signatures may confer protection
against malaria. Order of signatures: CPS preimmunization signature, CPS postimmunization signature, and RTS,S postimmunization signature. Purple and orange nodes
indicate genes that belong to the signature (purple and orange represent up-regulated and down-regulated genes, respectively). White nodes indicate genes participating
in the mode of action of the signature. Broken-lined nodes contain more than one gene, all of them acting in the mode of action in the same way. Purple arrows show
activation, whereas orange lines show inhibition. NFAC1, nuclear factor of activated T-cells, cytoplasmic 1; VWF, von willebrand factor. VEGFA, vascular endothelial
growth factor A. (C) ROC curves predicting malaria protection (based on the qRT-PCR validation data) for the selected CPS pre- and postimmunization and RTS,S post-
vaccination signatures of protection.
recall responses and with some different experimental conditions, zation (F2, CXCL10, and KL) with 100% accuracies. Some genes
which we cannot exclude to have an impact on gene expression). were found in several pre- and postimmunization CPS signatures
We validated 32 and 37 protein combinations (signatures) with such as CXCL10 and F2RL2 among others. The two CPS signatures
accuracies of >70% for CPS and RTS,S (data file S7), respectively, of protection above indicate a prominent role of the NF-B canonical
using the focused set of validation genes. pathway and an activation of the noncanonical pathway (Fig. 6B).
For CSP stimulation, two of the originally top performing signa- ITGA2 (CD49b) is an integrin that elicits several cellular signals,
tures for CPS immunization (Table 1) predicted the volunteers including the p38 MAPK cascade (29) that controls the transcrip-
from the validation set who were going to be protected before tional activity of RelA (NF-B p65) (30). Further contributing to
(CSF2, NFkBIE, TNFRSF11B, and ITGA2) and after CPS immuni- the canonical NF-B pathway, down-regulation of NFBIE (or IkB)
may sustain the activation of NF-B (31). GM-CSF (CSF2) signaling (52) that activates transcription of hypoxia-inducible genes, includ-
may also activate the canonical NF-B pathway (32) through IKK ing heme oxygenase 1 (HMOX1) (53), which, in malaria and other
that leads to the degradation of the NF-B inhibitor IB (32). NF- infections, is induced to limit iron availability and to provide other
B is a nuclear factor that, when activated through the canonical path- anti-inflammatory functions (54), possibly contributing to the ac-
way, induces an inflammatory response and is involved in innate and quisition of malaria protection. Furthermore, ERBB2 activation
acquired immune responses. Down-regulation of osteoprotegerin might lead to the transcription factor STAT3 activation through
(TNFRSF11B or TR11B), which inhibits the activation of the non- JAK3 (55). STAT3 participates in the induction of several molecules
canonical NF-B pathway through receptor activator of NF-B- involved in immune response, such as IL-6 (56, 57). In addition,
receptor activator of NF-B ligand (RANK-RANKL) binding (33), STAT3 might up-regulate FOXO1/FOXO3 expression (58) and fur-
would result in activation of the noncanonical NF-B pathway. ther contribute to protective responses as mentioned before. The ac-
In addition, GM-CSF (CSF2, a target gene of NF-B) (34) is im- tivation of JAK3 also results in phosphorylation of STAT5B (59, 60).
portant for development and maturation of DCs (35) that are key Last, ERBB2 can also dimerize with epidermal growth factor recep-
for antigen presentation upon vaccination and infections and is tor (EGFR) (61) that might also promote the activation of STAT3
critical for T cell functions. GM-CSF also increases IL-2R (36), (62, 63) and STAT5B (64). On the other hand, PRLR activation induces
potentiating IL-2 signaling, which, in turn, promotes the expres- JAK2 (65), which can phosphorylate EGFR (66). Last, reduced

sion of IL-2R subunit through signal transducers and activators activity of FAK2 might decrease the RAC1, RAC2, and RAC3 activa-
of transcription 5B (STAT5B) (37) in a positive feedback loop that tion (23), which can dampen MAPK signaling (MK01 and MK03)
can increase expansion of lymphocytes and IFN- production (38). (23), allowing higher FOXO1 and FOXO3 expression (23, 39, 40).
Therefore, GM-CSF would lead to a better vaccine take or enhanced MAPK are also involved in SMAD2 and SMAD3 inhibition (67);
protective responses upon vaccination because GM-CSF is related thus, reduced MAPK signaling pathways might increase SMAD2 and
to IFN- and other T helper 1 cell (TH1) responses (35). Last, p38 SMAD3 activity, which are involved in the regulatory T cell differ-
MAPK also activates forkhead box O (FOXO) transcription factors entiation process to avoid inflammation exacerbation in parasitic
(39, 40), among other proteins. FOXO is involved in regulating im- infection (68, 69). Reduced activity of FAK2 can also diminish the
mune responses to inflammation, in response to growth factors and FAK2-induced activation of SRC (proto-oncogene tyrosine protein
T cell–specific functions (41). FOXO3 has been associated with sus- kinase Src) (70). Since SRC is involved in the activation of MAPK
ceptibility to severe malaria (42). The mode of action of the postim- signaling (23) and CTNB1 (71), its reduced activation might con-
munization signature indicates that F2 (protothrombin) results tribute further to the inhibition of CTNB1 and MK01 and MK03.
in thrombin by enzymatic cleavage. Thrombin activates the ca- Some genes appeared in several RTS,S signatures. The most
nonical NF-B pathway (43), has a proinflammatory effect, induces frequent ones were TLR4, present in 24.3% of the signatures, SGK1
leukocyte recruitment, and stimulates chemotaxis and cytokine in 18.9%, MAPK13 in 8%, and EREG in 6%, the latter only in pre-
production by DCs (44, 45). Thrombin also promotes the activa- immunization signatures, whereas the others were found in both
tion of platelets through binding of von Willebrand factor (46). pre- and postimmunization signatures (data file S7).
In addition, activation of platelets is marked by an increase in vas- With the selected validated signatures, we constructed predictive
cular endothelial growth factor, which can induce the secretion of models by artificial neural networks (table S1), which are flexible
GM-CSF that activates the canonical NF-B pathway. CXCL10 is a and can easily provide mathematical formulas to be used as tools in
target gene of NF-B induced by IFN- that stimulates monocytes clinical trials. Tables S2 and S3 show the validation qRT-PCR values
and NK cells, modulates T cell migration through modulation of and prediction results using the models. Area under the curve
adhesion molecules, and activates CXCR3 (47) that is implicated in (AUC) in receiver operating characteristics (ROC) curves of both
T cell homing to the liver (48). KL is a transmembrane protein CPS signature models was excellent, with an AUC of 1 (Fig. 6C),
involved in endoplasmic reticulum stress (49), calcium homeostasis, and the ROC curve of the RTS,S signature model also showed a
modulation of inflammation, and attenuation of the activation of the remarkable predictive capacity, with an AUC of 0.8304 (Fig. 6C).
canonical NF-B pathway (50). Therefore, KL down-regulation in
protected individuals could result in increased activation of NF-B.
For RTS,S, one postimmunization signature for children and DISCUSSION
one for infants that were among the best signatures obtained initial- We identified and validated genes and signatures that predict vaccine-
ly (ERBB2, PRLR, FAK2 and GNAT3, SGK1, and TLR4; Table 1) induced protection in naïve adults and young African children from
were validated with accuracies of 90 and 96%, although GNAT3 different countries, before and after immunization, with high
could not be detected in any validation sample, probably due to generalization capabilities and accuracies, using samples from a se-
the different experimental conditions (e.g., shorter duration of the ries of malaria vaccine trials of two different malaria immunization
in vitro stimulations; see Materials and Methods). Because of the strategies targeting the preerythrocytic phase of P. falciparum.
lack of detection of GNAT3 for the infants’ signature and because Although the identification of predictive signatures of protection
the target population of RTS,S/AS01E is children, we selected the in samples collected before immunization for both immunization
postimmunization signature of this age group. The mode of action strategies was unexpected, previous studies have also found baseline
of the signature (ERBB2, PRLR, and FAK2) also indicates a role of predictors of vaccination responses (72, 73) and RTS,S efficacy (74).
the canonical NF-B pathway (Fig. 6B). Signaling through the In addition, African infants and children could have malaria-specific
kinase ERBB2 and the PRLR and inhibition of FAK2 could prevent responses at baseline as they may be previously exposed to Plasmodium
the inhibition of p50 (NFKB1) by catenin beta 1 (CTNB1) (51), in utero and during the first months of life, although malaria trans-
increasing NF-B activity. However, ERBB2 may provide protection mission intensity at that time for the three sites was low/moderate
by increasing the rate of hypoxia-inducible factor 1A synthesis (3). In contrast, CPS volunteers were true malaria-naïve adults. The
genes of the preimmunization signatures in both immunization juvant contains monophosphoryl lipid A (MPL), which is a deriva-
regimens and the BTMs associated with CPS protection at pre- tive of LPS and is therefore recognized by TLR4 (84). Thus, higher
immunization reveal innate, inflammatory responses and cell expression or activation of TLR4 in protected individuals could be
activation rather than antigen-specific memory responses. Thus, involved in a better response to RTS,S/AS01E vaccination. In another
CSP and PfRBC recall responses at baseline could reflect the systems biology study, TLR5 expression was associated with the
intrinsic capacity of the volunteers to respond effectively to magnitude of antibody responses to an inactivated influenza vaccine,
immunization (vaccine take). This is of utmost importance because and it was shown that TLR5-gut microbiota sensing was necessary
the absence of such signatures of protection at baseline could help for antibody responses to this and the inactivated polio vaccine (85).
identify individuals who require, for instance, a higher immuni- Thus, the presence of TLR4 in several preimmunization RTS,S signatures
zation dose to become protected [we know that all CPS volun- points to this receptor as a possible target to increase vaccine take,
teers likely would have been protected if immunized by optimal at least for TLR4-targeting vaccine adjuvants. Previous or simulta-
doses (8, 9)] or an additional immunomodulatory stimulus or neous treatments that increase TLR4 expression and signaling to
different adjuvants. all or selected infants and children could increase vaccine take and
The genes and signatures of both CPS immunization and RTS,S efficacy.
vaccine provided insights into the molecular mechanism necessary CPS-induced protection is mediated by preerythrocytic immu-

