ll
OPEN ACCESSiScience
ArticleIn vivo expansion of a CD9+ decidual-like NK cell
subset following autologous hematopoietic stem
cell transplantationAne Orrantia,
Enrique Va´zquez-
De Luis, Gabirel
Astarloa-Pando,
..., Ana Dopazo,
Olatz
Zenarruzabeitia,
Francisco Borrego
francisco.borregorabasco@
osakidetza.eus
Highlights
The transcriptome of NK
cells is significantly altered
immediately after
autoHSCT
The synthesis of
isoprenoids is
upregulated during NK
cell pool reconstitution
There is an expansion of
CD9+ decidual-like NK
cells following autoHSCT
Orrantia et al., iScience 25,
105235
October 21, 2022 ª 2022 The
Author(s).
https://doi.org/10.1016/
j.isci.2022.105235
ll
OPEN ACCESSiScienceArticleIn vivo expansion of a CD9+ decidual-like
NK cell subset following autologous
hematopoietic stem cell transplantation
Ane Orrantia,1,8 Enrique Va´zquez-De Luis,2,8 Gabirel Astarloa-Pando,1 In˜igo Terre´n,1 Ainhoa Amarilla-Irusta,1
Diego Polanco-Alonso,1 Carmen Gonza´lez,3 Alasne Uranga,3 Toma´s Carrascosa,4 Juan J. Mateos-Mazo´n,5
Juan C. Garcı´a-Ruiz,5 Sergio Callejas,2 Ana Quintas,2 Ana Dopazo,2,6 Olatz Zenarruzabeitia,1
and Francisco Borrego1,7,9,*1Immunopathology Group,
Biocruces Bizkaia Health
Research Institute, 48903
Barakaldo, Spain
2Genomics Unit, Spanish
National Center for
Cardiovascular Research
(CNIC), 28029 Madrid, Spain
3Biodonostia Health
Research Institute,
Hematology and
Hemotherapy Service,
Donostia University Hospital,
20014 Donostia-San
Sebastia´n, Spain
4Hematological Cancer
Group, Biocruces Bizkaia
Health Research Institute,
Hematology and
Hemotherapy Service,
Galdakao-Usansolo
University Hospital, 48960
Galdakao, Spain
5Hematological Cancer
Group, Biocruces Bizkaia
Health Research Institute,
Hematology and
Hemotherapy Service, Cruces
University Hospital, 48903
Barakaldo, Spain
6CIBER de Enfermedades
Cardiovasculares (CIBERCV),
Madrid, Spain
7Ikerbasque, Basque
Foundation for Science,
48013 Bilbao, Spain
8These authors contributed
equally
9Lead contact
*Correspondence:
francisco.borregorabasco@
osakidetza.eus
https://doi.org/10.1016/j.isci.
2022.105235SUMMARY
Autologous hematopoietic stem cell transplantation (autoHSCT) is a treatment
option for hematological disorders and pediatric solid tumors. After an au-
toHSCT, natural killer (NK) cells are the first lymphocyte subset returning to
normal levels. To uncover global changes during NK cell reconstitution after au-
toHSCT, we performed RNA-sequencing on NK cells before and after autoHSCT.
Results showed profound changes in the gene expression profile of NK cells
immediately after autoHSCT. Several biological processes including cell cycle,
DNA replication and the mevalonate pathway were enriched. Significantly, we
observed that following autoHSCT, NK cells acquired a decidual-like gene expres-
sion profile, including the expression of CD9. By using multiparametric flow cy-
tometry, we confirmed the expansion of NK cells expressing CD9 immediately af-
ter autoHSCT, which exhibited higher granzyme B and perforin expression levels
than CD9
 NK cells. These results provide insights into the physiopathology of
NK cells during their reconstitution after autoHSCT.
INTRODUCTION
Autologous hematopoietic stem cell transplantation (autoHSCT) is a therapy indicated to treat a diverse
group of hematological malignances and pediatric solid tumors (Bazinet and Popradi, 2019; Copelan,
2006; Hale, 2005). An early immune recovery after autoHSCT and specifically, day 15 absolute lymphocyte
count (ALC-15) after autoHSCT, have been described as an independent prognostic indicator for different
types of cancers (Porrata et al., 2001a, 2001b, 2002a, 2002b). Among lymphocytes, natural killer (NK) cells
are the first to recover during immune reconstitution after transplantation (Porrata et al., 2001c; Storek
et al., 2008) and are recognized as the main lymphocyte subset in ALC-15 that affects outcome after au-
toHSCT (Porrata et al., 2008).
NK cells are innate lymphocytes with the capacity to recognize and eliminate virus-infected and malignant
cells without prior sensitization (Cooper et al., 2001). Although traditionally divided in two main subsets
(CD56bright and CD56dim), NK cells are a diverse population of cells that exhibit different phenotypic and
functional features depending on tissue localization, maturation stage, environmental stimuli, viral infec-
tions, cancer, etc. (Freud et al., 2017). For example, tissue-resident NK cells express very high levels of
CD56 and markers of residency such as CD103, CD69, and CD49a (Crinier et al., 2018; Freud et al., 2017).
Decidual NK (dNK) cells are an interesting tissue-resident NK cell subset that exhibit a distinct phenotype
and unique transcriptional profile in comparison to peripheral blood NK (pNK) cells (Koopman et al., 2003).
dNK cells have been described as poorly cytotoxic, although they express functional activating receptors
and complete lytic machinery (Jabrane-Ferrat, 2019). Moreover, they are able to produce several cytokines,
such as interferon-gamma (IFNg), tumor necrosis factor (TNF), granulocyte-macrophage colony-stimulating
factor (GM-CSF), transforming growth factor b (TGF-b), and interleukin (IL)-10; chemokines, including, IL-8,
macrophage inflammatory protein 1 alpha (MIP1a), MIP1b, CCL5, CXCL10 andCXCL12; and angiogenic fac-
tors including angiopoietin-2, placental growth factor (PLGF) and vascular endothelial growth factor (VEGF),
at least after stimulation with IL-2 or IL-15 (Jabrane-Ferrat, 2019). dNK cells have been shown to regulate keyiScience 25, 105235, October 21, 2022 ª 2022 The Author(s).
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1
ll
OPEN ACCESS
iScience
Articledevelopmental processes during pregnancy, such as trophoblast invasion and vascular remodeling in the
decidua (Hanna et al., 2006). On the other hand, it is well known that both tumor-associated and tumor-infil-
trating NK cells (TANK and TINK, respectively) exhibit altered phenotype, metabolic profile, and effector
functions (Bruno et al., 2013; Terre´n et al., 2019, 2020; Terre´n and Borrego, 2022).
The phenotype diversity has also been observed when NK cell subsets reconstitution after autoHSCT was
analyzed in patients with multiple myeloma (MM) and malignant lymphomas (Jacobs et al., 2015; Orrantia
et al., 2021). Specifically, immediately after autoHSCT, NK cells exhibited an immature phenotype and
higher proliferation capacity (Orrantia et al., 2021) and showed differential expression of killer immuno-
globulin-like receptors (KIRs) (Jacobs et al., 2015) compared to NK cells pre-transplant. However, these
phenotypic changes were not maintained overtime (Jacobs et al., 2015; Orrantia et al., 2021). Importantly,
we have previously described an association between NK cells’ maturation stage after autoHSCT and the
clinical outcome of patients withMM (Orrantia et al., 2021). Thus, with the aim of acquiring a broader knowl-
edge of NK cell biology and physiopathology, we performed RNA-sequencing (RNA-seq) of NK cells dur-
ing their reconstitution after autoHSCT. Our results demonstrate that the gene expression profile of NK
cells exhibits very significant changes and that an intriguing decidual-like NK cell subset is expanded
following autoHSCT.RESULTS
The transcriptome landscape of natural killer cells is significantly changed during the
hematopoietic reconstitution after autologous hematopoietic stem cell transplantation
NK cells undergo profound phenotypical changes following autoHSCT (Jacobs et al., 2015; Orrantia et al.,
2021). To gain further insight into other aspects of NK cell biology, as for example proliferation and meta-
bolism, during subset reconstitution after autoHSCT, the gene expression profile of NK cells from patients
with MM (cohort 1) undergoing autoHSCT were compared at three different time points: T1 (pre-transplan-
tation), T2 (after leukocyte recovery or >1000 leukocytes/mL, usually around day 14 after autoHSCT) and T3
(30 days after autoHSCT). The clinical characteristics of patients from cohort 1 are listed in Table 1.
