Coordinate actions of innate immune responses …...SYSTEMS IMMUNOLOGY Coordinate actions of innate immune responses oppose those of the adaptive immune system during Salmonella infection

R E S E A R C H A R T I C L E

S Y S T E M S I M M U N O L O G Y

Coordinate actions of innate immune responsesoppose those of the adaptive immune systemduring Salmonella infection of miceAndrew N. Hotson,1,2* Smita Gopinath,2* Monica Nicolau,1 Anna Khasanova,1

Rachel Finck,1,2 Denise Monack,2 Garry P. Nolan1,2†

http://stke.sD

ownloaded from

The immune systemenacts a coordinated responsewhen facedwith complex environmental and pathogenicperturbations. We used the heterogeneous responses of mice to persistent Salmonella infection to modelsystem-wide coordination of the immune response to bacterial burden.We hypothesized that the variabilityin outcomes of bacterial growth and immune response across genetically identical mice could be used toidentify immune elements that serve as integrators enabling co-regulation and interconnectedness of theinnate and adaptive immune systems. Correlation analysis of immune response variation to Salmonella in-fection linked bacterial loadwith at least four discrete, interacting functional immune response “cassettes.”One of these, the innate cassette, in the chronically infected mice included features of the innate immunesystem, systemic neutrophilia, and high serum concentrations of the proinflammatory cytokine interleukin-6. Comparedwithmicewith amoderate bacterial load,micewith the highest bacterial burden exhibited highactivityof this innatecassette,whichwasassociatedwithadampenedactivityof theadaptiveTcell cassette—with fewer plasma cells and CD4+ T helper 1 cells and increased numbers of regulatory T cells—and with adampened activity of the cytokine signaling cassette. System-wide manipulation of neutrophil numbers re-vealed that neutrophils regulated signal transducer and activator of transcription (STAT) signaling in B cellsduring infection. Thus, a network-level approach demonstrated unappreciated interconnections that balancedinnateandadaptive immuneresponsesduring thedynamiccourseofdiseaseand identifiedsignalsassociatedwith pathogen transmission status, as well as a regulatory role for neutrophils in cytokine signaling.

cie
on A
ugust 21, 2020ncem

ag.org/

INTRODUCTION

Flow cytometric analysis with antibodies specific for phosphorylated pro-teins (phospho-flow) enables phospho-proteomic measurements of signaltransduction at the single-cell level (1, 2). By combining the detection ofthe activated, phosphorylated forms of signaling proteins with phenotypiccell surface markers that distinguish immune cell lineages, cell subset fre-quencies and signaling states are revealed. In addition, how disease perturbsthe functional responsiveness of cellsmaybe evaluated by studying immunecell signaling in response to stimulation with cytokines (3, 4). Cell type–specific signaling states can then be correlated across a population to iden-tify signaling nodes that act in concert (5). Here, immune cell populationfrequencies and their signaling capacities, as well as antibody and cytokineproduction and bacterial loads across multiple organs in mice infected withSalmonella, were analyzed using a systems biology approach to elucidateactivation of the immune network in response to chronic bacterial infection.

Infection with a given pathogen may elicit strikingly different outcomesin different individuals. For example, Salmonella Typhi causes potentiallylethal typhoid fever in most infected people, but a small subset of individ-uals (most famously, TyphoidMary) become asymptomatic carriers capableof bacterial transmission (6). Likewise, less than 20 colony-forming units(CFUs) ofSalmonella enterica serovarTyphimurium (S.Typhimurium) causelethal disease in C57BL/6 and BALB/c mouse strains (7), whereasNramp1+/+

129sv mice survive infection but are persistent carriers of the bacterium (8).Nramp1 (natural resistance–associated macrophage protein 1) is localized to

1The Baxter Laboratory for Stem Cell Biology, Stanford University School of Med-icine, Stanford, CA 94305, USA. 2Department of Microbiology and Immunology,Stanford University School of Medicine, Stanford, CA 94305, USA.*These authors contributed equally to this work.†Corresponding author. E-mail: [email protected]

ww

the lysosomesofmacrophages anddendritic cells and is necessary for inductionof proinflammatory cytokine and chemokine production in S. Typhimurium–infected 129sv mice and for resistance to Salmonella-induced cell death (9).However, the genetic variation of hosts does not fully dictate disease out-come, because even within the inbred 129sv strain, the amount of bacteriashed in the feces varies by 10 orders of magnitude (10). The mice with thehighest fecal loads of S. Typhimurium also exhibit gastrointestinal inflam-mation and are termed “super-shedders” because they can transmit infectionto uninfected mice (10). The factors that dictate disparate disease progres-sion remain unclear, although the natural microbiota certainly plays a rolebecause pretreatment with antibiotics causes all infected mice to becomesuper-shedders (10).

The host immune response to S. Typhimurium involves a complex in-terplay between innate and adaptive immunity, as well as negative feedbackto temper the immune response. Persistent infection enables us to analyzethe interaction between the innate and adaptive arms of the immune re-sponse because they are modulated by the pathogen over time. The innateimmune response to pathogenic assault is comprised of microbial recogni-tion and clearance as well as cytokine release, which shapes the adaptiveimmune response. Neutrophils are critical for containing bacteria at diseaseonset in susceptible mice, as exemplified by neutrophil-depleted mice hav-ing higher systemic loads of bacteria (11). In addition, S. Typhimurium in-duces a T helper 1 (TH1)–biased adaptive response, and neutralization of thekey TH1 cytokine interferon-g (IFN-g) causes increased bacterial burden inmultiple organs of chronically infected mice (8). At the same time, regula-tory T cells (Tregs) can abrogate the proliferation of CD4

+ TH1 cells (12). Inaddition, antibody production byB cells is required for a protective responseagainst S. Typhimurium (13).