to induce malaria protection through immunization. The two se- nity (8), although the underlying immunological mechanisms are
lected CPS signatures of protection (CSF2, NFkBIE, TNFRSF11B, not known. In previous studies, CD107a+ CD4+ T cells and CD8+
and ITGA2 and F2, CXCL10, and KL) indicate a prominent role of T cells producing GzmB upon PfRBC stimulation were associated
the NF-B canonical pathway in priming the immune system for a with protection upon PfRBC stimulation (9), whereas IFN-+ CD4+
protective vaccination response and an activation of the noncanonical T cells and  T cells, as well as CSP and MSP1-specific antibody
pathway via the RANK-RANKL system. The RANK-RANKL sys- responses, were associated with exposure during immunization but
tem is expressed in a wide variety of tissues, including T cells, DCs, not associated with protection (9, 21). Thus far, CPS-specific T cell
and memory B cells (75, 76). NF-B has a central role in inflammation, responses to the hallmark preerythrocytic antigen CSP have not
inducing the expression of proinflammatory molecules, including consistently been detected by either flow cytometry or enzyme-
cytokines and chemokines, and regulates the activation and differ- linked immune absorbent spot. Our differential gene expression
entiation of innate and acquired immune cells. The canonical path- analysis reporting no CSP responses and stronger PfRBC responses
way leads to acute responses, whereas the noncanonical NF-B is consistent with these data. Therefore, the negative association of
activation leads to sustained responses. We speculate that higher BTMs with protection upon CSP recall responses, which seems con-
NF-B activity leads to a more efficient immunization response, tradictory, could be due to noisy responses because of low signal
and NF-B–induced cytokines, such as IFN-, TNF and GzmB (9, 77), to background ratios. In contrast, BTMs induced upon PfRBC stim-
may directly contribute to malaria-protective responses. The cor- ulations related to signaling in T cells, DC activation, and signal
relation network analysis of CPS-induced protection also revealed transduction were found to be positively associated with protection
some gene sets related with the mode of action of the signatures in immunized volunteers and not before immunization and are
and NF-B, e.g., Reactome “TAK1 activates NFKB by phosphorylation thus in agreement with an adaptive malaria-specific CPS-induced
and activation of IKKS complex,” BioCarta “RELA pathway,” or the response. Therefore, only modules relevant to protection against
BioCarta “RANKL pathway.” In addition, we found the BioCarta late-liver (and blood-stage) infection seem to be induced by CPS
pathways “protothrombin (TPO) pathway” and “platelet-derived immunization, whereas no modules relevant to protection against
growth factor (PDGF) pathway,” which are related to F2 found in the sporozoite infection appear to be elicited.
the postimmunization CPS signature. Recently, platelets have been Regarding RTS,S, previous studies mainly showed antibody
shown to control Plasmodium parasitemia through killing of intra responses (7) and some T cell responses (86, 87), with only weak
erythrocytic parasites (78). associations of cell responses with protection. We found many
The RTS,S postvaccination signature of protection (ERBB2, PRLR, BTMs correlating with protection in RTS,S vaccinees and not in
and FAK2) and other genes found in the other signatures (e.g., comparators that were associated with immunogenicity, suggesting
TLR4) also indicates an important role of the canonical NF-B vaccine-induced antigen-specific responses. Most of these BTMs
pathway and activation of STAT3 and STAT5B transcription fac- were related to DC activation, cell cycle, and IFN/antiviral sensing.
tors. These transcription factors are also involved in the expression The genes of these BTMs were comprised in the whole blood IFN
of inflammatory cytokines, such as IL-6, and receptors (37, 56) and signatures that were associated with RTS,S vaccination and protec-
differentiation and generation of TH17, T follicular helper (T FH) tion in a trial with naïve adults (18). Another analysis of the same
cells, and B cells (79, 80). Studies in murine models have shown dataset also found IFN- signaling and NF-B pathways associated
that IL-6 mediates protective immunity against the preerythrocytic with RTS,S-induced protection (19). RTS,S/AS01 studies have shown
stages of malaria by inducing IL-1 and TNF, and during the eryth- no antigen-specific production of IFN- from CD4+ T cells (86),
rocytic stage of disease by controlling parasitemia through boost- but IFN- could be produced by NK cells via bystander activation
ing of specific IgG antibodies (81). TLR4 is an innate receptor that through CD4+ T cells producing IL-2 (88). In another RTS,S trial
recognizes pathogen-associated molecular patterns such as lipopoly- with malaria-naïve adults, Kazmin et al. (12) found BTMs of B
saccharide (LPS) and induces proinflammatory responses through and plasma cells positively associated with immunogenicity and
the canonical NF-B pathway and type I IFN responses through in- protection, and NK cells negatively associated with both protec-
terferon regulatory factor 3 (IRF3) (both responses found in the tion and antibody responses. We also detected RTS,S-induced re-
BTMs associated with protection) (82). It has also been shown that sponses in some NK BTMs, although they were not associated with
LPS induces the noncanonical NF-B pathway (83). The AS01E ad- protection. Our experimental strategy is different from the previous
studies in which they performed transcriptomics in whole blood or suggests potential for global protective responses to malaria. Pre-
PBMCs without stimulation, directly ex vivo, and early after vacci- immunization signatures and BTMs associated with protection could
nation. In contrast, we used in vitro recall responses at 1 month after indicate that the physiological status of the host at baseline could
the third vaccination, and we corrected for background responses determine vaccine efficacy and infection outcomes. Therefore, our
for each individual using the same number of cells. results could be valuable for identifying upfront low or nonvaccine
Some BTMs associated with protection in RTS,S vaccinees were responders and finding strategies to make nonresponders more reactive
also found in comparator vaccinees, particularly BTMs related to and hence improve vaccine take and protection. In addition, the
inflammatory/TLR/chemokine responses and monocytes. This postvaccination signatures of protection identified could be useful
probably reflects non–vaccine-specific responses. Most of these for vaccine clinical trials as surrogate markers to predict protection.
particular BTMs were the ones negatively enriched in RTS,S compared
to comparator vaccinees. Similar to RTS,S, we found repression of
BTMs of inflammatory/TLR responses and monocytes, and addition- MATERIALS AND METHODS
ally, antigen presentation and DC activation for CPS immunization. Detailed Supplementary Materials and Methods are available in the
However, as in RTS,S, many of these BTMs were positively associated Supplementary Materials.
with protection. Thus, despite down-regulation by immuniza-