Differential gene expression analysis revealed changes in the gene expression profile of NK cells during
their reconstitution, mainly at T2. Specifically, 194 genes were upregulated and 220 were downregulated
in T2 compared to T1, and 197 genes were upregulated and 257 were downregulated in T2 compared
to T3 (Figure 1A). In contrast, only few genes were altered in T3 compared to T1 (Figure 1B). Moreover, un-
supervised hierarchical cluster analysis demonstrated a shift in gene expression pattern at T2 (Figure 1C). In
general, genes were clearly separated into two main groups with different gene expression profiles. The
first group (G1) consisted of genes with high expression in T2 compared to T1 and T3, and in contrast,
the fourth group (G4) consisted of genes with low expression in T2 that were highly expressed at T1 and
T3. However, as revealed by genes in groups G2 and G3 that showed lower expression in T3 than in T1
and higher expression in T3 than in T1, respectively, NK cell gene profile was similar but not identical at
T3 compared to T1 (Figure 1C). This suggests that some changes persist in NK cells gene profile onemonth
(T3) after autoHSCT.Cell division is a highly relevant biological process in natural killer cells early after autologous
hematopoietic stem cell transplantation
Over representation analysis (ORA) performed against the Reactome database (Gillespie et al., 2022) re-
vealed that the 10 most significantly enriched pathways for upregulated genes at T2, compared to both
T1 and T3, were pathways associated with cell cycle and DNA replication (Figure 1D). Similar results
were obtained with Gene Ontology database (Carbon et al., 2021) (Figure S1). However, cell cycle and
DNA replication pathways were not enriched when ORA was performed with upregulated genes in T3
compared to T1, indicating that while early after autoHSCT NK cells are vigorously dividing, one month af-
ter autoHSCT (T3) NK cells recover pre-transplant (T1) division rate, confirming our previous results (Orran-
tia et al., 2021).The synthesis of isoprenoids is significantly upregulated during natural killer cell pool
reconstitution
Although cell division and DNA replication were the most significantly enriched pathways at T2
compared to T1, gene set enrichment analysis (GSEA) against the Kyoto Encyclopedia of Genes and2 iScience 25, 105235, October 21, 2022
Table 1. Cohort 1: Multiple myeloma patients’ clinical characteristics
n (%)
Gender Male 10 (52.6%)
Female 9 (47.4%)
Myeloma classification: ISS ISS1 4 (21.1%)
ISS2 9 (47.4%)
ISS3 5 (26.3%)
Myeloma classification: Durie-Salmon Staging System I-A 2 (10.5%)
II-A 4 (21.1%)
III-A 9 (47.4%)
III-B 1 (5.3%)
Mobilization regimen G-CSF 18 (94.7%)
Plerixafor 1 (5.3%)
Conditioning regimen Melphalan 200 17 (89.5%)
Melphalan 140 + BUMEL 2 (10.5%)
CMV serostatus Yes 18 (94.7%)
No 1 (5.3%)
Maintenance or consolidation regimen Yes 14 (73.7%)
No 5 (26.3%)
Tandem autoHSCT Yes 9 (47.4%)
No 10 (52.6%)
Disease progression Yes 11 (57.9%)
No 8 (42.1%)
Median (interquartile range)
Age 59 (46–72)
Infused CD34+ cells (x106 cells/kg) 2.73 (2.01–5.59)
ISS indicates International Staging System; G-CSF: granulocyte colony-stimulating factor BUMEL, busulfan-melphalan; CMV,
cytomegalovirus.
ll
OPEN ACCESS
iScience
ArticleGenomes (KEGG) database (Kanehisa et al., 2021) also revealed an upregulation of metabolic pathways
(normalized enrichment score or NES: 2.76; adjusted p value: 3x10
10). Among differentially expressed
genes (DEGs) we found that GAPDH and ENO1, which encode two key enzymes of glycolysis, were up-
regulated at T2 compared to T1; SUCLA2 and FH, which encode enzymes that are part of the tricarbox-
ylic acid cycle and ACAA2, which encode an enzyme involved in fatty acid b-oxidation, were also upre-
gulated at T2 (Figure 2A). Specifically, ORA of upregulated genes against the KEGG modules database
showed an overrepresentation of the mevalonate pathway and pathways associated with nucleotide
metabolism (Figure 2B). The mevalonate pathway, which is part of the terpenoid backbone biosynthetic
pathway, is responsible for the synthesis of isopentenyl pyrophosphate, the essential building block of all
isoprenoids including cholesterol. As shown in Figure 2C, genes encoding most of the key enzymes of
the mevalonate pathway appeared upregulated at T2 compared to T1: ACAT1, ACAT2, HMGCS1,
HMGCR, PMVK, andMVD. These results suggest that the synthesis of isoprenoids is key for NK cells early
after autoHSCT.
Natural killer cells acquire a decidual-like gene expression profile early after autologous
hematopoietic stem cell transplantation
Once the functional enrichment analysis was completed, we focused on the analysis of individual DEGs to
further characterize alterations in gene expression profiles that occur during NK cell reconstitution
following autoHSCT. We observed that a number of genes related to NK cell development and maturation
(LEF1, ZEB2, FCRL6), cytotoxicity (LGALS1, LGALS3, GZMB, TNFSF10, FCGR3A), genes encoding acti-
vating receptors and activation markers (CD300C, CD38, CD160) and chemokine receptors (CCR1,
CCR5, CCR7) were differentially expressed at T2 compared to T1 (Figure 2A). Surprisingly, among theiScience 25, 105235, October 21, 2022 3
Figure 1. The gene expression profile of NK cells significantly changes immediately after autoHSCT
(A) Venn diagrams showing the number of differentially expressed genes (DEGs) in T2 compared to T1 and T3.
(B) Venn diagrams showing the number of DEGs in T3 compared to T1 and T2.
(C) Heatmap showing the relative expression of DEGs in the three-time points (T1, T2, and T3). Genes were hierarchically grouped according to their
homology. Patients’ samples (in columns) were grouped based on the time points (T1, T2, and T3). Four gene groups were identified according to the gene
expression pattern profile among time points (G1, G2, G3, and G4). G1 consists of genes with high expression in T2 compared to T1 and T3; G2 consists of
ll
OPEN ACCESS
4 iScience 25, 105235, October 21, 2022
iScience
Article
Figure 1. Continued
genes with lower expression in T3 than in T1; G3 consists of genes with higher expression in T3 than in T1; and G4 consists of genes with lower expression
in T2 compared to T1 and T3.
(D) Enrichment dot plot graphs showing the 10 most significant upregulated biological processes at T2 compared to T1 (upper panel) and T3 (lower panel).
Reactome database was used for this analysis. Samples from 10 patients with MM, from cohort 1, were used for RNA-seq analysis.
ll
OPEN ACCESS
iScience
Articletop 10 DEGs with lower p value and higher Log2FC value in T2 vs T1 comparison, we found CD9, which en-
codes a protein from the tetraspanin family. Although CD9 is ubiquitously expressed by all the major sub-
sets of leukocytes, it is considered a marker to be mostly expressed on decidual NK (dNK) cells (Koopman
et al., 2003). The upregulation of CD9 at T2 was confirmed by qPCR (Figure S2).
In contrast to peripheral blood NK (pNK) cell subsets (CD56bright and CD56dim NK cells), dNK cells have
been described to, among others, express CD9, CD151, tetraspan-5 (TSPAN5), Glycodelin A (PAEP) and
Core 2b-1,6-N-acetylglucosaminyl transferase (GCNT1) (Koopman et al., 2003). In addition, as other tis-
sue-resident NK cells, dNK cells also show expression of CD103 (ITGAE) and CD49a (ITGA1) (Jabrane-Fer-
rat, 2019). As shown in Figure 3A, NK cells showed a significantly higher expression ofCD9 andCD151 at T2
compared to both T1 and T3. Moreover, although not statistically significant, the mean expression of
TSPAN5, GCNT1, and PAEP tended to be slightly higher at T2. Furthermore, they showed higher expres-
sion of galectin-1 (LGALS1), which has been described to be overexpressed in dNK cells in comparison to
pNK cells (Koopman et al., 2003). Like CD56dim NK cells, dNK are known to express KIRs although the reper-
toire seems to be different compared with pNK cells (Koopman et al., 2003). The analysis of the detected
KIRs revealed an interesting increment in the expression of KIR2DL4 at T2 compared to T1 (Figure 3A). The
ligand for this KIR is the non-classical HLA-G molecule whose expression is restricted to the fetal tropho-
blast cells that invade the maternal decidua during early pregnancy (Rajagopalan and Long, 2012). Similar
results were obtained for ITGA1 (CD49a) and ITGAE (CD103) (Figure 3A). The chemokine receptors CXCR3
and CXCR4 seem to mediate the traffic of dNK toward maternal-fetal interface (Albini and Noonan, 2021).
Results also demonstrated a tendency to an upregulation of CXCR3 at T2 and a tendency to an upregula-
tion of CXCR4 at T3 (Figure 3A). Together, these data suggest that NK cells acquire a decidual-like (dNK-
like) gene expression profile early after autoHSCT (T2) that is not maintained overtime.A subset of CD9+ natural killer cells is expanded early after autologous hematopoietic stem
cell transplantation
Given the functional differences between pNK cells and dNK cells (Jabrane-Ferrat, 2019), we were interested
in studying the effector functions of CD9+ NK cells expanded after autoHSCT and comparing them with the
CD9
NK cell subset. For that, we first examined degranulation (CD107a), cytokine (IFNg, TNF, and IL-10), and
chemokine production (IL-8, MIP1b) after stimulation with PMA and ionomycin in patients with MM (cohort 1).
Of note, this unspecific stimulation is not a substitute for a cytotoxic assay, but it allowed to measure NK cell
effector functions regardless of the possible differences that CD9+ andCD9
NK cellsmay show in the expres-
sion of activating and inhibitory receptors. In addition, we have determined effector functions (CD107a, IFNg,
TNF, and MIP1b) after stimulating NK cells with the 721.221 target cell line in a heterogeneous cohort of pa-
tients (cohort 2) who also underwent autoHSCT (Table 2). Furthermore, we evaluated the content of lytic gran-
ules of non-stimulated NK cells by measuring granzyme B and perforin expression levels.
First, we observed that patients before the autoHSCT (T1) exhibited a higher frequency of CD9+ NK cells
than healthy donors (Figure 3B), which is consistent with previous data showing that CD9+ NK cells are
expanded in patients with cancer (Albini and Noonan, 2021; Bruno et al., 2018; Gallazzi et al., 2021; Kim
et al., 2016). Second, flow cytometry data are in agreement with the results obtained with RNA-seq and
clearly showed an expansion of CD9+ NK cells early after autoHSCT (T2) (Figure 3C). The frequency of
CD9+ NK cells at T2 was higher than the frequency before the autoHSCT (T1) and 30 days after (T3) the au-
toHSCT (Figure 3C). Interestingly, at T3 it was observed that the frequency of CD9+ NK cells was still higher
than before the autoHSCT (T1) (Figure 3C), which is somehow consistent with the qPCR data (Figure S2).