Immune system studies have almost exclusively focused on single-cellanalyses by flow cytometry and correlating a few cell types at a time with a

w.SCIENCESIGNALING.org 12 January 2016 Vol 9 Issue 410 ra4 1

http://stke.sciencemag.org/


http://stke.sciencD

ownloaded from

givenmechanistic or clinical outcome.The advent of phospho-flowandoth-er intracellular staining techniques added signaling to the repertoire ofcellular events that can be measured; however, it is clear that the immunesystem works as a concerted network of dynamically diverse playerscontrolled over time to lead to a given outcome. High-throughput data col-lection of immune system attributes, such as gene expression profiles, couldpotentially be used to construct correlation networks of cellular attributes toattempt to discern systems-level functionalities. In this manner, we wouldnot look at the immune system as a set of cells with attributes, but rather as aset of cellular attributes organized into correlated functions. Here, we ap-plied this principle to S. Typhimurium infection as a model to determinethe potential of the general approach.

In a previous study, we compared the host immune response betweensuper-shedder and non–super-shedder mice and found that super-sheddershad greater neutrophilia (increased numbers of neutrophils) and adampened TH1 response (14). Here, both the dynamic progression of thehost immune response during chronic Salmonella infection and the differ-ences in connectivity between an infected and uninfected host immunenetwork were examined. Markers of a TH1 response, antibody response,cytokine signaling, and innate immunity were measured and correlatedacross mice with established infection and, separately, across uninfectedmice to evaluate how infection perturbs the immune system network. Infec-tion induced innate immunity attributes that were greatest inmice harboringhigh fecal bacterial loads, whereas mice with better controlled gastro-intestinal infection exhibited an enhanced adaptive immune response andcytokine signaling responses. Interventions to manipulate neutrophils wereused to validate an infection-specific neutrophil regulation of B cellsignaling. This approach reveals new intercellular relationships as well aspreviously unappreciated systems-level interactions that comprise anetwork of functional immune response characteristics.

on August 21, 2020

emag.org/

RESULTS

Persistent S. Typhimurium infection induces sustainedactivation of both innate and adaptive immune responsesMice were infected with S. Typhimurium, and over a 28-day time course,splenocyteswere isolated for staining with a panel of antibodies against phe-notypic cell surface markers for analysis by traditional fluorescence-basedflow cytometry. The cell populations were delineated and organized usingthe spanning-tree progression analysis of density-normalized events (SPADE)algorithm (15). Specific cell types were organized on the basis of the inten-sity of surface marker staining to identify cell lineages (Fig. 1, A and B).

SPADE enables visualization of the entire hematopoietic cell map, in-cluding cells with intermediate amounts of surface markers, which reflectthe transition states of cell progression (16). To determine how cell subsetfrequenciesweremodulated by infection,we normalized the number of cellsin each cell type cluster to those in the uninfected state (Fig. 1C). At 4 dayspostinfection (d.p.i.), there was an increase in the numbers of cells in thegranulocyte and macrophage populations relative to those in uninfectedmice, which became fully established by 8 d.p.i. (Fig. 1C). Activation ofcells of the adaptive immune system (the adaptive compartment) alsooccurred during the time course, with the expansion (increase in number)of Ly-6C+ plasma cells (17, 18), which was apparent at 6 d.p.i. (Fig. 1C andfig. S1A). CD4+ T cells positive for CD44, CD49d, Ly-6C, and Ki-67 wereincreased in number at 8 d.p.i. (Fig. 1C). Long-term memory CD4+ T cellsexpress Ly-6C on the cell surface (19), CD44 is a marker for effector/memory cells, CD49d is increased in abundance as activatedT cells undergocellular divisions (20, 21), and Ki-67 is expressed during the antigen-specificinduction of T cell proliferation (22). In chronically infected mice, TH cells

ww

underwent continued expansion; the CD44+ Ly-6Cmid-hi population in-creased continuously, from 8% of the total CD4+ T cells at 8 d.p.i. to 21%at 29 d.p.i. (Fig. 1C and fig. S1B).

Infection-induced CD4+ T cells were TH1-biased as indicated by de-tection of the transcription factor T-bet and of the phosphorylation (andactivation) of signal transducer and activator of transcription 4 (STAT4)in response to stimulation with interleukin-12 (IL-12) (fig. S1, C and D).In addition, infection induced a population of T-bet+Foxp3+CD4+ T cellsthat were highly responsive to IL-12. This phenotype is consistent withthat of TH1-like Tregs (23). CD4

+ T cells with antigen specificity towardS. Typhimurium continuously expanded through 30 days of infection(fig. S1D). To identify phenotypic markers associated with antigen-specificTH1 cells, we cultured splenocytes from infected mice with bone marrow–derived macrophages that were primed to present Salmonella antigen. In-tracellular cytokine staining revealed that within the CD44+ effector/memoryTH cell population, IFN-g production increasedwith Ly-6C abun-dance (fig. S1E). Thus, Ly-6C could be used in conjunction with CD44 todefine TH cell subsets enriched for the potential to secrete IFN-g. Together,these data suggest that Salmonella infection induces a rapid, sustained in-nate immune response in the host. The proliferation of activated T cells wasdetected aweek after infection, and their numbers increased through the firstmonth. Thus, the homeostatic state during chronic Salmonella infectionwascharacterized by increased numbers of innate myeloid cells, plasma cells,and activated TH cells. This opens two key questions:What are the relation-ships between the functions of these changes to cell subsets, and to whatextent is there a dynamic set of relationships that precisely define suchchanges?

Persistent S. Typhimurium infection remodels thesignaling potentials of multiple splenocyte subsetsDuring infection, the ability of cells to respond to cytokines may be directlymodulated by the pathogen or as a secondary effect of the host immune re-sponse and cytokine production. To evaluate the cell signaling potential dur-ing chronic Salmonella infection, we exposed splenocytes isolated frommiceat 30 d.p.i. to a set of cytokines that activate the maximum number of cellsubsets with distinct roles in immune processes: IFN-g (TH1-biased), IL-6(acute-phase response), IL-10 (anti-inflammatory), and IL-21 (antibody re-sponse in B cells). The ability of cells to respond to these cytokines wasdetermined by measurement of the relative abundances of phosphorylatedSTAT1 and STAT3 proteins (pSTAT1 and pSTAT3, respectively).