tion, protected immunized individuals had less down-regulation Study design
than unprotected immunized volunteers. Many of these BTMs were We used samples from malaria-naïve adults from a CPS clinical
also found before immunization (as in comparators for the RTS,S trial (9) and children and infants (two different age cohorts) from
study), suggesting protection mechanisms independent of immunization. three African countries participating in the RTS,S/AS01E phase 3
For instance, protected individuals may have monocytes of higher clinical trial (MAL055, trial registered with ClinicalTrials.gov,
capacity to mediate Fc receptor/antibody-dependent responses, number NCT00866619) (3). Samples from all individuals partici-
leading to better control of infection. This finding highlights a po- pating in the CPS clinical trial were included (fig. S1). Samples from
tentially crucial role of monocytes in malaria protective responses. three sites and two age cohorts from the RTS,S/AS01E phase 3 trial
Limitations of our study include the lack of additional cohorts to were selected. Specifically, all malaria cases detected 1 to 12 months
prove the universal predictive capacity of the identified signatures. after vaccination from whom at least two aliquots of PBMC samples
Additionally, the transcriptomic analysis was conducted on in vitro were available 1 month after or before vaccination were included
stimulations instead of ex vivo whole blood. This experimental de- (fig. S2). Malaria cases were defined as individuals who sought care
sign may introduce more variability and difficult the interpretation at a health facility and had any P. falciparum asexual parasitemia by
of the identified signatures, particularly the ones at baseline. However, blood smear. Up to four controls were selected for each case based
it allows to attribute differential gene expression to the antigen- on site and age group. At the time of sample selection, all study
specific recall response, and we are more certain that differences in investigators were blinded to vaccination, and selection was per-
gene expression were less likely due to basal differences, in cell formed randomly. As expected, per MAL055 trial design, about
subset counts for instance, than in an ex vivo approach without two-third of children had received RTS,S/AS01E vaccination, and
stimulation. Nevertheless, we did observe some changes in cell one-third had received the comparator vaccine. Sample sizes and
subsets upon CPS immunization and between protected and non- study design are detailed in Fig. 1, figs. S1 to S3, and the Supplemen-
protected CPS-immunized volunteers, which could affect in part tary Materials. Previously cryopreserved PBMCs were stimulated
some of the module responses. A related limitation is the modest in vitro with CSP peptide pool or PfRBC and their respective back-
acquired cellular responses induced by the immunizations, particularly ground controls, DMSO, and uRBC. RNA was extracted for gene
for RTS,S. Weak CSP-specific cell responses (87) and low frequency microarray analysis. Differential gene expression analysis using
of CSP-specific CD4+ T cells were detected in RTS,S vaccinees in linear regression models and subsequently GSEA was performed
the phase 3 trial using similar stimulation conditions (86), which is (see Supplementary Materials and Methods). Of 688 hybridized
probably the reason behind the absence of statistical significance in samples from the RTS,S study, 2 had no CEL data, and 31 were
the differential gene expression analysis. Despite the overall weak discarded because they were outliers (did not pass the quality check
differential responses elicited upon in vitro recall, we have obtained criteria). All microarray data were also used for protein network–
meaningful responses through both gene set analyses and the systems based models. Proteins behaving differently in the models were
biology approach. The fact that we did not detect a differential expression down-selected using data science methods, leading to identification
in the genes encoding the proteins from the signatures of protection of pre- and postimmunization signatures of protection, consisting
obtained with the network-based analysis is not unexpected, because of three to five proteins, with accuracies of >70%. For validation of
the network-based analysis can highlight other proteins that are signatures of protection, samples from another CPS trial (10) and
highly related to differentially expressed genes, despite not being samples from independent children and infants participating in the
found significant by conventional differential gene expression analysis. RTS,S/AS01E phase 3 trial were used. The gene expression of 70
Our results confirm the potential and feasibility of systems biol- selected genes from the signatures of protection was measured by
ogy approaches to identifying molecular signatures of protection qRT-PCR. Among the validated signatures, three were selected for
against complex diseases in field studies, where individuals are un- development of predictive models using artificial neural networks.
der heterogeneous conditions, and even in vulnerable populations
such as children. The finding that BTMs induced by CSP recall Samples
responses and associated with RTS,S protection in infants and chil- We used cryopreserved PBMCs obtained by density gradient cen-
dren from endemic areas were also found in PfRBC recall responses trifugation from venous whole blood. PBMCs were stimulated with
associated with protection in the CPS trial in malaria-naïve adults P. falciparum NF54-infected RBC for 16 hours and with CSP peptides
for 24 hours. uRBCs and DMSO were used as negative control stimuli Table S1. Predictive models obtained by artificial neural networks.
Table S2. Validation qRT-PCR values and prediction results for the selected CPS
for each individual and time point. RNA was extracted for microarray
preimmunization signature of protection.
analysis. For RTS,S/AS01E transcriptional validation, RNA from Table S3. Validation qRT-PCR values and prediction results for the selected CPS
12-hour stimulations with CSP and DMSO performed freshly on postimmunization signature of protection.
site of blood collection were used. Remaining PBMCs were stained Table S4. Validation qRT-PCR values and prediction results for the selected RTS,S/AS01E
and acquired on a Cyan ADP nine-color flow cytometer (Beckman postimmunization signature of protection.
Data file S1. Gene lists differential gene expression CPS immunogenicity and protection.
Coulter) at Radboud University Medical Center (RUMC) and an LSRII Data file S2. Gene list correlations PfRBC FC and CPS outcomes and frequencies of gene
cytometer (BD Biosciences) at ISGlobal. Flow cytometry data were ontology biological processes terms and BTMs of genes correlated with CPS outcomes.
analyzed using FlowJo software. A representative gating strategy for Data file S3. Gene lists differential gene expression RTS,S/AS01E immunogenicity and
the panel is shown in fig. S10A. protection.
Data file S4. Summary of GSEA and differential gene expression results.
Data file S5. Summary of GSEA and WGCNA results.
Gene expression microarrays and qRT-PCR Data file S6. Selected genes for validation for CPS and RTS,S/AS01E.
Human Gene 2.1 ST microarrays in 96-array plates (Affymetrix) Data file S7. Postvalidation signatures of protection with accuracy data for CPS and RTS,S/
and qRT-PCR (Fluidigm System Dynamics Array) for validation AS01.
were performed using provider’s recommendations. Data file S8. CSP 15-mer peptides, corresponding to CSP region of RTS,S and predicted CD4+
and CD8+ T cell epitopes.