Next, we analyzed granzyme B and perforin expression and found that CD9+NK cells exhibited higher expres-
sion of these two important components of lytic granules compared to CD9
 NK cells, as determined by the
median fluorescence intensity (MFI) of these markers (Figure 3D). Although the differences were not statisti-
cally significant when we analyzed samples from cohort 2, probably owing to low number of samples, CD9+
NK cells have a clear tendency to express higher levels of perforin and granzyme B (Figure 3D).iScience 25, 105235, October 21, 2022 5
Figure 2. Metabolic pathways are upregulated at T2
(A) Heatmap showing the mean relative expression of genes related to metabolism (GAPDH, ENO1, SUCLA2, FH,
ACAA2), to NK cell development and maturation (LEF1, ZEB2, FCRL6), cytotoxicity (LGALS1, LGALS3, GZMB, TNFSF10,
FCGR3A) and genes encoding activating receptors and activation markers (CD300C, CD38, CD160) and chemokine
receptors (CCR1, CCR5, CCR7) at the three-time points (T1, T2 and T3). (B) Enrichment dot plot graphs showing
upregulated biological processes at T2 over T1. KEGG modules database was used for this analysis. Modules are ranked
based on the gene ratio. The size of the dot represents the number of genes upregulated within each category and the
color of the dot represents the significance of the enrichment.
(C) A portion of the KEGG pathway map of the terpenoid backbone biosynthesis (hsa00900), showing the mevalonate
pathway at T2 over T1. The color inside the boxes refers to the Log2FC. The number inside the boxes refers to KEGG
identification numbers, which represent the gene or gene family found at that position in the pathway. Samples from 10
patients with MM, from cohort 1, were used for RNA-seq analysis.
ll
OPEN ACCESS
iScience
ArticleRegarding effector functions in response to PMA and ionomycin, our results showed lower frequency
of CD107a+ cells, indicating a lower degranulation, in addition to a tendency to a lower frequency of
MIP1b+ cells within CD9+ NK cells compared to CD9
 NK cells (Figure S3A). No significant IL-8
expression was observed in NK cells (data not shown). Furthermore, no clear differences in cytokine
production (IFNg, TNF, IL-10) were observed between CD9+ and CD9
 NK cells (Figure S3A), except6 iScience 25, 105235, October 21, 2022
T1 T2 T3
Cohort 1 Cohort 2
A B
D
C
CD9+
CD9-
T1 HD
0
10
20
30
(%
)C
D9
+
NK
ce
lls
✱✱
T1 T2 T3
0
20
40
60
80
100
(%
)C
D9
+
NK
ce
ll s
✱✱✱✱
✱
T1 T2 T3
0
20000
40000
50000
100000
150000
M
FI
G
ra
nz
ym
e
B
in
NK
ce
lls
✱ ns ns
T1 T2 T3
0
1000
2000
3000
4000
5000
M
FI
Pe
rfo
rin
in
NK
ce
l ls
✱
ns ns
T1 T2 T3
0
1000
2000
3000
4000
5000
M
FI
Pe
rfo
rin
in
NK
ce
lls ✱
✱ ✱
T1 T2 T3
0
10000
20000
30000
40000
M
FI
G
ra
nz
ym
e
B
in
NK
ce
l ls
✱ ✱
✱
Figure 3. NK cells acquire a dNK-like phenotype immediately after autoHSCT
(A) Heatmap showing the mean relative expression of genes related to dNK cell phenotype at the three-time points (T1,
T2, and T3). Samples from 10 patients with MM, from cohort 1, were used for RNA-seq analysis.
(B) Bar graph showing the frequency of CD9+ NK cells from 12 patients (6 from cohort 1 and 6 from cohort 2) and 8 healthy
donors (HD). Each dot represents a donor. Significance of the data was determined by using Mann-Whitney test
comparing ranks for unpaired and non-parametric data. Data were represented as bar plots showing the median and
interquartile range. **p < 0.01.
(C) Bar graph showing the frequency of CD9+ NK cells at T1, T2, T3 from 12 patients (6 patients from cohort 1 and 6
patients from cohort 2) undergoing autoHSCT. Significance of the data was determined by usingWilcoxon matched pairs
signed rank test. Each dot represents a patient. Data were represented as bar plots showing the median and interquartile
range. *p < 0.05, **p < 0.01.
(D) Bar graphs showing theMFI of granzyme B (upper line) and perforin (lower line) within CD9+ or CD9
NK cells at T1, T2,
T3. Each dot represents a patient. Samples from 6 patients from cohort 1 (left) and 6 patients from cohort 2 (right) were
used in this analysis. Significance of the data was determined by using Wilcoxon matched pairs signed rank test. Bars
represent the mean. ns: not significant, *p < 0.05.
ll
OPEN ACCESS
iScience
Articlefor IFNg at T3 and TNF at T2. Although we do not have an explanation for these data, it may be
possible that, although statistically significant, the biological relevance may be minimal. On the other
hand, there were no significant differences between CD9+ and CD9
 NK cells regarding degranulation
(CD107a) and cytokine production (MIP1b, TNF, and IFNg) following the stimulation with the 721.221
target cell line (Figure S3B). Together, these results suggest that CD9+ NK cells that expanded early
after autoHSCT have higher levels of granzyme B and perforin and may exhibit some altered effector
functions (CD107a and MIP1b) that depend on the type of stimulus (PMA and ionomycin versus 721.221
target cell line).iScience 25, 105235, October 21, 2022 7
Table 2. Cohort 2: Patients’ clinical characteristics
n (%)
Gender Male 4 (66.7%)
Female 2 (33.3%)
Cancer type POEMS syndrome 2 (33.3%)
Amyloidosis 2 (33.3%)
Multiple myeloma 1 (16.7%)
AML 1 (16.7%)
Mobilization regimen G-CSF 4 (66.7%)
Plerixafor 1 (16.7%)
G-CSF + Plerixafor 1 (16.7%)
Median (interquartile range)
Age 58.50 (49.00–66.25)
Infused CD34+ cells (x106 cells/kg) 2.51 (2.19–2.82)
POEMS stands for polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes. G-CSF: granulocyte col-
ony-stimulating factor; AML: Acute myeloid leukemia.
ll
OPEN ACCESS
iScience
ArticleDISCUSSION
In this study, we have examined the global changes occurring in NK cells during their reconstitution after
human HSCT, by analyzing their transcriptome using RNA-seq technology. We have documented that
the gene expression profile of NK cells changes immediately after leukocyte recovery, being the cell cycle
themain biological process at this point. Moreover, we have also described the upregulation of somemeta-
bolic pathways, such as the mevalonate pathway. We have also found that NK cells acquire a dNK-like
phenotype shortly after autoHSCT and that these cells show some differences when comparedwith conven-
tional NK cells.
Flow cytometry-based studies have previously reported that NK cells undergo a phenotypic change after
autoHSCT (Jacobs et al., 2015; Orrantia et al., 2021). In line with these data, our RNA-seq results have
demonstrated that the gene expression profile of NK cells changes early after autoHSCT (T2). Moreover,
although the transcriptome landscapes of NK cells at pre-transplant (T1) and 30 days after (T3) autoHSCT
are similar, they show certain differences, which is also in accordance with previous studies (Orrantia et al.,
2021). Among the overexpressed genes on T3 vs T1 some examples includeCD9,CD55, KIR2DL3, KIR3DL1,
KIR2DL4, KIR2DS4, and GZMB. Among the downregulated genes on T3 vs T1 some examples include
CCR7, PECAM1, FCRL6 and TGFBR1.
An early immune recovery after both alloHSCT and autoHSCT is key for patients’ outcome (Pavletic et al.,
1998; Porrata et al., 2001a, 2001b, 2002a, 2002b). Early lymphocyte recovery, and more importantly, early
NK cell recovery were reported to predict superior survival after autoHSCT (Porrata et al., 2008; Rueff
et al., 2014). Moreover, NK cells have been described as the first lymphocyte population to be recovered
to normal levels after autoHSCT, beingNK cell count and functionality restored as early as day 14 post-trans-
plant (Porrata et al., 2001c; Storek et al., 2008). Therefore, it is not unexpected that the major changes in the
gene expression profile of NK cells occur before T3 (30 days after autoHSCT). Our data demonstrate that,
immediately after transplantation, NK cells are in the active phase of the cell cycle and that DNA replication
andnucleotidemetabolism are upregulated.However, the upregulation of thesebiological processes is not
maintained 30 days after autoHSCT. These findings are in agreement with the dramatic increase in the fre-
quency of Ki67-expressing cells immediately after autoHSCT (Orrantia et al., 2021), suggesting that one of
the biological processes that NK cells are undergoing is cell division to reach pre-transplant counts, process
that appear to be completed one-month post-autoHSCT (T3).
NK cells have been described to exhibit an immature phenotype at leukocyte recovery andNK cell maturation
occurs during their reconstitution after autoHSCT (Orrantia et al., 2021). Similar to what occurs with T cells
(Bantug et al., 2018), studies conducted in the last years indicate that NK cell differentiation status is closely
associated with their metabolic signature (Marc¸ais et al., 2014; Terre´n et al., 2020). Moreover, NK cell8 iScience 25, 105235, October 21, 2022
ll
OPEN ACCESS
iScience
Articlemetabolism can also bemodified by other processes, such as cell activation and somepathological conditions
(O’Brien and Finlay, 2019; Terre´n et al., 2019, 2020). In this regard, our data demonstrate an upregulation of
metabolic pathways, specifically of the mevalonate pathway, immediately after autoHSCT. Mevalonate is
the precursor of isoprenoids, such as cholesterol that is a key component of the eukaryotic cell membranes.