We observed that S. Typhimurium infection resulted in changes to cyto-kine responses in nearly everymajor cell type. For example, IL-21–inducedSTAT1 and STAT3 phosphorylation was greater in lymphocytes frominfected mice than in lymphocytes from uninfected controls [P < 0.05 forpopulations of B cells, CD8+ T cells, and naïve CD4+ T cells; Kolmogorov-Smirnov (K-S) test] (Fig. 2A). Cells from infected mice also exhibitedenhanced responsiveness to IL-10. IL-10 is primarily classified as an anti-inflammatory cytokine, and it acts through STAT3. B cells and T cells frominfected mice demonstrated enhanced IL-10–mediated STAT3 phosphoryl-ation compared to the same populations from uninfectedmice (P < 0.5, K-Stest) (Fig. 2A). IL-10 also evoked stronger activation of pSTAT1 in macro-phages from infectedmice than in macrophages from naïve mice (P < 0.01,K-S test). As such, overall IL-10 and IL-21 responses were augmented dur-ing infection.

Note that the cells from uninfected mice responded quite uniformly; foreach cell type, the resting and stimulated intensities of pSTAT1 and pSTAT3were consistent across mice [coefficient of variation (CV) >10% for 4 of50 conditions]. In contrast, the responses of splenocytes from infectedmice exhibited marked variability (CV >10% for 49 of 50 conditions),both in their amounts of pSTATunder unstimulated conditions and in




their responses to cytokines (Fig. 2A). Consistent with these data, the var-iance in STAT phosphorylation was statistically significantly greater acrossinfected mice than uninfected mice for 32 of 50 conditions (Levene test).For example, the variation in the median fluorescence intensity (MFI) ofpSTAT3 in IL-6–stimulated B cellswas significantly (P< 0.01, Levene test)lower in uninfectedmice (MFI range of 79 to 98) than in infectedmice (MFIrange of 71 to 168). Similarly, in IFN-g–stimulated samples, the MFI of

on August 21, 2020


Dow

nloaded from

--

--

-r

lr

.

r-

l

.-

r-

-

-

-.--

.

-

ww

pSTAT1 of effector/memory CD4+ T cells ranged from 61 to 69 in un-infected mice, but expanded to between 42 and 246 in infected mice. Fur-thermore, infected mice were uniformly hyperresponsive either in both ofthese signaling nodes or in neither (Fig. 2B), demonstrating correlated re-sponses (P < 0.01, Spearman’s correlation test) to cytokines with special-ized functional outcomes: IL-6 promotes B cell growth and antibodyproduction, whereas IFN-g reinforces TH1 biasing in TH cells. Therefore,

Log10

B cells

Granulocytes

Monocytes

Macrophage

CD4 T cells

CD4+

CD44, CD49d

CD62L

CD62L

CD62LLy-6C,

Ly-6C

Ly-6C+

CD44+, Ly-6C+, CD49d

CD44+

Ly-6C-mid

Ki-67+

Ki-67+

CD8 T cells

pDC

Ki-67-mid

Ly-6C+

CD49d+

Plasma

CD44,Ki67

Infected cell percentageUninfected cell percentage

Ery

thro

cyte

s

Log10

−1 0 1

B cells

Granulocytes

Monocytes

Macrophage

CD4 T cells

CD4+

CD44, CD49d

CD62L

CD62L

CD62LLy-6C,

Ly-6C

Ly-6C+


CD44+

Ly-6C-mid

Ki-67+

Ki-67+

CD8 T cells

pDC

Ki-67-mid

Ly-6C+

CD49d+

Plasma

CD44,Ki67


Ery

thro

cyte

s

Log10

−1 0 1

B cells

Granulocytes

Monocytes

Macrophage

CD4 T cells

CD4+

CD44, CD49d

CD62L

CD62L

CD62LLy-6C,

Ly-6C

Ly-6C+


CD44+

Ly-6C-mid

Ki-67+

Ki-67+

CD8 T cells

pDC

Ki67-mid

Ly-6C+

CD49d+

Plasma

CD44,Ki67


Ery

thro

cyte

s

Ly-6C expression

Min Max

CD8 expression

Min Max

A

B

C

Day 6

Day 13

Day 19

Day 4

Day 8

Day 29

Ery

thro

cyte

sE

ryth

rocy

tes

Log10

0 1−1

B cells

Granulocytes

Monocytes

Macrophage

CD4 T cells

CD4+

CD44, CD49d

CD62L

CD62L

CD62LLy-6C,

Ly-6C

Ly-6C+


CD44+

Ly-6C-mid

Ki-67+

Ki-67+

CD8 T cells

pDC

Ki-67-mid

Ly-6C+

CD49d+

Plasma

CD44,Ki-67


Ery

thro

cyte

s

Log10

B cells

Granulocytes

Monocytes

Macrophage

CD4 T cells

CD4+

CD44, CD49d

CD62L

CD62L

CD62LLy-6C,

Ly-6C

Ly-6C+


CD44+

Ly-6C-mid

Ki-67+

Ki-67+

CD8 T cells

pDC

Ki-67-mid

Ly-6C+

CD49d+

Plasma

CD44,Ki-67


Log10

B cells

Granulocytes

Monocytes

Macrophage

CD4 T cells

CD4+

CD44, CD49d

CD62L

CD62L

CD62LLy-6C,

Ly-6C

Ly-6C+


CD44+

Ly-6C-mid

Ki-67+

Ki-67+

CD8 T cells

pDC

Ki-67-mid

Ly-6C+

CD49d+

Plasma

CD44,Ki-67


CD8 T cells

Ly-6

C

CD8 T cells

0 1−1

0 1−1

−1 0 1

Fig. 1. Rapid accumulation of innate cell types in the spleen precedes the adapt ive immuneresponse. (A to C) Splenocytesfrom mice infected with 5 × 108