Data file S9. Comparisons and limma contrasts.
Statistical analysis Data file S10. Gene list for validation.
Wilcoxon signed rank and Wilcoxon rank-sum tests were used to References (98–133)
compare cell frequencies between groups. Differential gene expression View/request a protocol for this paper from Bio-protocol.
(DGE) analysis was performed using linear models using the limma
package in R/Bioconductor (89), and empirical Bayes methods were REFERENCES AND NOTES
used. Spearman correlations were used to assess the association of 1. WHO, World Malaria Report 2019 (2019).
FC gene expression and total cumulative parasitemia during CPS 2. WHO Malaria Policy Advisory Committee and Secretariat, Malaria Policy Advisory
immunization and the time of patency. For GSEA, CAMERA, and Committee to the WHO: Conclusions and recommendations of March 2013 meeting.
Malar. J. 12, 213 (2013).
Tmod, analyses were performed using preranked analyses of DGE
3. RTS; S Clinical Trials Partnership, Efficacy and safety of RTS,S/AS01 malaria vaccine with or
(all genes regardless of statistical significance) and were performed without a booster dose in infants and children in Africa: Final results of a phase 3,
in R. Fisher’s exact tests were used to assess whether enrichments individually randomised, controlled trial. Lancet 386, 31–45 (2015).
were due to genes relatively relevant in terms of differential gene ex- 4. S. J. Draper, B. K. Sack, C. R. King, C. M. Nielsen, J. C. Rayner, M. K. Higgins, C. A. Long,
pression. Network-based mathematical models were built for non- R. A. Seder, Malaria vaccines: Recent advances and new horizons. Cell Host Microbe 24,
43–56 (2018).
protected and protected individuals for each of the cohorts using the 5. S. A. Jongo, S. A. Shekalaghe, L. W. P. Church, A. J. Ruben, T. Schindler, I. Zenklusen,
Therapeutic Performance Mapping System (Anaxomics Biotech) T. Rutishauser, J. Rothen, A. Tumbo, C. Mkindi, M. Mpina, A. T. Mtoro, A. S. Ishizuka,
(27, 28, 90–96). To center the network analysis, the known key pro- K. R. Kassim, F. A. Milando, M. Qassim, O. A. Juma, S. Mwakasungula, B. Simon, E. R. James,
teins with a relevant role in malaria immune responses were iden- Y. Abebe, N. Kc, S. Chakravarty, E. Saverino, B. M. Bakari, P. F. Billingsley, R. A. Seder,
C. Daubenberger, B. K. L. Sim, T. L. Richie, M. Tanner, S. Abdulla, S. L. Hoffman, Safety,
tified by a review of the scientific literature (data file S8). Proteins
immunogenicity, and protective efficacy against controlled human malaria infection
that behaved differently in the protected and nonprotected models of Plasmodium falciparum sporozoite vaccine in tanzanian adults. Am. J. Trop. Med. Hyg.
were selected. Combinations of three to five proteins were selected 99, 338–349 (2018).
after applying a data science analysis (different feature selection 6. J. G. Beeson, L. Kurtovic, C. Dobaño, D. H. Opi, J.-A. Chan, G. Feng, M. F. Good, L. Reiling,
approaches and base classifier methods using LOO cross-validation). M. J. Boyle, Challenges and strategies for developing efficacious and long-lasting malaria
vaccines. Sci. Transl. Med. 11, eaau1458 (2019).
To select the best gene candidates among the identified signatures 7. C. Dobaño, H. Sanz, H. Sorgho, D. Dosoo, M. Mpina, I. Ubillos, R. Aguilar, T. Ford,
for prospective validation, an approach based on Greedy algorithms N. Díez-Padrisa, N. A. Williams, A. Ayestaran, O. Traore, A. J. Nhabomba, C. Jairoce,
(97) was applied. The qRT-PCR data from the validation individuals J. Waitumbi, S. T. Agnandji, S. Kariuki, S. Abdulla, J. J. Aponte, B. Mordmüller, K. P. Asante,
were used to confirm the signatures of protection. The predictive S. Owusu-Agyei, H. Tinto, J. J. Campo, G. Moncunill, B. Gyan, C. Valim, C. Daubenberger,
Concentration and avidity of antibodies to different circumsporozoite epitopes correlate
mathematical functions developed for selected validated signatures
with RTS,S/AS01E malaria vaccine efficacy. Nat. Commun. 10, 2174 (2019).
were obtained through artificial neural networks. A LOO cross- 8. E. M. Bijker, G. J. H. Bastiaens, A. C. Teirlinck, G.-J. van Gemert, W. Graumans,
validation strategy was used to assess the accuracy. P values were M. van de Vegte-Bolmer, R. Siebelink-Stoter, T. Arens, K. Teelen, W. Nahrendorf,
adjusted by the false discovery rate (FDR) as described by Benjamini E. J. Remarque, W. Roeffen, A. Jansens, D. Zimmerman, M. Vos, B. C. L. van Schaijk,
and Hochberg and Bonferroni methods. J. Wiersma, A. J. A. M. van der Ven, Q. de Mast, L. van Lieshout, J. J. Verweij, C. C. Hermsen,
A. Scholzen, R. W. Sauerwein, Protection against malaria after immunization by
chloroquine prophylaxis and sporozoites is mediated by preerythrocytic immunity.
SUPPLEMENTARY MATERIALS Proc. Natl. Acad. Sci. U.S.A. 110, 7862–7867 (2013).
stm.sciencemag.org/cgi/content/full/12/543/eaay8924/DC1 9. E. M. Bijker, A. C. Teirlinck, R. Schats, G.-J. van Gemert, M. van de Vegte-Bolmer,
Materials and Methods L. van Lieshout, J. IntHout, C. C. Hermsen, A. Scholzen, L. G. Visser, R. W. Sauerwein,
Fig. S1. Experimental design and sample sizes. Cytotoxic markers associate with protection against malaria in human volunteers
Fig. S2. CPS study profile and sample flow chart. immunized with Plasmodium falciparum sporozoites. J. Infect. Dis. 210, 1605–1615
Fig. S3. RTS,S study profile and sample flow chart. (2014).
Fig. S4. PCA of CPS gene expression data. 10. E. M. Bijker, R. Schats, J. M. Obiero, M. C. Behet, G. J. Van Gemert, M. Van De Vegte-Bolmer,
Fig. S5. Heatmaps and unsupervised hierarchical clustering in gene expression after cell W. Graumans, L. Van Lieshout, G. J. H. Bastiaens, K. Teelen, C. C. Hermsen, A. Scholzen,
stimulation for all CPS immunized volunteers. L. G. Visser, R. W. Sauerwein, Sporozoite immunization of human volunteers under
Fig. S6. PCA of RTS,S/AS01E gene expression data. mefloquine prophylaxis is safe, immunogenic and protective: A double-blind
Fig. S7. Heatmaps and unsupervised hierarchical clustering in gene expression after cell randomized controlled clinical trial. PLOS ONE 9, e112910 (2014).
stimulation for all RTS,S and comparator vaccinees. 11. H. I. Nakaya, B. Pulendran, Vaccinology in the era of high-throughput biology. Philos.
Fig. S8. Frequencies of leukocyte subsets after resting of PBMCs. Trans. R. Soc. B Biol. Sci. 370, 20140146 (2015).
12. D. Kazmin, H. I. Nakaya, E. K. Lee, M. J. Johnson, R. van der Most, R. A. van den Berg, on adipose tissue from obese patients as age-, diabetes- and gender-specific hallmarks.
W. R. Ballou, E. Jongert, U. Wille-Reece, C. Ockenhouse, A. Aderem, D. E. Zak, J. Sadoff, Sci. Rep. 6, 25756 (2016).
J. Hendriks, J. Wrammert, R. Ahmed, B. Pulendran, Systems analysis of protective immune 29. A. Madamanchi, S. A. Santoro, M. M. Zutter, in Advances in Experimental Medicine and
responses to RTS,S malaria vaccination in humans. Proc. Natl. Acad. Sci. U.S.A. 114, Biology (2014), vol. 819, pp. 41–60.
2425–2430 (2017). 30. S. Karunakaran, V. Ravindranath, Activation of p38 MAPK in the substantia nigra leads
13. T. D. Querec, R. S. Akondy, E. K. Lee, W. Cao, H. I. Nakaya, D. Teuwen, A. Pirani, K. Gernert, to nuclear translocation of NF-B in MPTP-treated mice: Implication in Parkinson’s
J. Deng, B. Marzolf, K. Kennedy, H. Wu, S. Bennouna, H. Oluoch, J. Miller, R. Z. Vencio, disease. J. Neurochem. 109, 1791–1799 (2009).
M. Mulligan, A. Aderem, R. Ahmed, B. Pulendran, Systems biology approach predicts 31. Z. Li, G. J. Nabel, A new member of the I kappa B protein family, I kappa B epsilon,
immunogenicity of the yellow fever vaccine in humans. Nat. Immunol. 10, 116–125 inhibits RelA (p65)-mediated NF-kappa B transcription. Mol. Cell. Biol. 17, 6184–6190
(2009). (1997).
14. H. I. Nakaya, T. Hagan, S. S. Duraisingham, E. K. Lee, M. Kwissa, N. Rouphael, D. Frasca, 32. K. Ebner, A. Bandion, B. R. Binder, R. de Martin, J. A. Schmid, GMCSF activates NF-kB via
M. Gersten, A. K. Mehta, R. Gaujoux, G.-M. Li, S. Gupta, R. Ahmed, M. J. Mulligan, direct interaction of the GMCSF receptor with IkB kinase. Blood 102, 192–199 (2003).
S. Shen-Orr, B. B. Blomberg, S. Subramaniam, B. Pulendran, Systems analysis of immunity 33. S. Kohli, V. Kohli, Role of RANKL-RANK/osteoprotegerin molecular complex in bone
to influenza vaccination across multiple years and in diverse populations reveals shared remodeling and its immunopathologic implications. Indian J. Endocrinol. Metab. 15,
molecular signatures. Immunity 43, 1186–1198 (2015). 175–181 (2011).
15. G. Obermoser, S. Presnell, K. Domico, H. Xu, Y. Wang, E. Anguiano, L. Thompson-Snipes, 34. R. Schreck, P. A. Baeuerle, NF-KB inducible transcriptional activator of the granulocyte-
R. Ranganathan, B. Zeitner, A. Bjork, D. Anderson, C. Speake, E. Ruchaud, J. Skinner, macrophage colony-stimulating factor gene. Mol. Cell. Biol. 10, 1281–1286 (1990).
L. Alsina, M. Sharma, H. Dutartre, A. Cepika, E. Israelsson, P. Nguyen, Q.-A. Nguyen, 35. Y. Shi, C. H. Liu, A. I. Roberts, J. Das, G. Xu, G. Ren, Y. Zhang, L. Zhang, Z. R. Yuan,