This pathway is known to be essential for lymphocyte proliferation and cell cycle progression (Chakrabarti
and Engleman, 1991; Crosbie et al., 2013; Cuthbert and Lipsky, 1990) which, in turn, is consistent with the
above-presented data. In addition, the mevalonate pathway is also important for NK cell cytotoxicity (Crosbie
et al., 2013; Poggi et al., 2013). Fluvastatin, an inhibitor of the 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-
CoA) encoded by the HMGCR gene, is known to inhibit actin redistribution, RhoA activation (needed for
acting assembly), and lipid raft formation (Poggi et al., 2013). In addition, fluvastatin inhibits the intracellular
free calcium increase, Akt1/PKB activation mediated by activating receptors engagement, and perforin/gran-
zyme-containing granules release (Poggi et al., 2013). All these effects were reverted with the addition of me-
valonate (Poggi et al., 2013). Thus, the synthesis of mevalonate is essential for the cytoskeleton assembly and
influence NK cell activation and cytotoxic granules release (Poggi et al., 2013).
Our data revealed, to our knowledge, a previously undescribed shift in NK cells phenotype toward a dNK-
like phenotype immediately after autoHSCT. pNK can acquire characteristics of dNK cells, such as the
expression of CD9, by their culture with not only conditioned medium from decidual stromal cells, but
also with TGF-b1 (Keskin et al., 2007). Conversion of pNK into a dNK-like phenotype is also achieved under
hypoxia and TGF-b1 culture conditions (Cerdeira et al., 2013). In addition, induction of the expression of
CD9 is also observed when NK cells are activated with IL-2 and PHA (Hanna et al., 2004). Very recently, it
has been reported that, in co-culture experiments, the NK-92 cell line acquires CD9 from ovarian tumor
cell lines through trogocytosis (Gonzalez et al., 2021). However, the acquisition of dNK-like phenotype after
autoHSCT may not be explained by these mechanisms. In fact, the TGF-b1 serum levels were reported to
be reduced during autologous bone marrow transplantation conditioning regimen and returned to normal
levels between days 20 and 50 post-transplant (Coomes and Moore, 2010; Liem et al., 1999). Nevertheless,
the combination of TGF-b1 with IL-15 and/or IL-18, cytokines also produced by decidual stroma cells, simi-
larly induced the acquisition of dNK cell markers (Du et al., 2022; Hawke et al., 2020; Siewiera et al., 2015).
Very significantly, IL-15 plasma levels are highly increased after autoHSCT (Orrantia et al., 2021; Porrata
et al., 2010) that, together with TGF-b1, may in part contribute to the acquisition of the dNK-like phenotype.
Thus, we hypothesize that the unique cytokine milieu that is present immediately after autoHSCT may be
responsible for this change in the NK cell phenotype.
Besides a distinct phenotype, dNK cells also show different functional characteristics compared to pNK
cells. dNK cells play crucial roles during pregnancy, such as regulating trophoblast invasion and promoting
vascular growth in the decidua (Hanna et al., 2006). These functions are mediated by the secretion of
different cytokines and chemokines. For example, dNK cells secrete IL-8 and interferon-inducible protein
10, which, upon binding to their receptors expressed on the trophoblast, regulate its invasion. Moreover,
dNK cells secrete angiogenic growth factors, such as VEGF and PLGF, which promote vascular remodeling
(Jabrane-Ferrat, 2019; Zhang et al., 2021). Although they display abundant cytolytic granules (Koopman
et al., 2003), dNK cells present low cytotoxic effect toward the K562 cell line (King et al., 1996) and failed
to polarize their microtubules organizing centers and cytotoxic granules to the immunological synapse
(Kopcow et al., 2005). However, very recently it was demonstrated that dNK cells protect placenta from in-
fections via selective transfer of granulysin (Crespo et al., 2020; Liu et al., 2021). In addition to dNK cells,
TINK and tumor-associated NK cells from patients with non-small-cell lung carcinoma (NSCLC) were
also found to secrete proangiogenic factors, such as VEGF, PLGF, and IL-8 (Bruno et al., 2013, 2014). In
this case, the proangiogenic activity of NK cells in tumors could have detrimental actions (Albini and
Noonan, 2021). Finally, the acquisition of CD9 from high-grade serous ovarian tumor cells via trogocytosis
was shown to confer NK cells immunosuppressive properties (Gonzalez et al., 2021).
Although CD9+ NK cells express higher levels of perforin and granzyme B than CD9
 NK cells, it is un-
clear to what extent the effector functions of these subsets are different, especially regarding degranu-
lation (CD107a) and MIP1b production. Several possibilities may explain the different results we have ob-
tained with different stimuli and cohorts (Figure S3): 1) PMA and ionomycin stimulation is a very strong
and non-specific stimulus, whereas the stimulation of 721.221 target cells is specific and, it may occur,
that other types of stimuli (i.e., other target cells, proinflammatory cytokines, etc.) give rise to different
results. 2) The two cohorts of patients with different types of cancer may affect NK cell activity differently.iScience 25, 105235, October 21, 2022 9
ll
OPEN ACCESS
iScience
Article3) Overnight priming with IL-15 (see STAR Methods section) could obliterate differences between CD9+
and CD9
cells in response to 721.221 target cells. Regardless, what is clear is that more research is
needed to fully define this highly expanded NK cell subset after autoHSCT.
The expansion of the CD9+ NK cell subset early after autoHSCTmay have clinical implications. Several pub-
lications have signaled that the increased frequency of CD9+ circulating immune cells associates with worse
disease-free survival in patients with cancer (Albini and Noonan, 2021; Bruno et al., 2018; Gallazzi et al.,
2021; Kim et al., 2016). Also, an important issue is the impact of the infused autograft absolute lymphocyte
count. In addition to stem cells, infusion of autograft immune effector cells produces an autologous graft-
versus-tumor effect (Porrata, 2022). We have not determined the number of CD9+ NK cells infused from the
autograft and, undoubtedly, this is a point that needs to be investigated in future studies to determine the
role of this NK cell subset in the evolution of patients undergoing autoHSCT.
In conclusion, to our knowledge, this is the first study to perform RNA-seq in NK cells from patients under-
going HSCT. We have demonstrated a change in the gene expression profile of NK cells shortly after au-
toHSCT, with an upregulation of biological processes such as cell cycle and metabolism. We have also re-
vealed that NK cells acquire a phenotype similar to that of dNK cells, characterized by the expression of
CD9. These CD9+ NK cells express higher levels of perforin and granzyme B and, depending on the stimuli,
they may exhibit different effector functions when compared with CD9
 NK cells. Future studies will deter-
mine the relevance of decidual-like NK cells in HSCT. Despite the limitation in the number of samples
analyzed, the results presented here provide insights into the biology and physiopathology of NK cells dur-
ing their reconstitution after autoHSCT.Limitations of the study
In this study, we have shown that the NK cell transcriptome changes significantly immediately after au-
toHSCT (around days 14-15). Thirty days after transplantation, NK cell gene expression is very similar to,
but not the same as, that of pre-transplant NK cells. We have analyzed the bulk NK cell transcriptome
using RNA-seq, which assesses changes owing to experimental conditions (before and after transplanta-
tion). However, these analyzes did not allow us to assess differences between NK cell subsets. For that,
single-cell RNA-seq analyses (scRNA-seq) must be performed. Also, extensive flow cytometry, mass cy-
tometry, or spectral cytometry studies can be helpful in phenotypically defining CD9+ decidual NK cells.
In addition, we have only preliminarily defined the functional profile of CD9+ NK cells, so it is essential to
fully characterize their effector functions, including testing their proangiogenic activity, in response to
different stimuli. Finally, if the expansion of CD9+ decidual-type NK cells, or other characteristics, are
associated with disease prognosis after autoHSCT, it is necessary to carry out this type of study in a
larger cohort of patients.STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:
d KEY RESOURCES TABLE
d RESOURCE AVAILABILITY10B Lead contact
B Materials availability
B Data and code availability
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Patients’ characteristics and study design
d METHOD DETAILS
B Sample preparation
B NK cell isolation
B RNA extraction and sequencing
B Reverse transcription (RT) reaction
B Real time qPCR
B Flow cytometry: Functional assay
d QUANTIFICATION AND STATISTICAL ANALYSIS
B RNA-seq data analysis
B Flow cytometry data analysisiScience 25, 105235, October 21, 2022
ll
OPEN ACCESS
iScience
ArticleSUPPLEMENTAL INFORMATION
Supplemental information can be found online at https://doi.org/10.1016/j.isci.2022.105235.
ACKNOWLEDGMENTS
We thank all patients who participated in this study and the staff from the Basque Biobank for Research. We
thank the staff of the Cell Analytics Facility from Achucarro Basque Center for Neuroscience for NK cell
sorting.
Funding: Supported by the following grants: AECC-Spanish Association Against Cancer (PROYE16074-
BORR) and Health Department, Basque Government (2021333006). GA-P is the recipient of a predoctoral
contract funded by AECC-Spanish Association Against Cancer (PRDVZ21440ASTA). DP-A is a recipient of a
fellowship from the AECC-Spanish Association Against Cancer (PPLAB212164POLA), AA-I and GA-P are
recipient of a fellowship from the Jesu´s de Gangoiti Barrera Foundation (FJGB20/007, FJGB21/001 and
FJBG21/005). IT is recipient of a predoctoral contract funded by the Department of Education, Basque
Government (PRE_2021_2_0215). OZ is the recipient of a postdoctoral contract funded by ‘‘Instituto de
Salud Carlos III-Contratos Sara Borrell 2017 (CD17/00128)’’ and the European Social Fund (ESF)-The ESF
invests in your future. FB is an Ikerbasque Research Professor, Ikerbasque, Basque Foundation for Science.