Salmonella bacteria were prepared for flow cytometry and analyzed by SPADE. Each bubblerepresents a group of cells defined by a specific lineage markeexpression profile and is called acell circle. The coloring of each celcircle represents the surface markestaining intensity (A and B) or thechange in cell frequency (C) fromfive mice sacrificed at each timepoint over the 29-day time courseEach plot was generated from anequal number of cell eventsmergedin silico from five mice. The numbeof cells in each cell circle is represented by the size. (A) In the SPADEplot, cell circles are colored by CD8staining intensity, demonstratingthe grouping of cell circles into celtypes by lineage-specific markers(represented by a magenta box)(B) Cell circles are colored by Ly6C staining intensity, as an exampleof a gradient of phenotypic markeintensity across a cell type (represented with a black arrow). (C) Thisfigureusescolor toshow thechangein cell frequency across the infection time course. Cell circles arecolored by the change in the frequency of cells occurring in eachcircle at an infection time point compared to those in uninfected miceThe formula used to color cell circles was as follows: log10(percentage of cells in the cell circle duringinfection/percentage of cells inthe cell circle in uninfected mice)An increased frequency is indicatedby orange, no change by olive, anda decrease by purple. pDC, plasmacytoid dendritic cell.




even at the observation level, it was apparent that signaling events associatedwith distinct immune consequences were correlated within infected mice.This provoked us to ask whether deeper or more extensive correlationsacross the immune system existed that could enable a more thorough, orsystems-level, interpretation of immune actions in infected mice.

Correlation maps demonstrate extensiveSalmonella-driven remodeling of host immune responseSalmonella infection, whether acute or chronic, would be expected to initi-ate co-regulated changes to cytokine responsiveness across various cell

ww

types. We hypothesized that the diversity of bacterial loads and infectionoutcomes across genetically identical mice would reveal variable immuneresponses, with specific immune attributes being more or less activated inmore highly infected mice. To objectively illustrate coordinated immuneevents during persistent Salmonella infection, we determined correlationvalues between measurements of 125 attributes of the immune system in-cluding STAT signaling, cell frequencies across organs, markers of T cellandB cell activation, serum cytokine concentrations, and bacterial load. Theintensities of each of these attributes were correlated across 20 uninfectedmice and, separately, across 19 mice infected with Salmonella for 30 to

w.SCIENCESIGNALING.org 1

on August 21, 2020


Dow

nloaded from

35 days. For each node-node pair (wherein anode is a cell frequency, bacterial load, serumcytokine value, or a cellular attribute such asSTAT1 phosphorylation in a given cell typein response to a perturbing cytokine), it wasdetermined whether the correlation wasstatistically significantly positive, signifi-cantly negative, or insignificant using datafrom two processing batches (see Materialsand Methods). Of 7750 possible correla-tions, 2223 positive and 443 negative signif-icant correlations were observed in infectedmice (fig. S2A), whereas there were only835 positive and 318 negative attribute cor-relations in uninfected mice (fig. S2B) asrepresenting the “basal” state. This findingsuggests that infection induces coordinatedactivity of the immune system above the ba-sal state.

Many correlations between immuneattributes were similar between uninfectedand infected mice. Of the correlations ob-served in uninfected mice, 44% (512 of1153) were maintained during infection,whereas only 5.8% (68 of 1153) of the cor-relations in uninfected mice were of the op-posite direction in infected animals (Fig. 3A).The correlations that reversed direction dur-ing infection were associated with fournodes: splenic Treg frequency, splenic neutro-phil frequency, basal STAT3 phosphoryl-ation in B cells, and IL-21–induced STAT1phosphorylation in granulocytes. These find-ings suggest that the regulatory units thatcontrol the coordinated actions of multipleimmune functions are primarily conservedand amplified during infection, although asmall subset of the immune response, suchas Tregs, becomes differentially regulated.

A large number of the negative correla-tions observed in the infected state involvedeither fecal bacterial load or splenic neutro-philia (Fig. 3B), demonstrating that thesefactors generally opposed other measuredevents of the immune system. For example,both fecal CFU and neutrophil frequencieswere negatively correlated with splenic TH1frequencies and with cytokine-inducedSTAT1 and STAT3 phosphorylation in mul-tiple splenic cell subsets. In addition, fecal

B c

ell

B c

ell

Unstim IFN-γ IL-6 IL-10 IL-21

Uninfected Salmonella{ { Uninfected Salmonella{

CD

8C

D8

CD

4 m

em/e

ffC

D4

mem

/eff

CD

4 n

aC

D4

naï

veve





A

B

B c

ells

IL-6

→ p

STA

T3 M

FI

CD4 mem/eff IFN-γ → pSTAT1 MFI

4 mice 12 mice 4 mice 12 mice

Mac

rop

hag

eM

acro

ph

age

pSTAT1 (MFI)0 271 869

pSTAT1 (MFI)

pSTAT1 (MFI)

pSTAT1 (MFI)

pSTAT1 (MFI)

0 73 704

0 45 827

0 126 443

0 41 883

pSTAT3 (MFI)0 444 2028

pSTAT3 (MFI)

pSTAT3 (MFI)

pSTAT3 (MFI)

pSTAT3 (MFI)

0 114 828

0 64 547

0 112 566

0 92 507

Salmonella

Uninfected

R = 0.96P < 0.01

2

Fig. 2. Salmonella infection inducesvariability in the signaling responsesof splenocytes to cytokines. (A) Sple-nocytes from 12 mice infected withSalmonella for 30 days and from4un-infected mice were left unstimulated(unstim) or were stimulated for 15 minwith IFN-g, IL-6, IL-21 (all at 40 ng/ml),or IL-10 (80 ng/ml) before being pre-pared for flow cytometric analysis. Theheat map shows the MFIs of pSTAT1and pSTAT3 for the listed cell types.Mem/eff, memory/effector. (B) Corre-lation between the IL-6–induced

phosphorylation of STAT3 in B cells and the IFN-g–induced phosphorylation of STAT1 in effector/memoryCD4+ T cells.