A. C. Harrod, S. M. Zurawski, V. Pascual, H. Ueno, G. T. Nepom, C. Quinn, D. Blankenship, H. S. W. Tan, G. Das, S. Devadas, Granulocyte-macrophage colony-stimulating factor
K. Palucka, J. Banchereau, D. Chaussabel, Systems scale interactive exploration reveals (GM-CSF) and T-cell responses: What we do and don’t know. Cell Res. 16, 126–133
quantitative and qualitative differences in response to influenza and pneumococcal (2006).
vaccines. Immunity 38, 831–844 (2013). 36. E. G. Elias, J. L. Zapas, S. L. Beam, S. D. Brown, GM-CSF and IL-2 combination as adjuvant
16. S. Li, N. Rouphael, S. Duraisingham, S. Romero-Steiner, S. Presnell, C. Davis, D. S. Schmidt, therapy in cutaneous melanoma: Early results of a phase II clinical trial. Oncology 19,
S. E. Johnson, A. Milton, G. Rajam, S. Kasturi, G. M. Carlone, C. Quinn, D. Chaussabel, 15–18 (2005).
A. K. Palucka, M. J. Mulligan, R. Ahmed, D. S. Stephens, H. I. Nakaya, B. Pulendran, 37. T. Kanai, S. Seki, J. A. Jenks, A. Kohli, T. Kawli, D. P. Martin, M. Snyder, R. Bacchetta,
Molecular signatures of antibody responses derived from a systems biology study of five K. C. Nadeau, Identification of STAT5A and STAT5B target genes in human T cells.
human vaccines. Nat. Immunol. 15, 195–204 (2014). PLOS ONE 9, e86790 (2014).
17. D. E. Zak, E. Andersen-Nissen, E. R. Peterson, A. Sato, M. K. Hamilton, J. Borgerding, 38. D. Kimura, M. Miyakoda, K. Honma, Y. Shibata, M. Yuda, Y. Chinzei, K. Yui, Production
A. T. Krishnamurty, J. T. Chang, D. J. Adams, T. R. Hensley, A. I. Salter, C. A. Morgan, of IFN- by CD4+ T cells in response to malaria antigens is IL-2 dependent. Int. Immunol.
A. C. Duerr, S. C. De Rosa, A. Aderem, M. J. McElrath, Merck Ad5/HIV induces broad innate 22, 941–952 (2010).
immune activation that predicts CD8+ T-cell responses but is attenuated by preexisting 39. S. Asada, H. Daitoku, H. Matsuzaki, T. Saito, T. Sudo, H. Mukai, S. Iwashita, K. Kako, T. Kishi,
Ad5 immunity. Proc. Natl. Acad. Sci. U.S.A. 109, E3503–E3512 (2012). Y. Kasuya, A. Fukamizu, Mitogen-activated protein kinases, Erk and p38, phosphorylate
18. D. Rinchai, S. Presnell, M. Vidal, S. Dutta, V. Chauhan, D. Cavanagh, G. Moncunill, and regulate Foxo1. Cell. Signal. 19, 519–527 (2007).
C. Dobaño, D. Chaussabel, Blood interferon signatures putatively link lack of protection 40. B. Cai, Z. Xia, p38 MAP kinase mediates arsenite-induced apoptosis through FOXO3a
conferred by the RTS,S recombinant malaria vaccine to an antigen-specific ige response. activation and induction of Bim transcription. Apoptosis 13, 803–810 (2008).
F1000Research 4, 919 (2015). 41. S. M. Hedrick, R. H. Michelini, A. L. Doedens, A. W. Goldrath, E. L. Stone, FOXO
19. R. A. van den Berg, M. Coccia, W. R. Ballou, K. E. Kester, C. F. Ockenhouse, J. Vekemans, transcription factors throughout T cell biology. Nat. Rev. Immunol. 12, 649–661 (2012).
E. Jongert, A. M. Didierlaurent, R. G. van der Most, Predicting RTS,S vaccine-mediated 42. C. N. Nguetse, P. G. Kremsner, T. P. Velavan, FOXO3A regulatory polymorphism
protection from transcriptomes in a malaria-challenge clinical trial. Front. Immunol. 8, 557 and susceptibility to severe malaria in Gabonese children. Immunogenetics 67, 67–71
(2017). (2015).
20. T. M. Tran, E. M. Bijker, M. C. Haks, T. H. M. Ottenhoff, L. Visser, R. Schats, P. Venepally, 43. P. C. Delekta, I. J. Apel, S. Gu, K. Siu, Y. Hattori, L. M. McAllister-Lucas, P. C. Lucas,
H. Lorenzi, P. D. Crompton, R. W. Sauerwein, Whole-blood transcriptomic signatures Thrombin-dependent NF-b activation and monocyte/endothelial adhesion are
induced during immunization by chloroquine prophylaxis and Plasmodium falciparum mediated by the CARMA3·Bcl10·MALT1 signalosome. J. Biol. Chem. 285, 41432–41442
sporozoites. Sci. Rep. 9, 8386 (2019). (2010).
21. W. Nahrendorf, A. Scholzen, E. M. Bijker, A. C. Teirlinck, G. J. H. Bastiaens, R. Schats, 44. P. Pontrelli, M. Cariello, F. Rascio, M. Gigante, R. Verrienti, T. Tataranni, G. Castellano,
C. C. Hermsen, L. G. Visser, J. Langhorne, R. W. Sauerwein, Memory B-cell and antibody C. Divella, E. Ranieri, G. Stallone, L. Gesualdo, G. Grandaliano, Thrombin may modulate
responses induced by Plasmodium falciparum sporozoite immunization. J. Infect. Dis. dendritic cell activation in kidney transplant recipients with delayed graft function.
210, 1981–1990 (2014). Nephrol. Dial. Transplant. 30, 1480–1487 (2015).
22. E. M. Bijker, K. Nganou-Makamdop, G.-J. van Gemert, F. Zavala, I. Cockburn, 45. X. Li, T. Simmet, Y. Laumonnier, S. Paskas, T. Syrovets, Mature dendritic cells express
R. W. Sauerwein, Studying the effect of chloroquine on sporozoite-induced protection functional thrombin receptors triggering chemotaxis and CCL18/pulmonary
and immune responses in Plasmodium berghei malaria. Malar. J. 14, 130 (2015). and activation-regulated chemokine induction. J. Immunol. 181, 1215–1223 (2014).
23. M. Kanehisa, M. Furumichi, M. Tanabe, Y. Sato, K. Morishima, KEGG: New perspectives 46. H. R. Gralnick, S. B. Williams, B. S. Coller, Fibrinogen competes with von Willebrand factor
on genomes, pathways, diseases and drugs. Nucleic Acids Res. 45, D353–D361 (2017). for binding to the glycoprotein IIb/IIIa complex when platelets are stimulated
24. D. Croft, A. F. Mundo, R. Haw, M. Milacic, J. Weiser, G. Wu, M. Caudy, P. Garapati, with thrombin. Blood 64, 797–800 (1984).
M. Gillespie, M. R. Kamdar, B. Jassal, S. Jupe, L. Matthews, B. May, S. Palatnik, K. Rothfels, 47. J. R. Groom, A. D. Luster, CXCR3 in T cell function. Exp. Cell Res. 317, 620–631 (2011).
V. Shamovsky, H. Song, M. Williams, E. Birney, H. Hermjakob, L. Stein, P. D’Eustachio, The 48. M. G. Overstreet, Y. C. Chen, I. A. Cockburn, Y. C. Chen, F. Zavala, Protective CD8+ T cells
Reactome pathway knowledgebase. Nucleic Acids Res. 42, D472–D477 (2014). against Plasmodium liver stages: Immunobiology of an ‘unnatural’ immune response.
25. D. Chaussabel, C. Quinn, J. Shen, P. Patel, C. Glaser, N. Baldwin, D. Stichweh, Immunol. Rev. 225, 272–283 (2008).
D. Blankenship, L. Li, I. Munagala, L. Bennett, F. Allantaz, A. Mejias, M. Ardura, E. Kaizer, 49. P. W. Gold, J. Licinio, M. G. Pavlatou, Pathological parainflammation and endoplasmic
L. Monnet, W. Allman, H. Randall, D. Johnson, A. Lanier, M. Punaro, K. M. Wittkowski, reticulum stress in depression: Potential translational targets through the CNS insulin,
P. White, J. Fay, G. Klintmalm, O. Ramilo, A. K. Palucka, J. Banchereau, V. Pascual, A klotho and PPAR- systems. Mol. Psychiatry 18, 154–165 (2013).
modular analysis framework for blood genomics studies: Application to systemic lupus 50. Y. Maekawa, K. Ishikawa, O. Yasuda, R. Oguro, H. Hanasaki, I. Kida, Y. Takemura, M. Ohishi,
erythematosus. Immunity 29, 150–164 (2008). T. Katsuya, H. Rakugi, Klotho suppresses TNF--induced expression of adhesion
26. M. J. Mason, G. Fan, K. Plath, Q. Zhou, S. Horvath, Signed weighted gene co-expression molecules in the endothelium and attenuates NF-B activation. Endocrine 35, 341–346
network analysis of transcriptional regulation in murine embryonic stem cells. (2009).
BMC Genomics 10, 327 (2009). 51. B. Ma, M. Fey, M. O. Hottiger, WNT/-catenin signaling inhibits CBP-mediated RelA
27. G. Jorba, J. Aguirre-Plans, V. Junet, C. Segú-Vergés, J. L. Ruiz, A. Pujol, N. Fernández-Fuentes, acetylation and expression of proinflammatory NF-B target genes. J. Cell Sci. 128,
J. M. Mas, B. Oliva, In-silico simulated prototype-patients using TPMS technology 2430–2436 (2015).
to study a potential adverse effect of sacubitril and valsartan. PLOS ONE 15, e0228926 52. E. Laughner, P. Taghavi, K. Chiles, P. C. Mahon, G. L. Semenza, HER2 (neu) signaling
(2020). increases the rate of hypoxia-inducible factor 1 (HIF-1) synthesis: Novel mechanism
28. M. Gómez-Serrano, E. Camafeita, E. García-Santos, J. A. López, M. A. Rubio, for HIF-1-mediated vascular endothelial growth factor expression. Mol. Cell. Biol. 21,
A. Sánchez-Pernaute, A. Torres, J. Vázquez, B. Peral, Proteome-wide alterations 3995–4004 (2001).
53. G. L. Semenza, F. Agani, G. Booth, J. Forsythe, N. Iyer, B. H. Jiang, S. Leung, R. Roe, unsolved mysteries of a deadly infectious disease. Annu. Rev. Immunol. 32, 157–187
C. Wiener, A. Yu, Structural and functional analysis of hypoxia-inducible factor 1. (2014).
Kidney Int. 51, 553–555 (1997). 78. B. J. McMorran, V. M. Marshall, C. de Graaf, K. E. Drysdale, M. Shabbar, G. K. Smyth,
54. E. E. Johnson, M. Wessling-Resnick, Iron metabolism and the innate immune response J. E. Corbin, W. S. Alexander, S. J. Foote, Platelets kill intraerythrocytic malarial parasites
to infection. Microbes Infect. 14, 207–216 (2012). and mediate survival to infection. Science 323, 797–800 (2009).
55. J. Liu, J. A. Kern, Neuregulin-1 activates the JAK-STAT pathway and regulates lung 79. E. K. Deenick, S. J. Pelham, A. Kane, C. S. Ma, Signal transducer and activator
epithelial cell proliferation. Am. J. Respir. Cell Mol. Biol. 27, 306–313 (2002). of transcription 3 control of human T and B cell responses. Front. Immunol. 9, 168 (2018).
56. Y. Xia, P. N. Khoi, H. J. Yoon, S. Lian, Y. E. Joo, K. O. Chay, K. K. Kim, Y. Do Jung, Piperine 80. D. R. McIlwain, M. Grusdat, V. I. Pozdeev, H. C. Xu, P. Shinde, C. Reardon, Z. Hao, M. Beyer,
inhibits IL-1-induced IL-6 expression by suppressing p38 MAPK and STAT3 activation A. Bergthaler, D. Häussinger, G. P. Nolan, K. S. Lang, P. A. Lang, T-cell STAT3 is required
in gastric cancer cells. Mol. Cell. Biochem. 398, 147–156 (2015). for the maintenance of humoral immunity to LCMV. Eur. J. Immunol. 45, 418–427 (2015).
57. F. Li, X. J. Li, L. Qiao, F. Shi, W. Liu, Y. Li, Y. P. Dang, W. J. Gu, X. G. Wang, W. Liu, MiR-98 81. S. Pied, A. Civas, F. Berlot-Picard, L. Renia, F. Miltgen, M. Gentilini, J. Doly, D. Mazier, IL-6
suppresses melanoma metastasis through a negative feedback loop with its target gene induced by IL-1 inhibits malaria pre-erythrocytic stages but its secretion is down-regulated
IL-6. Exp. Mol. Med. 46, e116 (2014). by the parasite. J. Immunol. 148, 197–201 (1992).
58. H. M. Oh, C. R. Yu, N. Golestaneh, A. Amadi-Obi, Y. S. Lee, A. Eseonu, R. M. Mahdi, 82. S. E. Doyle, S. A. Vaidya, R. O’Connell, H. Dadgostar, P. W. Dempsey, T. T. Wu, G. Rao,
C. E. Egwuagu, STAT3 protein promotes T-cell survival and inhibits interleukin-2 R. Sun, M. E. Haberland, R. L. Modlin, G. Cheng, IRF3 mediates a TLR3/TLR4-specific
production through up-regulation of class O forkhead transcription factors. J. Biol. Chem. antiviral gene program. Immunity 17, 251–263 (2002).
286, 30888–30897 (2011). 83. B. Mordmüller, D. Krappmann, M. Esen, E. Wegener, C. Scheidereit, Lymphotoxin
59. B. A. Witthuhn, M. D. Williams, H. Kerawalla, F. M. Uckun, Differential substrate and lipopolysaccharide induce NF-B-p52 generation by a co-translational mechanism.