AUTHOR CONTRIBUTIONS
F.B. and O.Z. conceived the project; F.B., A.O., E.V-DL., S.C., A.Q., and A.D. designed experiments; C.G.,
A.U., T.C., J.J.M-M. and J.C.G-R. obtained the clinical samples and clinical data from patients; A.O., G.A-P
and D.P-A. performed experiments; A.O. and E.V-DL. performed data analysis; A.O. and E.V-DL. designed
figures; F.B., I.T., G.A-P., A.A-I., and O.Z. provided intellectual input; F.B. and A.O. wrote the article; all au-
thors critically reviewed the article.
DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: April 7, 2022
Revised: August 12, 2022
Accepted: September 25, 2022
Published: October 21, 2022REFERENCES
Albini, A., and Noonan, D.M. (2021). Decidual-like
NK cell polarization: from cancer killing to cancer
nurturing. Cancer Discov. 11, 28–33. https://doi.
org/10.1158/2159-8290.CD-20-0796.
Bantug, G.R., Galluzzi, L., Kroemer, G., and Hess,
C. (2018). The spectrum of T cell metabolism in
health and disease. Nat. Rev. Immunol. 18, 19–34.
https://doi.org/10.1038/nri.2017.99.
Bazinet, A., and Popradi, G. (2019). A general
practitioner’s guide to hematopoietic stem-cell
transplantation. Curr. Oncol. 26, 187–191. https://
doi.org/10.3747/co.26.5033.
Bruno, A., Bassani, B., D’Urso, D.G., Pitaku, I.,
Cassinotti, E., Pelosi, G., Boni, L., Dominioni, L.,
Noonan, D.M., Mortara, L., and Albini, A. (2018).
Angiogenin and the MMP9-TIMP2 axis are up-
regulated in proangiogenic, decidual NK-like
cells from patients with colorectal cancer. FASEB
J. 32, 5365–5377. https://doi.org/10.1096/fj.
201701103R.
Bruno, A., Ferlazzo, G., Albini, A., and Noonan,
D.M. (2014). A think tank of TINK/TANKs: tumor-
infiltrating/tumor-associated natural killer cells in
tumor progression and angiogenesis. J. Natl.Cancer Inst. 106. dju200–13. https://doi.org/10.
1093/jnci/dju200.
Bruno, A., Focaccetti, C., Pagani, A., Imperatori,
A.S., Spagnoletti, M., Rotolo, N., Cantelmo, A.R.,
Franzi, F., Capella, C., Ferlazzo, G., et al. (2013).
The proangiogenic phenotype of natural killer
cells in patients with non-small cell lung cancer.
Neoplasia 15, 133–142. https://doi.org/10.1593/
neo.121758.
Carbon, S., Douglass, E., Good, B.M., Unni, D.R.,
Harris, N.L., Mungall, C.J., Basu, S., Chisholm,
R.L., Dodson, R.J., Hartline, E., et al. (2021). The
GeneOntology resource: enriching a GOld mine.
Nucleic Acids Res. 49, D325–D334. https://doi.
org/10.1093/nar/gkaa1113.
Cerdeira, A.S., Rajakumar, A., Royle, C.M., Lo, A.,
Husain, Z., Thadhani, R.I., Sukhatme, V.P.,
Karumanchi, S.A., and Kopcow, H.D. (2013).
Conversion of peripheral blood NK cells to a
decidual NK-like phenotype by a cocktail of
defined factors. J. Immunol. 190, 3939–3948.
https://doi.org/10.4049/jimmunol.1202582.
Chakrabarti, R., and Engleman, E.G. (1991).
Interrelationships between mevalonate
metabolism and themitogenic signaling pathwayin T lymphocyte proliferation. J. Biol. Chem. 266,
12216–12222. https://doi.org/10.1016/S0021-
9258(18)98884-8.
Coomes, S.M., and Moore, B.B. (2010).
Pleiotropic effects of transforming growth factor-
b in hematopoietic stem-cell transplantation.
Transplantation 90, 1139–1144. https://doi.org/
10.1097/TP.0b013e3181efd018.
Cooper, M.A., Fehniger, T.A., and Caligiuri, M.A.
(2001). The biology of human natural killer-cell
subsets. Trends Immunol. 22, 633–640. https://
doi.org/10.1016/s1471-4906(01)02060-9.
Copelan, E.A. (2006). Hematopoietic stem-cell
transplantation. N. Engl. J. Med. 354, 1813–1826.
https://doi.org/10.1056/NEJMra052638.
Crespo, Aˆ.C., Mulik, S., Dotiwala, F., Ansara, J.A.,
Sen Santara, S., Ingersoll, K., Ovies, C., Junqueira,
C., Tilburgs, T., Strominger, J.L., and Lieberman,
J. (2020). Decidual NK cells transfer granulysin to
selectively kill bacteria in trophoblasts. Cell 182,
1125–1139.e18. https://doi.org/10.1016/j.cell.
2020.07.019.
Crinier, A., Milpied, P., Escalie`re, B., Piperoglou,
C., Galluso, J., Balsamo, A., Spinelli, L., Cervera-iScience 25, 105235, October 21, 2022 11
ll
OPEN ACCESS
iScience
ArticleMarzal, I., Ebbo, M., Girard-Madoux, M., et al.
(2018). High-dimensional single-cell analysis
identifies organ-specific signatures and
conserved NK cell subsets in humans and mice.
Immunity 49, 971–986.e5. https://doi.org/10.
1016/j.immuni.2018.09.009.
Crosbie, J., Magnussen, M., Dornbier, R.,
Iannone, A., and Steele, T.A. (2013). Statins inhibit
proliferation and cytotoxicity of a human
leukemic natural killer cell line. Biomark. Res. 1,
33. https://doi.org/10.1186/2050-7771-1-33.
Cuthbert, J.A., and Lipsky, P.E. (1990). Inhibition
by 6-fluoromevalonate demonstrates that
mevalonate or one of the mevalonate
phosphates is necessary for lymphocyte
proliferation. J. Biol. Chem. 265, 18568–18575.
https://doi.org/10.1016/S0021-9258(17)44789-2.
Du, X., Zhu, H., Jiao, D., Nian, Z., Zhang, J., Zhou,
Y., Zheng, X., Tong, X., Wei, H., and Fu, B. (2022).
Human-induced CD49a+ NK cells promote fetal
growth. Front. Immunol. 13, 821542–821613.
https://doi.org/10.3389/fimmu.2022.821542.
Freud, A.G., Mundy-Bosse, B.L., Yu, J., and
Caligiuri, M.A. (2017). The Broad spectrum of
human natural killer cell diversity. Immunity.
https://doi.org/10.1016/j.immuni.2017.10.008.
Gallazzi, M., Baci, D., Mortara, L., Bosi, A., Buono,
G., Naselli, A., Guarneri, A., Deho`, F.,
Capogrosso, P., Albini, A., et al. (2021). Prostate
cancer peripheral blood NK cells show enhanced
CD9, CD49a, CXCR4, CXCL8, MMP-9 production
and secrete monocyte-recruiting and polarizing
factors. Front. Immunol. 11, 586126–586217.
https://doi.org/10.3389/fimmu.2020.586126.
Gillespie, M., Jassal, B., Stephan, R., Milacic, M.,
Rothfels, K., Senff-Ribeiro, A., Griss, J., Sevilla, C.,
Matthews, L., Gong, C., et al. (2022). The
reactome pathway knowledgebase 2022. Nucleic
Acids Res. 50, D687–D692. https://doi.org/10.
1093/nar/gkab1028.
Gonzalez, V.D., Huang, Y.-W., Delgado-
Gonzalez, A., Chen, S.-Y., Donoso, K., Sachs, K.,
Gentles, A.J., Allard, G.M., Kolahi, K.S., Howitt,
B.E., et al. (2021). High-grade serous ovarian
tumor cells modulate NK cell function to create
an immune-tolerant microenvironment. Cell Rep.
36, 109632. https://doi.org/10.1016/j.celrep.
2021.109632.
Hale, G.A. (2005). Autologous hematopoietic
stem cell transplantation for pediatric solid
tumors. Expert Rev. Anticancer Ther. 5, 835–846.
https://doi.org/10.1586/14737140.5.5.835.
Hanna, J., Bechtel, P., Zhai, Y., Youssef, F.,
McLachlan, K., andMandelboim, O. (2004). Novel
insights on human NK cells’ immunological
modalities revealed by gene expression profiling.
J. Immunol. 173, 6547–6563. https://doi.org/10.
4049/jimmunol.173.11.6547.
Hanna, J., Goldman-Wohl, D., Hamani, Y.,
Avraham, I., Greenfield, C., Natanson-Yaron, S.,
Prus, D., Cohen-Daniel, L., Arnon, T.I., Manaster,
I., et al. (2006). Decidual NK cells regulate key
developmental processes at the human fetal-
maternal interface. Nat. Med. 12, 1065–1074.
https://doi.org/10.1038/nm1452.
Hawke, L.G., Mitchell, B.Z., and Ormiston, M.L.
(2020). TGF-B and IL-15 synergize through MAPK
pathways to drive the conversion of human NK12 iScience 25, 105235, October 21, 2022cells to an innate lymphoid cell 1–like phenotype.
J. Immunol. 204, 3171–3181. https://doi.org/10.
4049/jimmunol.1900866.
Jabrane-Ferrat, N. (2019). Features of human
decidual NK cells in healthy pregnancy and
during viral infection. Front. Immunol. 10, 1397–
1410. https://doi.org/10.3389/fimmu.2019.01397.
Jacobs, B., Tognarelli, S., Poller, K., Bader, P.,
Mackensen, A., and Ullrich, E. (2015). NK cell
subgroups, phenotype, and functions after
autologous stem cell transplantation. Front.
Immunol. 6, 583. https://doi.org/10.3389/fimmu.
2015.00583.