2 January 2016 Vol 9 Issue 410 ra4 4



on August 21, 2020


Dow

nloaded from

MacsSTAT1

MacsSTAT3 B cells

STAT1

CD4 mem/effSTAT1

B cellsSTAT3

CD4 mem/effSTAT3

Gr1STAT1

CD4 naïveSTAT1

CD8STAT1 CD8

STAT3 CD4 naïveSTAT3

CD4 mem/effpSTAT1

CD4 mem/effpSTAT3

CD4 naïvepSTAT1

B cellspSTAT1

Gr1pSTAT1

CD8pSTAT3

CD8pSTAT1 CD4 naïve

pSTAT3

B cellspSTAT3

MacspSTAT3

MacspSTAT1

CD8

CD11b+

Naïve% of CD4

T-bet+

% of CD4

Mem/eff% of CD4

CD4

cDC

B cells

Gr1pSTAT3

Gr1STAT3

Macs

TH1

IL-6

Ki-67+ IL-10

Naïve

IgG Tregs

IFN-γ

mLN CFU

pSTAT5+ pSTAT4−

unstim

T-bet+

% of CD4B cells

FoxP3+

% of CD4

pSTAT5+ pSTAT4−

IL-2+IL-12

pSTAT5− pSTAT4−

IL-2+IL-12pSTAT5− pSTAT4−

unstim

Gr1

Spleen CFU

CD4

CD8

B cells

cDC

pDC

Gr1

Macs

Intestinal IgG

FoxP3+

% of CD4

CD4

CD11b+

Gr1

B cells

Feces CFU

pSTAT5+ pSTAT4+

IL-2+IL-12

B cellspSTAT1

CD4 naïvepSTAT3

CD8pSTAT1

CD4 naïvepSTAT1

CD8pSTAT3

Gr1pSTAT1

CD4 mem/effpSTAT3

B cellspSTAT3

CD4 naïve pSTAT1

CD4 naïve pSTAT3

MacspSTAT3

MacspSTAT1

Gr1pSTAT3

CD8pSTAT3

CD4 mem/effpSTAT1

Gr1pSTAT1

B cellspSTAT1

CD8pSTAT1

MacspSTAT1

Gr1pSTAT1

CD44+

Ly-6C-int

CD44−

Ly-6C-int

CD44+

Ly-6C-hi

CD8pSTAT1

MacspSTAT1

Gr1pSTAT3

MacspSTAT3 B cells

pSTAT1

B cellspSTAT3

CD4 mem/effpSTAT1

CD4 mem/effpSTAT3

CD44−

Ly-6C−

CD4 naïvepSTAT3

CD4 naïve pSTAT1

CD4 mem/effpSTAT3

CD4 mem/effpSTAT1

CD4 mem/effpSTAT1

B cellspSTAT3

CD44+

Ly-6C−

CD4 naïvepSTAT1

Gr1pSTAT3

Gr1pSTAT1

MacspSTAT1

CD8pSTAT3

CD8pSTAT1

Ly-6C+

B cellspSTAT1

MacspSTAT3

Ki-67+

pSTAT5+ pSTAT4+

unstim

pSTAT5− pSTAT4+

IL-2+IL-12pSTAT5− pSTAT4+

unstim

Unstim

TotalSTAT

IL-10

IL-21

IL-6

IFN γ

Total IκBα

mLN

Intestine

Serum

% ofSpleen

CD4 subsets

CD4 Ly-6C subsets

B cell subsets

Tregs

Gr1pStat1

Gr1

Correlation reversals betweeninfected and unifected: Turns negative in infection Turns positive in infection

B cellspSTAT3

A

B

Selected infection correlations: Negative in infection Positive in infection

Fig. 3. Transmission status and splenic neutrophil frequency share a similarcorrelation pattern during Salmonella infection. (A) The map of correlationsthat reversed directions between uninfected and infected mice (20 un-infected mice and 19 mice infected for 30 to 35 days). Green circles arenodes with four or more reversals. Blue lines represent negative correla-tions during infection that were positive in uninfected mice, and red linesrepresent positive correlations during infection that were negative in un-infected mice. (B) The correlation map of infection shows all correlations

ww

with fecal bacterial load (feces CFU) and splenic neutrophil frequencies(Gr1) (filled magenta circles) obtained from data from 19 mice infectedfor 30 to 35 days. Red lines signify positive correlations, and blue linessignify negative correlations. Outlined blue or red circles share correla-tions with either fecal CFU or splenic neutrophil frequency, solid bluecircles share negative correlations with both, and gray circles are notcorrelated with either. int, intermediate; hi, high; cDC, conventional den-dritic cell; Macs, macrophages.




http://stke.sciencemag

Dow

nloaded from

Salmonella CFU was negatively correlated with CD4 Ki-67, a measureof T cell proliferation, and Ly-6C, indicative of IFN-g production,whereas neutrophil frequencies were associated with reduced plasmacell frequency, dampened IL-2 and IL-12 activity in CD4+ T cells frommesenteric lymph nodes (mLNs), and increased Treg numbers. Together,these data reveal that increased bacterial fecal shedding is associatedwith neutrophilia, reduced Janus-activated kinase (JAK)–STAT signaling,and a suppressed adaptive immune response. Thus, correlation analysisrevealed that the immune network that forms during persistent Salmonel-la infection was balanced between an enhanced neutrophil response andan enhanced responsiveness of T cells.