recognition capabilities of Janus family protein tyrosine kinases within the interleukin 2 EMBO Rep. 4, 82–87 (2003).
receptor (I12R) system: Jak3 as a potential molecular target for treatment of leukemias 84. A. M. Didierlaurent, C. Collignon, P. Bourguignon, S. Wouters, K. Fierens, M. Fochesato,
with a hyperactive Jak-Stat signaling machinery. Leuk. Lymphoma 32, 289–297 (1999). N. Dendouga, C. Langlet, B. Malissen, B. N. Lambrecht, N. Garçon, M. Van Mechelen,
60. H. Cheng, J. A. Ross, J. A. Frost, R. A. Kirken, Phosphorylation of human Jak3 at tyrosines S. Morel, Enhancement of adaptive immunity by the human vaccine adjuvant AS01
904 and 939 positively regulates its activity. Mol. Cell. Biol. 28, 2271–2282 (2008). depends on activated dendritic cells. J. Immunol. 193, 1920–1930 (2014).
61. J. L. Macdonald-Obermann, L. J. Pike, Different epidermal growth factor (EGF) receptor 85. J. Z. Oh, R. Ravindran, B. Chassaing, F. A. Carvalho, M. S. Maddur, M. Bower, P. Hakimpour,
ligands show distinct kinetics and biased or partial agonism for homodimer K. P. Gill, H. I. Nakaya, F. Yarovinsky, R. B. Sartor, A. T. Gewirtz, B. Pulendran, TLR5-
and heterodimer formation. J. Biol. Chem. 289, 26178–26188 (2014). mediated sensing of gut microbiota is necessary for antibody responses to seasonal
62. Z. Ren, T. S. Schaefer, ErbB-2 activates Stat3 in a Src- and JAK2-dependent manner. influenza vaccination. Immunity 41, 478–492 (2014).
J. Biol. Chem. 277, 38486–38493 (2002). 86. G. Moncunill, S. C. De Rosa, A. Ayestaran, A. J. Nhabomba, M. Mpina, K. W. Cohen,
63. H. W. Lo, S. C. Hsu, M. Ali-Seyed, M. Gunduz, W. Xia, Y. Wei, G. Bartholomeusz, J. Y. Shih, C. Jairoce, T. Rutishauser, J. J. Campo, J. Harezlak, H. Sanz, N. Díez-Padrisa, N. A. Williams,
M. C. Hung, Nuclear interaction of EGFR and STAT3 in the activation of the iNOS/NO D. Morris, J. J. Aponte, C. Valim, C. Daubenberger, C. Dobaño, M. J. McElrath, RTS,S/AS01E
pathway. Cancer Cell 7, 575–589 (2005). malaria vaccine induces memory and polyfunctional T cell responses in a pediatric
64. M. A. Olayioye, I. Beuvink, K. Horsch, J. M. Daly, N. E. Hynes, ErbB receptor-induced African phase III trial. Front. Immunol. 8, 1008 (2017).
activation of Stat transcription factors is mediated by Src tyrosine kinases. J. Biol. Chem. 87. G. Moncunill, M. Mpina, A. J. Nhabomba, R. Aguilar, A. Ayestaran, H. Sanz, J. J. Campo,
274, 17209–17218 (1999). C. Jairoce, D. Barrios, Y. Dong, N. Díez-Padrisa, J. F. Fernandes, S. Abdulla, J. Sacarlal,
65. A. Pezet, H. Buteau, P. A. Kelly, M. Edery, The last proline of Box 1 is essential N. A. Williams, J. Harezlak, B. Mordmüller, S. T. Agnandji, J. J. Aponte, C. Daubenberger,
for association with JAK2 and functional activation of the prolactin receptor. C. Valim, C. Dobaño, Distinct helper T cell type 1 and 2 responses associated with malaria
Mol. Cell. Endocrinol. 129, 199–208 (1997). protection and risk in RTS,S/AS01E vaccinees. Clin. Infect. Dis. 65, 746–755 (2017).
66. T. Yamauchi, K. Ueki, K. Tobe, H. Tamemoto, N. Sekine, M. Wada, M. Honjo, M. Takahashi, 88. A. Horowitz, J. C. R. Hafalla, E. King, J. Lusingu, D. Dekker, A. Leach, P. Moris, J. Cohen,
T. Takahashi, H. Hirai, T. Tushima, Y. Akanuma, T. Fujita, I. Komuro, Y. Yazaki, T. Kadowaki, J. Vekemans, T. Villafana, P. H. Corran, P. Bejon, C. J. Drakeley, L. von Seidlein, E. M. Riley,
Tyrosine phosphorylation of the EGF receptor by the kinase Jak2 is induced by growth Antigen-specific IL-2 secretion correlates with NK cell responses after immunization
hormone. Nature 390, 91–96 (1997). of Tanzanian children with the RTS,S/AS01 malaria vaccine. J. Immunol. 188, 5054–5062
67. M. Funaba, C. M. Zimmerman, L. S. Mathews, Modulation of Smad2-mediated signaling (2012).
by extracellular signal-regulated kinase. J. Biol. Chem. 277, 41361–41368 (2002). 89. M. E. Ritchie, B. Phipson, D. Wu, Y. Hu, C. W. Law, W. Shi, G. K. Smyth, limma powers
68. L. Adalid-Peralta, G. Fragoso, A. Fleury, E. Sciutto, Mechanisms underlying the induction differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res.
of regulatory T cells and its relevance in the adaptive immune response in parasitic 43, e47 (2015).
infections. Int. J. Biol. Sci. 7, 1412–1426 (2011). 90. R. Valls, A. Pujol, L. Artigas, J. M. Mas, “Anaxomics’ Methodologies—Understanding the
69. R. V. Luckheeram, R. Zhou, A. D. Verma, B. Xia, CD4+ T cells: Differentiation and functions. complexity of biological processes,” White Paper (2017).
Clin. Dev. Immunol. 2012, 925135 (2012). 91. D. Romeo-Guitart, J. Forés, M. Herrando-Grabulosa, R. Valls, T. Leiva-Rodríguez, E. Galea,
70. I. Dikic, G. Tokiwa, S. Lev, S. A. Courtneidge, J. Schlessinger, A role for Pyk2 and Src F. González-Pérez, X. Navarro, V. Petegnief, A. Bosch, M. Coma, J. M. Mas, C. Casas,
in linking G-protein-coupled receptors with MAP kinase activation. Nature 383, 547–550 Neuroprotective drug for nerve trauma revealed using artificial intelligence. Sci. Rep. 8,
(1996). 1879 (2018).
71. S. M. Saud, M. R. Young, Y. L. Jones-Hall, L. Ileva, M. O. Evbuomwan, J. Wise, N. H. Colburn, 92. O. Iborra-Egea, C. Gálvez-Montón, S. Roura, I. Perea-Gil, C. Prat-Vidal, C. Soler-Botija,
Y. S. Kim, G. Bobe, Chemopreventive activity of plant flavonoid isorhamnetin in colorectal A. Bayes-Genis, Mechanisms of action of sacubitril/valsartan on cardiac remodeling:
cancer is mediated by oncogenic Src and -catenin. Cancer Res. 73, 5473–5484 (2013). A systems biology approach. npj Syst. Biol. Appl. 3, 12 (2017).
72. J. S. Tsang, P. L. Schwartzberg, Y. Kotliarov, A. Biancotto, Z. Xie, R. N. Germain, E. Wang, 93. S. Perera, L. Artigas, R. Mulet, J. M. Mas, T. Sardón, Systems biology applied to
M. J. Olnes, M. Narayanan, H. Golding, S. Moir, H. B. Dickler, S. Perl, F. Cheung; Baylor HIPC non-alcoholic fatty liver disease (NAFLD): Treatment selection based
Center; CHI Consortium, Global analyses of human immune variation reveal baseline on the mechanism of action of nutraceuticals. Forum Nutr. 13, 61–68 (2014).
predictors of postvaccination responses. Cell 157, 499–513 (2014). 94. M. Herrando-Grabulosa, R. Mulet, A. Pujol, J. M. Mas, X. Navarro, P. Aloy, M. Coma,
73. HIPC-CHI Signatures Project Team; HIPC-I Consortium, Multicohort analysis reveals C. Casas, Novel neuroprotective multicomponent therapy for amyotrophic lateral
baseline transcriptional predictors of influenza vaccination responses. Sci. Immunol. 2, sclerosis designed by networked systems. PLOS ONE 11, e0147626 (2016).
eaal4656 (2017). 95. Therapeutic Performance Mapping System (TPMS) Anaxomics; www.anaxomics.com/
74. G. M. Warimwe, H. A. Fletcher, A. Olotu, S. T. Agnandji, A. V. S. Hill, K. Marsh, P. Bejon, tpms.php.
Peripheral blood monocyte-to-lymphocyte ratio at study enrollment predicts efficacy 96. J. M. M. Benavente, A. P. Torras, P. A. Calaf, J. Farrés, “Methods and systems for identifying
of the RTS,S malaria vaccine: Analysis of pooled phase II clinical trial data. BMC Med. 11, molecules or processes of biological interest by using knowledge discovery in biological
184 (2013). data. 2010.” U.S. Patent Application No. 12/912,535.
75. L. E. Theill, W. J. Boyle, J. M. Penninger, RANK-L and RANK: T cells, bone loss, 97. T. H. Cormen, C. E. Leiserson, R. L. Rivest, Introduction to Algorithms, Chapter 16, Greedy
and mammalian evolution. Annu. Rev. Immunol. 20, 795–823 (2002). Algorithms (The MIT Press Cambridge, McGraw-Hill Book Company Boston, ed. 2, 2001).
76. N. Meednu, H. Zhang, T. Owen, W. Sun, V. Wang, C. Cistrone, J. Rangel-Moreno, L. Xing, 98. RTS; S Clinical Trials Partnership, Efficacy and safety of the RTS,S/AS01 malaria vaccine
J. H. Anolik, Production of RANKL by memory B cells: A link between B cells and bone during 18 months after vaccination: A phase 3 randomized, controlled trial in children
erosion in rheumatoid arthritis. Arthritis Rheum. 68, 805–816 (2016). and young infants at 11 African sites. PLOS Med. 11, e1001685 (2014).
77. P. D. Crompton, J. Moebius, S. Portugal, M. Waisberg, G. Hart, L. S. Garver, L. H. Miller, 99. R. Wang, J. Epstein, F. M. Baraceros, E. J. Gorak, Y. Charoenvit, D. J. Carucci, R. C. Hedstrom,
C. Barillas-Mury, S. K. Pierce, Malaria immunity in man and mosquito: Insights into N. Rahardjo, T. Gay, P. Hobart, R. Stout, T. R. Jones, T. L. Richie, S. E. Parker, D. L. Doolan,
J. Norman, S. L. Hoffman, Induction of CD4+ T cell-dependent CD8+ type 1 responses 25. C. Cortes, V. Vapnik, Support-vector networks. Mach. Learn. 20, 273–297 (1995).
1
in humans by a malaria DNA vaccine. Proc. Natl. Acad. Sci. U.S.A. 98, 10817–10822 126. L. Breiman, J. Friedman, R. Olshen, C. Stone, Classification and Regression Trees (CRC Press,
(2001). 1984).
100. J. M. González, K. Peter, F. Esposito, I. Nebié, J. M. Tiercy, A. Bonelo, M. Arévalo-Herrera, 127. S. Haykin, Neural Networks: A Comprehensive Foundation (Prentice Hall, 1998).
D. Valmori, P. Romero, S. Herrera, G. Corradin, J. A. López, HLA-A*0201 restricted CD8+ 128. H. Madsen, P. Thyregod. Introduction to General and Generalized Linear Models (Chapman
T-lymphocyte responses to malaria: Identification of new Plasmodium falciparum & Hall/CRC, 2011).
epitopes by IFN- ELISPOT. Parasite Immunol. 22, 501–514 (2000). 129. Y. Chang, C.-J. Hsieh, K.-W. Chang, M. Ringgaard, C.-J. Lin, Training and testing
101. S. Lê, J. Josse, F. Husson, FactoMineR: An R package for multivariate analysis. J. Stat. Softw. low-degree polynomial data mappings via linear SVM. J. Mach. Learn. Res. 11,
25, 1–18 (2008). 1471–1490 (2010).
102. G. R. Warnes, B. Bolker, L. Bonebakker, R. Gentleman, W. Huber, A. Liaw, T. Lumley, 130. H. Zou, T. Hastie, Regularization and variable selection via the elastic net. J. R. Stat. Soc.
M. Maechler, A. Magnusson, S. Moeller, M. Schwartz, gplots: Various R programming tools Ser. B 67, 301–320 (2005).
for plotting data (2019). 131. A. Y. Ng, in Twenty-First International Conference on Machine Learning - ICML ‘04 (ACM
103. S. Zhao, Y. Guo, Q. Sheng, Y. Shyr, Heatmap3: An improved heatmap package with more Press, 2004), p. 78.
powerful and convenient features. BMC Bioinformatics 15, P16 (2014). 132. R. Tibshirani, Regression shrinkage and selection via the lasso: A retrospective. J. R. Stat.
104. A. Subramanian, P. Tamayo, V. K. Mootha, S. Mukherjee, B. L. Ebert, M. A. Gillette, Soc. Ser. B 73, 273–282 (2011).
A. Paulovich, S. L. Pomeroy, T. R. Golub, E. S. Lander, J. P. Mesirov, Gene set enrichment 133. S. N. Wood, Fast stable direct fitting and smoothness selection for generalized additive
analysis: A knowledge-based approach for interpreting genome-wide expression models. J. R. Stat. Soc. Ser. B 70, 495–518 (2008).
profiles. Proc. Natl. Acad. Sci. U.S.A. 102, 15545–15550 (2005).