Kanehisa, M., Furumichi, M., Sato, Y., Ishiguro-
Watanabe, M., and Tanabe, M. (2021). KEGG:
integrating viruses and cellular organisms.
Nucleic Acids Res. 49, D545–D551. https://doi.
org/10.1093/nar/gkaa970.
Keskin, D.B., Allan, D.S.J., Rybalov, B., Andzelm,
M.M., Stern, J.N.H., Kopcow, H.D., Koopman,
L.A., and Strominger, J.L. (2007). TGFbeta
promotes conversion of CD16+ peripheral blood
NK cells into CD16- NK cells with similarities to
decidual NK cells. Proc. Natl. Acad. Sci. USA 104,
3378–3383. https://doi.org/10.1073/pnas.
0611098104.
Kim, K.-J., Kwon, H.J., Kim, M.C., and Bae, Y.K.
(2016). CD9 expression in colorectal carcinomas
and its prognostic significance. J. Pathol. Transl.
Med. 50, 459–468. https://doi.org/10.4132/jptm.
2016.10.02.
King, A., Jokhi, P.P., Burrows, T.D., Gardner, L.,
Sharkey, A.M., and Loke, Y.W. (1996). Functions of
human decidual NK cells. Am. J. Reprod.
Immunol. 35, 258–260. https://doi.org/10.1111/j.
1600-0897.1996.tb00041.x.
Koopman, L.A., Kopcow, H.D., Rybalov, B.,
Boyson, J.E., Orange, J.S., Schatz, F., Masch, R.,
Lockwood, C.J., Schachter, A.D., Park, P.J., and
Strominger, J.L. (2003). Human decidual natural
killer cells are a unique NK cell subset with
immunomodulatory Potential. J. Exp. Med. 198,
1201–1212. https://doi.org/10.1084/jem.
20030305.
Kopcow, H.D., Allan, D.S.J., Chen, X., Rybalov, B.,
Andzelm, M.M., Ge, B., and Strominger, J.L.
(2005). Human decidual NK cells form immature
activating synapses and are not cytotoxic. Proc.
Natl. Acad. Sci. USA 102, 15563–15568. https://
doi.org/10.1073/pnas.0507835102.
Li, B., and Dewey, C.N. (2011). RSEM: accurate
transcript quantification from RNA-Seq data with
or without a reference genome. BMC Bioinf. 12,
323. https://doi.org/10.1186/1471-2105-12-323.
Liem, L.M., Fibbe, W.E., van Houwelingen, H.C.,
and Goulmy, E. (1999). Serum transforming
growth factor-beta1 levels in bone marrow
transplant recipients correlate with blood cell
counts and chronic graft-versus-host disease.
Transplantation 67, 59–65. https://doi.org/10.
1097/00007890-199901150-00009.
Liu, Y., Gao, S., Zhao, Y., Wang, H., Pan, Q., and
Shao, Q. (2021). Decidual natural killer cells: a
good nanny at thematernal-fetal interface during
early pregnancy. Front. Immunol. 12, 663660–
663711. https://doi.org/10.3389/fimmu.2021.
663660.Luo, W., and Brouwer, C. (2013). Pathview: an
R/Bioconductor package for pathway-based data
integration and visualization. Bioinformatics 29,
1830–1831. https://doi.org/10.1093/
bioinformatics/btt285.
Marc¸ais, A., Cherfils-Vicini, J., Viant, C., Degouve,
S., Viel, S., Fenis, A., Rabilloud, J., Mayol, K.,
Tavares, A., Bienvenu, J., et al. (2014). The
metabolic checkpoint kinase mTOR is essential
for IL-15 signaling during the development and
activation of NK cells. Nat. Immunol. 15, 749–757.
https://doi.org/10.1038/ni.2936.
O’Brien, K.L., and Finlay, D.K. (2019).
Immunometabolism and natural killer cell
responses. Nat. Rev. Immunol. 19, 282–290.
https://doi.org/10.1038/s41577-019-0139-2.
Orrantia, A., Terre´n, I., Astarloa-Pando, G.,
Gonza´lez, C., Uranga, A., Mateos-Mazo´n, J.J.,
Garcı´a-Ruiz, J.C., Rin˜o´n, M., Rey, M., Pe´rez-
Fernandez, S., et al. (2021). NK cell reconstitution
after autologous hematopoietic stem cell
transplantation: association between NK cell
maturation stage and outcome in multiple
myeloma. Front. Immunol. 12, 748207–748213.
https://doi.org/10.3389/fimmu.2021.748207.
Orrantia, A., Terre´n, I., Izquierdo-Lafuente, A.,
Alonso-Cabrera, J.A., Sanda´, V., Vitalle´, J.,
Moreno, S., Tasias, M., Uranga, A., Gonza´lez, C.,
et al. (2020). A NKp80-based identification
strategy reveals that CD56neg NK cells are not
completely dysfunctional in health and disease.
iScience 23, 101298. https://doi.org/10.1016/j.
isci.2020.101298.
Pavletic, Z.S., Joshi, S.S., Pirruccello, S.J.,
Tarantolo, S.R., Kollath, J., Reed, E.C., Bierman,
P.J., Vose, J.M., Warkentin, P.I., Gross, T.G., et al.
(1998). Lymphocyte reconstitution after
allogeneic blood stem cell transplantation for
hematologic malignancies. Bone Marrow
Transplant. 21, 33–41. https://doi.org/10.1038/sj.
bmt.1701037.
Poggi, A., Boero, S., Musso, A., and Zocchi, M.R.
(2013). Selective role of mevalonate pathway in
regulating perforin but not FasL and TNFalpha
release in human natural killer cells. PLoS One 8,
e62932. https://doi.org/10.1371/journal.pone.
0062932.
Porrata, L.F., Ingle, J.N., Litzow, M.R., Geyer, S.,
and Markovic, S.N. (2001a). Prolonged survival
associated with early lymphocyte recovery after
autologous hematopoietic stem cell
transplantation for patients with metastatic
breast cancer. Bone Marrow Transplant. 28,
865–871. https://doi.org/10.1038/sj.bmt.
1703236.
Porrata, L.F., Litzow,M.R., Tefferi, A., Letendre, L.,
Kumar, S., Geyer, S.M., and Markovic, S.N.
(2002a). Early lymphocyte recovery is a predictive
factor for prolonged survival after autologous
hematopoietic stem cell transplantation for acute
myelogenous leukemia. Leukemia 16, 1311–1318.
https://doi.org/10.1038/sj.leu.2402503.
Porrata, L.F. (2022). The impact of infused
autograft absolute numbers of immune effector
cells on survival post-autologous stem cell
transplantation. Cells 11, 2197. https://doi.org/
10.3390/cells11142197.
Porrata, L.F., Gertz, M.A., Inwards, D.J., Litzow,
M.R., Lacy, M.Q., Tefferi, A., Gastineau, D.A.,
ll
OPEN ACCESS
iScience
ArticleDispenzieri, A., Ansell, S.M., Micallef, I.N., et al.
(2001b). Early lymphocyte recovery predicts
superior survival after autologous hematopoietic
stem cell transplantation in multiple myeloma or
non-Hodgkin lymphoma. Blood 98, 579–585.
https://doi.org/10.1182/blood.V98.3.579.
Porrata, L.F., Inwards, D.J., Ansell, S.M., Micallef,
I.N., Johnston, P.B., Gastineau, D.A., Litzow, M.R.,
Winters, J.L., and Markovic, S.N. (2008). Early
lymphocyte recovery predicts superior survival
after autologous stem cell transplantation in non-
hodgkin lymphoma: a Prospective study. Biol.
Blood Marrow Transplant. 14, 807–816. https://
doi.org/10.1016/j.bbmt.2008.04.013.
Porrata, L.F., Inwards, D.J., Micallef, I.N., Ansell,
S.M., Geyer, S.M., and Markovic, S.N. (2002b).
Early lymphocyte recovery post-autologous
haematopoietic stem cell transplantation is
associated with better survival in Hodgkin’s
disease. Br. J. Haematol. 117, 629–633. https://
doi.org/10.1046/j.1365-2141.2002.03478.x.
Porrata, L.F., Inwards, D.J., Micallef, I.N.,
Johnston, P.B., Ansell, S.M., Hogan, W.J., and
Markovic, S.N. (2010). Interleukin-15 affects
patient survival through natural killer cell recovery
after autologous hematopoietic stem cell
transplantation for non-hodgkin lymphomas.
Clin. Dev. Immunol. 914945–13. https://doi.org/
10.1155/2010/914945.
Porrata, L.F., Litzow, M.R., and Markovic, S.N.
(2001). Immune reconstitution after autologoushematopoietic stem cell transplantation. Mayo
Clin. Proc. 76, 407–412. https://doi.org/10.4065/
76.4.407.
Rajagopalan, S., and Long, E.O. (2012). KIR2DL4
(CD158d): an activation receptor for HLA-G.
Front. Immunol. 3, 258–266. https://doi.org/10.
3389/fimmu.2012.00258.
Rueff, J., Medinger, M., Heim, D., Passweg, J.,
and Stern, M. (2014). Lymphocyte subset recovery
and outcome after autologous hematopoietic
stem cell transplantation for plasma cell
myeloma. Biol. Blood Marrow Transplant. 20,
896–899. https://doi.org/10.1016/j.bbmt.2014.
03.007.
Siewiera, J., Gouilly, J., Hocine, H.-R., Cartron, G.,
Levy, C., Al-Daccak, R., and Jabrane-Ferrat, N.
(2015). Natural cytotoxicity receptor splice
variants orchestrate the distinct functions of
human natural killer cell subtypes. Nat. Commun.
6, 10183. https://doi.org/10.1038/ncomms10183.