Known immune interactions are identified throughcorrelation analysisTo establish whether the correlated attributes of the host immune responseto Salmonella infection were reflective of biologically meaningful relation-ships, we compared a subset of the correlation network to known infection-induced immuneconnections. In theNramp1+/+model ofSalmonella infection,Tregs suppress the proliferation of antigen-specific T cells (12). In addition,increased fecal bacterial load induces increased numbers of neutrophils;both of these factors also cause a reduction in the numbers of TH1-biasedcells and dampen STAT1 responsiveness to IL-6 in naïve CD4+ T cells,whereas neutrophils inhibit the IL-2–mediated proliferation of activatedTH cells (14). These known causative relationships between attributes ofthe host-microbe interaction were constructed into a network for compari-son with the correlation network derived herein (Fig. 4). Our correlationnetwork identified the established positive relationship between fecalCFU and neutrophils, as well as five of six negative relationships (Fig. 4).CD44+Ly-6ChighCD4+ T cells were used as a measure of T cell prolifer-ation because this subset contains antigenically expanded TH cells duringinfection (Fig. 1 and fig. S1). Additionally, our analysis identified threecorrelations within this subnetwork that, to our knowledge, have not been

ww

previously reported (designated as unreported correlations): a negativecorrelation between bacterial shedding and T cell proliferation, and posi-tive correlations between TH1 frequency and both T cell proliferation andIL-6 responsiveness (Fig. 4). Functional intervention and measurement ofthe node pairs in these three correlations have not been reported, to ourknowledge; thus, it is unknown if there is a direct causal relationship be-tween these pairs, or if an alternative regulatory element coordinates theiractivity.

Clustering correlations reveal functionalimmune cassettesEvaluation of paired correlations is useful for determining howa given nodeconnects within a network as a whole (Fig. 3B). However, it may bechallenging to deduce the architecture of the entire network from a seriesof pairwise correlations.We hypothesized that master regulator functions ofthe immune system (cell types or cytokines, for example) would control theactivity of multiple downstream functions or elements; thus, the networkwould be composed of modules of co-regulated elements. Furthermore,the interactions across immune system pathways would organize bothmaster regulators and activated elements into biologically functional cas-settes that could each act in coordination or opposition to each other.

To assess whether key factors of Salmonella infection such as neutro-phils, TH1 response, antibody production, cytokine signaling, and fecalCFU shared common correlations, we constructed a correlation matrix.All 125 measured attributes of the immune system were organized by hier-archical clustering. Clustering revealed four groups of attributes that demon-strated similar correlation patterns; the attributes within each group sharedfunctional commonalities that could be considered as a module of bio-logical activity (Fig. 5A). For example, the Innate Immunity Cassetteconsisted of high fecal bacterial load and IL-6 serum concentrations,high frequencies of splenic antigen-presenting cells and Tregs, and reducedeffector/memory CD4+ T cell representation, which are all reported char-


on August 21, 2020

.org/

acteristics of super-shedders (14). The othercassettes identified were the STAT Signal-ing Cassette (consisting mostly of pSTAT1and pSTAT3 responses and certainB cell re-sponse characteristics), the Total STAT Pro-tein Cassette, and the Adaptive ImmunityCassette (consisting of TH cell and B cell re-sponses). There were also relationships ob-served between the cassettes of biologicalfunction. For example, the relationship be-tween the STAT Signaling Cassette andthe Innate Immunity Cassette was enrichedfor negative correlations between their indi-vidual elements (P < 0.01, t test) (Fig. 5A,gray box), indicating that the two cassettesact in opposition to each other. The InnateImmunity Cassette was also negatively cor-related with the Adaptive Immunity Cassette(P < 0.01, t test) but was not significantlycorrelated with the Total STAT Protein Cas-sette (Fig. 5B). STAT signalingwas positivelycorrelatedwith the adaptive immune responseandwith theTotal STATProteinCassette (Fig.5, B and C). Therefore, mice harboring thegreatest fecal Salmonella loads and transmis-sion capacity demonstrated an enhanced in-nate immune response that actively opposedadaptive immunity and STAT signaling.

Fig. 4. Discovered connectivity shares relationships with the known regulatory network. This subset ofnetwork relationships from the Salmonella-infected murine system illustrates the overlap between known re-lationships and those reported in this work. Lines with arrowheads show previously reported causal relation-ships as determined through experimental intervention. The solid red and blue arrows represent knownpositive and negative relationships, respectively, that were present in our correlation data; the dotted bluearrow is a known relationship that was not found. Thick lines without arrowheads indicate previously unchar-acterized (unreported) correlations reported here.




The presence of these cassettes in uninfected immune system networkswas then investigated (Fig. 6A). A c2 test was used to determine whethercassettes defined by the analysis of data from infectedmicewere statisticallysignificantly enriched for correlations in data from uninfected mice com-pared to the overall correlation rate. Although there were fewer correlationsin data from uninfected mice, the cassettes identified by the analysis of datafrom infectedmicewere largely statistically recapitulated (albeit at aweaker

ww

level). The STAT Signaling, Total STAT Protein, and Adaptive ImmunityCassetteswere significantly enriched for positive correlations, suggesting thatthe elements of these cassettes functioned in concert in uninfected mice. Fur-thermore, the intersections of these cassettes were also enriched for positivecorrelations. In the absence of infection, there was no significant correlationwithin the Innate Immunity Cassette, reflecting the absence of a regulatedinnate response in uninfected mice (Fig. 6A). A positive correlation between

on August 21, 2020


Dow

nloaded from

STAT signaling/ B cell cassette

Total STAT cassette

Adaptive immunity cassette

Innate immunity cassette

A

B C

Infected correlation matrix clustered using Ward algorithm

A STAT signaling/B cell cassetteMyeloid STAT1Lymphoid STAT1Lymphoid STAT3B cell proliferationSerum IgG