105. J. Weiner III, T. Domaszewska, tmod: An R package for general and multivariate
enrichment analysis. Peer J. 4, e2420v1 (2016). Acknowledgments: We thank all the trial volunteers and the staff from the Radboud
06.
1 M. Nakazawa, Package “fmsb,” 1–51 (2018). University Medical Center, Ifakara Health Institute and Bagamoyo Research and Training
107. B. H. Liu, H. Yu, K. Tu, C. Li, Y. X. Li, Y. Y. Li, DCGL: An R package for identifying differentially Centre (IHI-BRTC in Tanzania), Centre de Recherches Médicales de Lambaréné (CERMEL,
coexpressed genes and links from gene expression microarray data. Bioinformatics 26, Gabon) and Manhiça Health Research Center, and Fundação Manhiça (FM-CISM,
2637–2638 (2010). Mozambique), all of whom made this study possible. Specifically, we thank the principal
108. M. Kanehisa, S. Goto, M. Hattori, K. F. Aoki-Kinoshita, M. Itoh, S. Kawashima, T. Katayama, investigators of the RTS,S/AS01 phase 3 trial: S. Abdulla, P. Alonso, J. Sacarlal, P. Aide, and
M. Araki, M. Hirakawa, From genomics to chemical genomics: New developments E. Bijker, who lead the CPS trials. We also thank L. Puyol, D. Barrios, C. Amroune, and
in KEGG. Nucleic Acids Res. 34, D354–D357 (2006). C. Hernández from ISGlobal for laboratory logistics and management support and K. Stuart
109. S. Kerrien, B. Aranda, L. Breuza, A. Bridge, F. Broackes-Carter, C. Chen, M. Duesbury, and J. Carnes from the Center for Infectious Disease Research (CIDR, Seattle) who performed
M. Dumousseau, M. Feuermann, U. Hinz, C. Jandrasits, R. C. Jimenez, J. Khadake, some RNA extraction for validation. We thank the GlaxoSmithKline Biologicals S.A. for support
U. Mahadevan, P. Masson, I. Pedruzzi, E. Pfeiffenberger, P. Porras, A. Raghunath, in the conduct of the RTS,S study. Funding: This work was supported by the EU FP7-founded
B. Roechert, S. Orchard, H. Hermjakob, The IntAct molecular interaction database in 2012. SysMalVac project (grant number 305869-2); the Instituto de Salud Carlos III (ISCIII) (PI11
Nucleic Acids Res. 40, D841–D846 (2012). 00423); the Netherlands Organisation for Health Research and Development (ZonMw) (project
110. L. Salwinski, L. Licata, A. Winter, D. Thorneycroft, J. Khadake, A. Ceol, A. C. Aryamontri, 95110086); and the Agència de Gestió d’Ajuts Universitaris i de Recerca AGAUR (2014SGR991).
R. Oughtred, M. Livstone, L. Boucher, D. Botstein, K. Dolinski, T. Berardini, E. Huala, RTS,S sample collection was supported by the PATH Malaria Vaccine Initiative (MVI). C. Dobaño
M. Tyers, D. Eisenberg, G. Cesareni, H. Hermjakob, Recurated protein interaction datasets. was recipient of a Ramon y Cajal Contract from the Ministerio de Economía y Competitividad
Nat. Methods 6, 860–861 (2009). (RYC-2008-02631). G.M. was recipient of a Sara Borrell–ISCIII fellowship (CD010/00156) and
111. T. S. Keshava Prasad, R. Goel, K. Kandasamy, S. Keerthikumar, S. Kumar, S. Mathivanan, work was performed with the support of Department of Health, Catalan Government grant
D. Telikicherla, R. Raju, B. Shafreen, A. Venugopal, L. Balakrishnan, A. Marimuthu, (SLT006/17/00109). ISGlobal is a member of the CERCA Programme, Generalitat de Catalunya.
S. Banerjee, D. S. Somanathan, A. Sebastian, S. Rani, S. Ray, C. J. Harrys Kishore, S. Kanth, The funders had no role in study design, data collection and analysis, decision to publish, or
M. Ahmed, M. K. Kashyap, R. Mohmood, Y. L. Ramachandra, V. Krishna, B. A. Rahiman, preparation of the manuscript. Author contributions: G.M., A.S., J.J.C., L.O., L.P., J.M., E.M.P.,
S. Mohan, P. Ranganathan, S. Ramabadran, R. Chaerkady, A. Pandey, Human protein C.H.M.K., C. Daubenberger, B.M., S.T.A., R.S., and C. Dobaño designed the studies and/or
reference database–2009 update. Nucleic Acids Res. 37, D767–D772 (2009). experiments. A.N., C.J., J.J.C., M.M., C. Daubenberger, B.M., S.T.A., R.S., G.M., and C. Dobaño
112. E. Chautard, M. Fatoux-Ardore, L. Ballut, N. Thierry-Mieg, S. Ricard-Blum, MatrixDB, performed the clinical studies and collected samples and clinical data. G.M., A.S., M.M., and
the extracellular matrix interaction database. Nucleic Acids Res. 39, D235–D240 (2011). A.B.H. conducted laboratory experiments. G.M., A.S., L.O., J.M., L.P., D.A., R.V., J.M.M., and
113. H. W. Mewes, A. Ruepp, F. Theis, T. Rattei, M. Walter, D. Frishman, K. Suhre, M. Spannagl, M.D.-F. analyzed the data. H.S. and J.J.A. performed data management and sample selection.
K. F. X. Mayer, V. Stümpflen, A. Antonov, MIPS: Curated databases and comprehensive G.M., N.A.W., and C. Dobaño coordinated the study. G.M., C. Dobaño, R.V., and J.M.M.
secondary data resources in 2010. Nucleic Acids Res. 39, D220–D224 (2011). interpreted the data. G.M., A.S., and C. Dobaño wrote the manuscript. All authors read and
114. I. Xenarios, E. Fernandez, L. Salwinski, X. J. Duan, M. J. Thompson, E. M. Marcotte, approved the final manuscript. Competing interests: The authors declare that they have no
D. Eisenberg, DIP: The database of interacting proteins: 2001 update. Nucleic Acids Res. competing interests. Data and materials availability: All data associated with this study are
29, 239–241 (2001). present in the paper or the Supplementary Materials. Microarrays data are available in the
115. A. Chatr-Aryamontri, A. Zanzoni, A. Ceol, G. Cesareni, Searching the protein interaction manuscript and at the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo
space through the MINT database. Methods Mol. Biol. 484, 305–317 (2008). (accession no. GSE144826). Code for the systems biology analysis is based on the Therapeutic
116. V. Law, C. Knox, Y. Djoumbou, T. Jewison, A. C. Guo, Y. Liu, A. MacIejewski, D. Arndt, Performance Mapping System technology proprietary of Anaxomics Biotech. Code used for
M. Wilson, V. Neveu, A. Tang, G. Gabriel, C. Ly, S. Adamjee, Z. T. Dame, B. Han, Y. Zhou, the rest of analyses is available at GitHub (https://github.com/gmoncu/SysMalVac, DOI:
D. S. Wishart, DrugBank 4.0: Shedding new light on drug metabolism. Nucleic Acids Res. 10.5281/zenodo.3630760).
42, D1091–D1097 (2014).
117. M. Mitchell, An Introduction to Genetic Algorithms (MIT Press, 1998).
Submitted 29 July 2019
118. K. Kira, L. A. Rendell, in Proceedings of the Tenth National Conference on Artificial
Resubmitted 26 November 2019
Intelligence, AAAI’92 (AAAI Press, 1992), pp. 129–134.
Accepted 15 April 2020
19.
1 S. Theodoridis, K. Koutroumbas, Pattern Recognition, Fourth Edition (Academic Press, 2008).
Published 13 May 2020
120. M. P. Fay, M. A. Proschan, Wilcoxon-Mann-Whitney or t-test? On assumptions
10.1126/scitranslmed.aay8924
for hypothesis tests and multiple interpretations of decision rules. Stat. Surv. 4, 1–39
(2010).
121. J. A. Hanley, B. J. McNeil, A method of comparing the areas under receiver operating Citation: G. Moncunill, A. Scholzen, M. Mpina, A. Nhabomba, A. B. Hounkpatin, L. Osaba,
characteristic curves derived from the same cases. Radiology 148, 839–843 (1983). R. Valls, J. J. Campo, H. Sanz, C. Jairoce, N. A. Williams, E. M. Pasini, D. Arteta, J. Maynou,
122. K. Fukunaga, Introduction to Statistical Pattern Recognition (Academic Press, 1990). L. Palacios, M. Duran-Frigola, J. J. Aponte, C. H. M. Kocken, S. T. Agnandji, J. M. Mas,
123. S. Russell, P. Norvig, Artificial Intelligence: A Modern Approach (Prentice Hall, ed. 2, 2003). B. Mordmüller, C. Daubenberger, R. Sauerwein, C. Dobaño, Antigen-stimulated PBMC
124. D. Coomans, D. L. Massart, Alternative k-nearest neighbour rules in supervised pattern transcriptional protective signatures for malaria immunization. Sci. Transl. Med. 12, eaay8924
recognition. Anal. Chim. Acta 136, 15–27 (1982). (2020).
Antigen-stimulated PBMC transcriptional protective signatures for malaria immunization
Gemma Moncunill, Anja Scholzen, Maximillian Mpina, Augusto Nhabomba, Aurore Bouyoukou Hounkpatin, Lourdes
Osaba, Raquel Valls, Joseph J. Campo, Hèctor Sanz, Chenjerai Jairoce, Nana Aba Williams, Erica M. Pasini, David
Arteta, Joan Maynou, Lourdes Palacios, Miquel Duran-Frigola, John J. Aponte, Clemens H. M. Kocken, Selidji Todagbe
Agnandji, José Manuel Mas, Benjamin Mordmüller, Claudia Daubenberger, Robert Sauerwein and Carlota Dobaño
Sci Transl Med 12, eaay8924.