Storek, J., Geddes, M., Khan, F., Huard, B., Helg,
C., Chalandon, Y., Passweg, J., and Roosnek, E.
(2008). Reconstitution of the immune system after
hematopoietic stem cell transplantation in
humans. Semin. Immunopathol. 30, 425–437.
https://doi.org/10.1007/s00281-008-0132-5.
Terre´n, I., and Borrego, F. (2022). Role of NK cells
in tumor progression. In Experientia
Supplementum, Magdalena Klink and IzabelaSzulc-Kielbic, eds. (Springer), pp. 169–187.
https://doi.org/10.1007/978-3-030-91311-3.
Terre´n, I., Orrantia, A., Vitalle´, J., Astarloa-Pando,
G., Zenarruzabeitia, O., and Borrego, F. (2020).
Modulating NK cell metabolism for cancer
immunotherapy. Semin. Hematol. 57, 213–224.
https://doi.org/10.1053/j.seminhematol.2020.10.
003.
Terre´n, I., Orrantia, A., Vitalle´, J., Zenarruzabeitia,
O., and Borrego, F. (2019). NK cell metabolism
and tumor microenvironment. Front. Immunol.
10, 2278. https://doi.org/10.3389/fimmu.2019.
02278.
Yu, G., and He, Q.-Y. (2016). ReactomePA: an
R/Bioconductor package for reactome pathway
analysis and visualization. Mol. Biosyst. 12,
477–479. https://doi.org/10.1039/C5MB00663E.
Yu, G., Wang, L.-G., Han, Y., and He, Q.-Y. (2012).
clusterProfiler: an R Package for comparing
biological themes among gene clusters. Omi. A J.
Integr. Biol. 16, 284–287. https://doi.org/10.1089/
omi.2011.0118.
Zhang, X., Li, Y., Huang, C., Liu, S., Chen, X., Yu,
S., Diao, L., and Zeng, Y. (2021). The role of
decidual natural killer cell-derived soluble factors
in early pregnancy. Am. J. Reprod. Immunol. 86,
e13477. https://doi.org/10.1111/aji.13477.iScience 25, 105235, October 21, 2022 13
ll
OPEN ACCESS
iScience
ArticleSTAR+METHODS
KEY RESOURCES TABLEREAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
PE Mouse monoclonal antibody Anti-CD3 (UCHT1) BD Biosciences Cat#555333
RRID: AB_395740
PE Mouse monoclonal antibody Anti-CD14 (M4P9) BD Biosciences Cat#562691
RRID: AB_2737725
PE Mouse monoclonal antibody Anti-CD19 (HIB19) BD Biosciences Cat# 555413
RRID: AB_395813
PE-Cy5 Mouse monoclonal antibody Anti-CD56 (B159) BD Biosciences Cat# 555517
RRID: AB_395907
PE-Vio770 Mouse monoclonal antibody Anti-NKp80 (4A4.D10) Miltenyi Biotec Cat#130-105-068
RRID: AB_2659831
BV510 Mouse monoclonal antibody Anti-CD14 (MVP9) BD Biosciences Cat#563079
RRID: AB_2737993
BV510 Mouse monoclonal antibody Anti-CD19 (SJ25C1) BD Biosciences Cat#562947
RRID: AB_2737914
BV510 Mouse monoclonal antibody Anti-CD123 (9F5) BD Biosciences Cat#563072
RRID: AB_2728102
BV786 Mouse monoclonal antibody Anti-CD56 (NCAM 16.2) BD Biosciences Cat#564058
RRID: AB_2738569
PerCP/Cyanine5.5 Mouse monoclonal antibody Anti-CD3 (SK7) BioLegend Cat#344808
RRID: AB_10640736
PE/Dazzle 594 Mouse monoclonal antibody Anti-CD9 (HI9a) BioLegend Cat#312118
RRID: AB_2728258
BV421 Mouse monoclonal antibody Anti-IL-8 (G265-8) BD Biosciences Cat#563310
RRID: AB_2738131
BV421 Mouse monoclonal antibody Anti-CD56 (NCAM 16.2) BD Biosciences Cat#562752
RRID: AB_2732054
BV605 Mouse monoclonal antibody Anti-IFN-g (B27) BD Biosciences Cat#562974
RRID: AB_2737926
BV650 Rat monoclonal antibody Anti-IL-10 (JES3-9D7) BD Biosciences Cat#564051
RRID: AB_2738565
FITC Mouse monoclonal antibody Anti-MIP-1b (D21-1351) BD Biosciences Cat#560565
RRID: AB_1645489
R718 Mouse monoclonal antibody Anti-Granzyme B (GB11) BD Biosciences Cat#566964
RRID: AB_2869975
BV711 Mouse monoclonal antibody Anti-Perforin (dG9) BioLegend Cat#308130
RRID:AB_2687190
APC Mouse monoclonal antibody Anti-TNF-a (MAb11) BioLegend Cat#502912
RRID:AB_315264
APC-Vio770 Recombinant antibody RE-Afinity
Anti-CD107a (REA792)
Miltenyi Biotec Cat#130-111-623
RRID:AB_2654486
Biological samples
PBMCs from multiple myeloma patients Basque Biobank for Research N/A
(Continued on next page)
14 iScience 25, 105235, October 21, 2022
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
Chemicals, peptides, and recombinant proteins
LIVE/DEAD Fixable Green Dead Cell Stain
Kit for 488nm excitation
Invitrogen Cat#L34969
LIVE/DEAD Fixable Aqua Dead Cell
Stain Kit for 405nm excitation
Invitrogen Cat#L34957
LIVE/DEAD Fixable Near IR Dead Cell
Stain Kit for 633 or 635nm excitation
Invitrogen Cat#10119
Brilliant Stain Buffer BD Biosciences N/A
GolgiStop Protein Transport Inhibitor (monensin) BD Biosciences Cat#554724
GolgiPlug Protein Transport Inhibitor (brefeldin A) BD Biosciences Cat#555029
Fixation/Permeablization Kit BD Biosciences Cat# 554714
Ficoll Paque Plus GE Healthcare Cat#GE17-1440-03
Fetal Bovine Serum (FBS) GE Healthcare Hyclone Cat#SV30160.03
Dimethyl sulfoxide (DMSO) Thermo Scientific Scientific Cat#20688
RPMI 1640 medium containing GlutaMAX Gibco Cat#72400054
Penicillin-Streptomycin Gibco Cat#15140122
MEM Non-Essential Amino Acids Solution (100X) Gibco Cat# 11140035
Sodium Pyruvate (100 mM) Gibco Cat# 11360070
PBS (PBS) Gibco Cat#18912
BSA Millipore Cat#A3733
Phorbol 12-myristate 13-acetate (PMA) Merck Cat# 79346
Ionomycin calcium salt from Streptomyces conglobatus Merck Cat# I0634
Critical commercial assays
RNeasy Plus Micro Kit Qiagen Cat# 74034
NEBNext Single Cell/Low Input RNA Library
Prep Kit for Illumina
New England Biolabs Cat# E6420L
Super-Script IV VILO Master Mix with ezDNase enzyme Invitrogen Cat#11766050
Brilliant III Ultra-Fast SYBR Green qPCR Master Mix Agilent Technologies Cat# 600882
Deposited data
GEO (Gene Expression Omnibus) GSE199608
Oligonucleotides
Primer: PPIA Forward:
CACCGTGTTCTTCGACATTG
Invitrogen N/A
Primer: PPIA Reverse:
CTGTGAAAGCAGGAACCCTTA
Invitrogen N/A
Primer: CD9 Forward:
GGGGATATTCCCACAAGGAT
Invitrogen N/A
Primer: CD9 Reverse:
GCAGTTCAACGCATAGTGGA
Invitrogen N/A
Software and algorithms
FlowJo Version 10.4.1 FlowJo N/A
GraphPad Prism v9 GraphPad N/A
AriaMX Real Time Software v.A1.2 Agilent Technologies N/A
Rsem v.1.3.1 N/A N/A
Morpheus Broad Institute N/A
clusterProfiler R package v.3.18.1 N/A N/A
(Continued on next page)
ll
OPEN ACCESS
iScience 25, 105235, October 21, 2022 15
iScience
Article
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
ReactomePA R package v.1.34.0 N/A N/A
Pathview R package v1.30.1 N/A N/A
ll
OPEN ACCESS
iScience
ArticleRESOURCE AVAILABILITY
Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by
the lead contact, Francisco Borrego (francisco.borregorabasco@osakidetza.eus).
Materials availability
This study did not generate new unique reagents.
Data and code availability
d RNA-seq data have been deposited at GEO (Gene Expression Omnibus) and are publicly available as of
the date of publication. Accession numbers are listed in the key resources table.
d All custom scripts used for DGE and Functional Analysis are available in github repository: https://
github.com/twisen/Orrantia-et-al-2022-RNASeq
d Any additional information required to reanalyze the data reported in this paper is available from the
lead contact upon request.EXPERIMENTAL MODEL AND SUBJECT DETAILS
Patients’ characteristics and study design
Blood samples from 19 patients who suffered from MM and received autoHSCT were used in the study
(Cohort 1. Age: median 59 and interquartile range 46-72; 10 males and 9 females). From them, samples
from 10 patients were used for RNA-seq experiments, 3 samples were used for qPCR experiments and 6
samples were used for flow cytometry experiments. Samples were obtained at three different time points:
T1 was collected the day before starting with the conditioning treatment; T2 was collected after leukocyte
recovery (>1000 leukocytes/mL) and T3 was collected 30 days after cell infusion. MMpatients’ (cohort 1) clin-
ical data can be found in Table 1. In addition, blood samples from a heterogeneous group of 6 patients that
received an autoHSCT (Cohort 2. Age: median 58 and interquartile range 49-66; 4 males and 2 females)
were used for flow cytometry experiments. Information about patients from cohort 2 is in Table 2. More-
over, buffy coats from anonymous healthy adult donors were used. Sample collection was carried out
through the Basque Biobank for Research (http://www.biobancovasco.org), which complies with the qual-
ity management, traceability and biosecurity set out in the Spanish Law 14/2007 of Biomedical Research
and in the Royal Decree 1716/2011. The study was approved by the Basque Ethics Committee for Clinical
Research (BIO14/TP/003 and PI + CES + INC-BIOEF, 2017-03). All subjects provided written and signed
informed consent in accordance with the Declaration of Helsinki.