B Total STAT cassetteTotal STAT1/3 proteinMyeloid IL-21 responseCD4 IL-2 responseSpleen CFU

C Adaptive immunity cassetteT cell IL-6 and IL-10 responseCD4 activation, proliferation and T

H1

Plasma cells and Intetsinal IgGSerum IL-10NFκB inhibition

UncorrelatedMyeloid STAT3, serum IFN-γmLN CFUOther cell frequencies

D Innate imunity cassetteFecal CFU, serum IL-6Gr1, cDC, macrophage and B cell levelsT

reg and unactivated CD4 T cells

0 1−1

Innate cassette

Adaptive cassette

Total STAT cassette

STAT signalingcassette

Correlation values averaged across cassettes

Correlations averaged Infected immune network

A B C D

Fig. 5. Clustering of theSalmonella infection correlationmatrix reveals function-al cassettes comprised of attributeswith shared correlations. (A) The correlationnetwork from 19 infected mice shown in fig. S2B is given in matrix form. Thematrix was clustered and attributes that clustered together (black boxes) wereidentified and annotated on the basis of function. Red represents a positivecorrelation,white representsnocorrelation,andblue representsanegativecor-

relation. The gray box highlights the intersection of two clustered cassettes.(B) Theaverage valuewithin a cluster is representedcolorimetrically fromblue(negative) to red (positive). The rectangle at the intersection of two cassettesrepresents the relation between the twocassettes. (C) A pictorial representationof the relationships between the cassettes. A red line represents a positivecorrelation, whereas a blue line represents a negative correlation.




hD

ownloaded from

the Total STAT Protein Cassette and the Adaptive Immunity Cassette existedonly in the uninfected condition (Fig. 6B).

B cells are regulated by neutrophils during infectionOf the four functional cassettes identified, only the Innate ImmunityCassette (containing neutrophil frequency) was unique to infected mice(Fig. 6A), and correlations between neutrophil frequency and other nodeswere reversed during infection relative to correlations in the data from un-infected mice (Fig. 3A). Given the observations that neutrophil frequencyswitched to a negative correlation with the extent of phosphorylation ofSTAT1 in B cells during infection and that B cell basal pSTAT3 abundancealso switched correlationswithmanynodes (Fig. 3A),we investigatedBcellSTAT signaling. Mice were administered granulocyte colony-stimulatingfactor (G-CSF) to induce granulopoiesis and sacrificed 3 days later, and thensplenocytes were analyzed. In both infected and uninfected mice, G-CSFdampened baseline STAT1 and STAT3 phosphorylation, aswell as inhibitedthe IL-6– and IL-10–dependent phosphorylation of STAT3 in splenic Bcells (Fig. 7A). Furthermore, neutrophil frequencies were increased ininfected mice, and granulocyte depletion by a neutralizing antibody ledto augmented STAT signaling. Neutrophil depletion increased the basalphosphorylation of STAT1 and STAT3 and enhanced the IL-6–induced phos-phorylation of STAT3 in splenic B cells from these mice relative to that in Bcells from untreated mice (Fig. 7B). Together, these results suggest that neu-trophils suppress STAT signaling in B cells and that neutrophils are abundantenough during infection to exert this effect.

ww

DISCUSSION

Systems biology approaches reveal coordinated immuneactivity during infectionThiswork used the structured variation in a continuum of immune responses toSalmonella infection to discover organized biologically functional immune cas-settes that were associated with infection severity. The values representingmultiple attributes of immune system function were evaluated during chronicSalmonella infection and inuninfectedmice.Byclustering the correlationvaluesof these measurements of immune activation into functional cassettes, elementsof immunity that acted in concert during bacterial challengewere identified. In-fection induced the amplification of certain aspects of the naïve immunenetwork, whereas the relationships among key regulators, such as neutrophilsand Tregs, and the system were altered upon infection. In addition, differencesin the immune systemassociatedwith transmission statuswere detected becausesuper-shedders exhibited a globally increased innate response and reducedadaptive response and STAT signaling compared to those of moderate sheddersof bacterial load in feces. In addition, the identified relationships among aspectsof immunity enabled us to hypothesize and show by perturbation that the onsetof neutrophilia during infection in turn regulated B cell STAT signaling.

Cassette intensities predict super-shedderimmune phenotypeMice with high fecal loads of S. Typhimurium can transmit disease and areknown as super-shedders. Compared to infectedmice that do not shed large


on August 21, 2020

ttp://stke.sciencemag.org/

amounts of bacteria, super-shedders have asuppressed adaptive immune response (asmeasured by reduced numbers of TH1 andplasma cells and increased numbers of Tregs),a dampened STAT Signaling Cassette, andincreased neutrophils, serum IL-6, and nu-clear factor kB (NF-kB) pathway activation(characteristics of an activated innate immunesystem). Therefore, measurement of repre-sentativemarkers from the identifiedcassettesprovided information on intestinal disease se-verity and potential for pathogen transmis-sion. This could prove useful in cases ofhost-adapted pathogens, where identificationof a transmitter is of key importance to haltthe spread of disease, but consistent detectionof the pathogen is difficult. Identification oftransmitters by representative biomarkerswould enable more reliable detection andfaster intervention. Additionally, this systemsanalysis approach could likely be extended toother disease systems to guide the identifica-tion of peripheral blood markers that are pre-dictive of severity and prognosis.