DOI: 10.1126/scitranslmed.aay8924

Predicting protection
A highly effective malaria vaccine would save many lives but correlates of protection remain ill-defined.
Moncunill et al. studied peripheral blood cells isolated from individuals that had received sporozoites under
chemoprophylaxis or the RTS,S vaccine. Transcriptomic analysis of gene expression after in vitro stimulation of
cells revealed preimmunization and postimmunization signatures, which were validated with separate cohorts. The
preimmunization signatures hint at mechanisms of differential vaccine responses between individuals; once
validated in additional studies, the postimmunization signatures could be used as a surrogate for protection in
clinical trials, possibly accelerating vaccine development.
ARTICLE TOOLS http://stm.sciencemag.org/content/12/543/eaay8924
SUPPLEMENTARY http://stm.sciencemag.org/content/suppl/2020/05/11/12.543.eaay8924.DC1
MATERIALS
RELATED http://stm.sciencemag.org/content/scitransmed/12/529/eaaw9522.full
CONTENT
http://stm.sciencemag.org/content/scitransmed/11/495/eaav3963.full
http://stm.sciencemag.org/content/scitransmed/11/474/eaau1458.full
REFERENCES This article cites 116 articles, 26 of which you can access for free
http://stm.sciencemag.org/content/12/543/eaay8924#BIBL
PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions
Use of this article is subject to the Terms of Service
Science Translational Medicine (ISSN 1946-6242) is published by the American Association for the Advancement of
Science, 1200 New York Avenue NW, Washington, DC 20005. The title Science Translational Medicine is a
registered trademark of AAAS.
Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement
of Science. No claim to original U.S. Government Works

Antigen-Stimulated PBMC Transcriptional Protective Signatures For Malaria Immunization

Uploaded by

Copyright:

Available Formats

Antigen-Stimulated PBMC Transcriptional Protective Signatures For Malaria Immunization

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Antigen-Stimulated PBMC Transcriptional Protective Signatures For Malaria Immunization

Uploaded by

Copyright:

Available Formats

SCIENCE TRANSLATIONAL MEDICINE | RESEARCH ARTICLE

MALARIA Copyright © 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

C Experimental design D Computational analysis (detection of biomarkers)

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

A Immunogenicity B CSP PfRBC Protection negatively associated with protection

were found positively enriched at postim-

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

C Enriched in monocytes (M11.0) D TLR and inflammatory signaling (M16)

Before immunization Unprotected after immunization Protected after immunization

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

A Immunogenicity B Protection (CSP stimulation)

Activated dendritic cells (M67) Activated dendritic cells (M67)

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Suppression of MAPK signaling (M56)

IRF7 PARP9 IMPA2 CLEC4A

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Comparators RTS,S unprotected RTS,S protected

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Preimmunization Adults PfRBC/uRBC PTK2, TGF-3, and 70.83% 70.83%

A B C CPS pre vaccination signature

True-positive rate (1-specificity)

Immunization doses with 8 Pf mosquitoes

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

True-positive rate (1-specificity)

True-positive rate (1-specificity)

TNFA FOXO3 SMAD3

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

Sci Transl Med 12, eaay8924.

Downloaded from http://stm.sciencemag.org/ at UPPSALA UNIVERSITY on May 13, 2020

ARTICLE TOOLS http://stm.sciencemag.org/content/12/543/eaay8924

Use of this article is subject to the Terms of Service

You might also like