METHOD DETAILS
Sample preparation
Peripheral blood mononuclear cells (PBMCs) were isolated from blood samples or buffy coats by Ficoll (GE
Healthcare) density gradient centrifugation and cryopreserved in heat-inactivated Fetal Bovine Serum
(FBS) (GE Healthcare Hyclone) containing 10% Dimethyl sulfoxide (DMSO) (Thermo Scientific Scientific).
NK cell isolation
Cryopreserved PBMCs were thawed at 37⁰C in a water bath, washed with PBS (Lonza), filtered using 70mm
cell strainers and counted. Next, PBMCs were stained with LIVE/DEAD Fixable Green Cell Stain Kit (Invitro-
gen) to exclude dead cells, following manufacturer’s recommendations. Then, cells were washed with PBS
containing 2.5% BSA (Sigma-Aldrich). Extracellular staining was carried out by incubating PBMCs for
30 min on ice in the dark with the following fluorochrome-conjugated monoclonal antibodies: PE anti-
CD3 (UCHT1), PE anti-CD14 (M4P9), PE anti-CD19 (HIB19) and PE-Cy5 anti-CD56 (B159) from BD Biosci-
ence and PEVio770 anti-NKp80 (4A4.D10) from Miltenyi Biotec. NK cells were identified as viable16 iScience 25, 105235, October 21, 2022
ll
OPEN ACCESS
iScience
ArticleCD3-/CD14-/CD19-/NKp80+/CD56+/
 cells (Orrantia et al., 2020, 2021) and sorted using a BD FACSJazz
cell sorter (Figure S4).RNA extraction and sequencing
Total RNA was isolated from the sorted NK cells using RNeasy Plus Micro Kit (Qiagen) following manufac-
turer’s recommendations. Quality and integrity of each RNA sample was checked using both a Bioanalyzer
and a Nanodrop before proceeding to RNAseq protocol. 0.5 ng of total RNA were used to generate bar-
coded RNA-seq libraries using the NEBNext Single Cell/Low Input RNA Library Prep Kit for Illumina (New
England Biolabs). Briefly, poly A + RNA was purified using poly-T oligo-attached magnetic beads followed
by fragmentation and then first and second cDNA strand synthesis. Next, cDNA 30 ends were adenylated
and the adapters were ligated and performed an uracile excision from the adaptor followed by PCR ampli-
fication. The size of the libraries was checked using the Agilent 2100 Bioanalyzer High Sensitivity DNA chip
and their concentration was determined using the Qubit fluorometer (ThermoFisher Scientific). Libraries
were sequenced on a HiSeq 2500/HiSeq4000 (Illumina) to generate 49/60 bases single reads. FastQ files
for each sample were obtained using bcl2fastq 2.20 Software software (Illumina).Reverse transcription (RT) reaction
cDNA was synthetized from 3 ng of total RNA using Super-Script IV VILOMaster Mix with ezDNase enzyme
(ThermoFisher Scientific) following manufacturer’s recommendations. A DNA removal step is performed
using ezDNase before the reverse transcription.Real time qPCR
Real time qPCR was performed using Brilliant III Ultra-Fast SYBR Green qPCRMaster Mix (Agilent Technol-
ogies) following manufacturer’s recommendation on an Agilent AriaMX instrument (Agilent Technologies).
In each reaction, 2 mL of (1:10) diluted cDNA of each sample was loaded in triplicates into Micro-Amp Fast
96-well reaction plates (Applied Biosystems). In addition, 18 mL of qPCR master mix was added, which
included Brilliant III Ultra-Fast SYBR Green qPCR Master Mix (10 mL), housekeeping gene/probe gen
primers (600nM of each primer were used), and nuclease free DEPC-treated water (Fisher Scientific). Nega-
tive controls (no cDNA) were included to verify the absence of contamination. Quantitative analysis was
made based on the cycle threshold (Ct) value and calculated using AriaMX Real Time software v.A1.2
(Agilent Technologies). Ct mean values from triplicate reactions were calculated for further analysis. The
sequences of the primers used are shown in Table S1. PPIA was used as housekeeping gene. Primers
were designed with Primer3Plus software using FASTA sequences of each gene and synthetized by Invitro-
gen (Thermo Fisher).Flow cytometry: Functional assay
For functional assay, PBMCs were thawed, counted, plated at 1 x 106 cells/well in 48-well plates in NK cell
culturemedium (RPMI 1640mediumwith GlutaMAX, 10% heat-inactivated FBS, 1% penicillin-streptomycin,
1% non-essential amino acids and 1% Sodium-Pyruvate) and cultured overnight. Next, cells were stimu-
lated with 50 ng/mL PMA and 2 mM Ionomycin for 6h. APC-Vio770 anti-CD107a (REA792, Miltenyi Biotec)
was added at the start of the stimulation time in addition to GolgiStop (monensin) and GolgiPlug (brefeldin
A) protein transport inhibitors (BD Biosciences) following manufacturer’s recommendations. Next, PBMCs
were washed with PBS and stained with LIVE/DEAD Fixable Aqua Dead Cell Stain Kit (Invitrogen) reagent to
exclude dead cells, following manufacturer’s recommendations. Afterward, cells were washed with PBS
containing 2.5% BSA (Sigma-Aldrich) and extracellular staining was performed. For that, cells were incu-
bated for 30 min on ice protected from light with the following fluorochrome-conjugated mAbs: BV510
anti-CD14 (M4P9), BV510 anti-CD19 (SJ25C1), BV510 anti-CD123 (9F5) and BV786 anti-CD56 (NCAM
16.2) from BD Biosciences; PE/Dazzle 594 anti-CD9 (HI9a) and PerCP-Cy5.5 anti-CD3 (SK7) from
BioLegend and PE-Vio770 anti-NKp80 (4A4.D10) from Miltenyi Biotec. Cells were then washed again
with PBS containing 2.5% BSA, and fixed and permeabilized using Cytofix/Cytoperm Fixation/Permeabili-
zation solution from BD Biosciences following manufacturer’s recommendations. Next, cells were incu-
bated for 30 min on ice protected from light with the following fluorochrome conjugated mAbs: BV421
anti-IL-8 (G265-8), BV605 anti-IFNg (B27), BV650 anti-IL-10 (JES3-9D7), FITC anti-MIP1b (D21-1351), R718
anti-granzyme B (GB11) from BD Bioscience; BV711 anti-perforin (dG9) and APC anti-TNF (Mab11) from
BioLegend. Finally, cells were washed with 1x Perm/Wash Buffer (BD Biosciences) and resuspended iniScience 25, 105235, October 21, 2022 17
ll
OPEN ACCESS
iScience
ArticlePBS. Samples were then acquired in a LSRFortessa X-20 flow cytometer (BD Biosciences). Flow cytometry
panel used for the functional assay is shown in Table S2.
For the functional assay with the 721.221 target cell line we followed a protocol previously published by us
(Orrantia et al., 2021). Briefly, after thawing and counting the PBMCs, cells in NK cell culture medium were
plated at 0.5 x 106 cells/well in 96 well round (U) bottom plates. PBMCs were then primed with 10 ng/mL of
recombinant human (rh) interleukin (IL)-15 (Miltenyi Biotec) and cultured overnight. Then, 721.221 target
cells were added at Effector:Target (E:T) 1:1 ratio (0.5x106 PBMCs and 0.5x106 721.221 target cells) and
cultured for 8 additional hours. APC-Vio770 anti-CD107a was added at the start of the co-culture period
and GolgiStop and GolgiPlug were added 1 h later. Next, extracellular and intracellular staining was per-
formed as above described for PMA and ionomycin stimulation. However, in this set of experiments a
different panel of antibodies were used (see Tables S2 and S3).QUANTIFICATION AND STATISTICAL ANALYSIS
RNA-seq data analysis
Sequencing reads were aligned to the human reference transcriptome (GRCh38 v91) and quantified with
RSem v1.3.1 (Li and Dewey, 2011). Raw counts were normalized with TPM (Transcripts per Million) and
TMM (Trimmed Mean of M-values) methods, transformed into log2 expression (log2(rawCount+1)) and
compared to calculate fold-change and corrected p Value. Only those genes expressed with at least 1
count in 8 samples equal were taken into account. As we had an expected intra-group variability, we
considered for functional analysis genes with |Log2FC| > 0.58 and unadjusted p Value <0.05, although
we also calculate the Benjamini and Hochberg correction for all genes. The source code for the analysis
is available in the github repository: https://github.com/twisen/Orrantia-et-al-2022-RNASeq.
Heatmaps were generated withMorpheus online tool fromBroad Institute (https://software.broadinstitute.
org/morpheus). For functional analysis, we used clusterProfiler (version 3.18.1), ReactomePA (version 1.34.0)
and pathview R (version 1.30.1) packages (Luo and Brouwer, 2013; Yu et al., 2012; Yu and He, 2016).Flow cytometry data analysis
Flow cytometry data were analyzed using FlowJo v10.8.1. software. GraphPad Prism v.9.3.1 was used for
graphical representation and statistical analysis. Non-parametric Wilcoxon matched-pairs signed rank
test were used to determine significant differences. Data were represented as bar plots showing the me-
dian and interquartile range.18 iScience 25, 105235, October 21, 2022