Healthy and diseasednetworks differIn comparing the correlation network thatformed the basis of immune homeostasis inuninfected mice to Salmonella infection, no-table differences in both pairwise correlationsand relationships between functional cas-settes were detected. The number of nodecorrelations increased during infection, likelybecause of the activity and variability of the

A Uninfected correlation matrix arranged as infected

A STAT signaling/B cell cassette

P = 0.001

B Total STAT cassette

P = 0.001

C Adaptive immunity cassette

P = 0.001

Uncorrelated

P > 0.05 (ns)

D Innate imunity cassette

P > 0.05 (ns)

STAT signaling/ B cell cassette

Total STAT cassette

Adaptive immunity cassette

Innate immunity cassette

B Uninfected immune network Fig. 6. Uninfected mice exhibit co-regulation of adaptive, but not innate,immunity markers. (A) The correlationnetwork data obtained from 20 un-infected mice in fig. S2A is shown inmatrix form. The matrix was arrangedin the same order as that for infectedmice in Fig. 5. n.s., not significant.(B)Representationof the relationshipsbetween cassettes.




hD

ownloaded from

immune response. Four of 125 nodes switched the directionality ofmultiple correlations after the onset of infection. For example, splenicneutrophil frequency was positively correlated with B cell STAT1signaling and plasma cell frequencies in uninfected mice, but these factorswere inversely related during infection. Granulocyte depletion experi-ments demonstrated that neutrophils affected these aspects of B cell biology.Furthermore, neutrophils reduced the basal pSTAT3 abundance in B cells,and this node reversed the directionality of many correlations during infec-tion. This finding suggests that B cell pSTAT3andother nodes are co-regulatedin the uninfected state, but that during infection, the influence of neutrophiliaon B cell signaling overrides alternative regulation. Indeed, the discovery byPuga et al. (24) of B helper neutrophils (NBH) demonstrates that neutrophilsaffect splenic B cell function. The authors implicate mucosal bacteria in driv-ing NBH production, which in turn secrete IL-21, BAFF (B cell–activatingfactor), and APRIL (a proliferation-inducing ligand) to support B cell anti-body production. This finding shows that neutrophils influenceB cell activityand may provide a basis for the regulation of B cell signaling observed in thecontext of infection.

Salmonella infection also influenced the relationship between groups ofimmune system attributes that worked in concert. First, the correlationsamong the markers identified as the Innate Immunity Cassette were ob-served only when the immune systemwas challenged by infection, presum-ably because of the absence of a basal innate immune response in uninfectedmice. Second, the Adaptive Immunity Cassette and Total STAT ProteinCassette were positively correlated with each other in unchallenged mice,

ww

but this association was lost during infection. Therefore, we infer thatSTAT1 and STAT3 proteins are co-regulated with adaptive immunity inthe basal state, but that during infection, additional mechanisms disrupt thiscorrelation.

Crosstalk occurs between immune system organsDuring infection, some extracellular Salmonella exit the host through theintestinal tract, and a reservoir of intracellular bacteria persists in macro-phages located in the spleen and mLN. Here, the bacterial load in each ofthese organs was associated with a separate cassette. Fecal CFUwas part ofthe Innate Immunity Cassette, splenic CFU was incorporated within theTotal STAT Protein Cassette, and CFU in the mLN was not included in afunctional cassette. Therefore, the local bacterial burden in the spleen wascorrelated with the increased abundance of STAT1 and STAT3 proteins insplenocytes, yet the increased signaling potential of these transcriptionfactors in splenocytes was associated with low quantities of Salmonella inthe lower intestine. IL-2–induced STAT5 phosphorylation in CD4 T cellsfrom themLNwas correlatedwith splenicCFUand total amounts of STAT1and STAT3 proteins. IL-2 induces the proliferation of activated T cells, sug-gesting that the ability to mount an adaptive response in the mLN is asso-ciated with bacterial load in the spleen, but not in the intestine or the mLNitself. Serum cytokine concentrationswere also associated with the immunestatus of individual organs. For example, neutrophils in the spleenwere pos-itively correlated with the acute-phase concentrations of IL-6 but wereinversely correlated with the amount of the anti-inflammatory cytokine


on August 21, 2020

ttp://stke.sciencemag.org/

IL-10. The immune network of uninfectedmice also exhibited correlations betweenorgans. In one example of these interactions,the frequency of neutrophils in the colonshared 18 negative correlations with splenicSTAT signaling. Thus, in the basal state,there was co-regulation between neutrophilrecruitment to the colon and cytokinesignaling potential in the spleen. Further-more, this relationship between colonic neu-trophils and splenocytes was abrogatedduring infection. Thus, infection disruptsthe underlying systemic regulation of theimmune system.

A network-level approachreveals interlockingrelationships that balance innateand adaptive immunitySalmonella causes a TH1-biasing infection,and the immune response to this organismrequires both IFN-g and T-bet (25). Salmonella-specific CD4+ T cells produce IFN-g ex vivo(26), and we detected these cells at 8 d.p.i.,and they continued to increase in numberthrough the first month of infection in themodel used here. Characterization of theCD4+ T cells that produced IFN-g indicatedthat cytokine production increased with Ly-6C abundance in the effector/memory popu-lation. Ly-6C is a marker for professionalmemory CD4+ T cells that reside primarilyin the bone marrow (19). The increase inthe number of cells in this population in thespleens of infected mice relative to that in

Unstim IL-6 IL-100

50

100

150

200

250

*

**

*

UnstimUnstim IL-60

50

100

150

200

250*

****

0

50

100

150

200

250** *

Unstim

0

50

100

150

200

250

UninfectedUninfected + G-CSFInfectedInfected + G-CSF

A

B

Anti–Ly-6GAnti-Gr1

PBS

pSTA

T3

MF

I in

B c

ells

pSTA

T3

MF

I in

B c

ells

pSTA

T1

MF

I in

B c

ells

pSTA

T1

MF

I in

B c

ells

Fig. 7. Neutrophils regulate STAT signaling in B cells. (A) Uninfected and 30-day Salmonella-infected non–super-shedder mice were left untreated or were given 1 mg of G-CSF intraperitoneally daily for 3 days beforebeing sacrificed on the fourth day. Data are means ± SD of four mice per group. The splenocytes were leftuntreated or were treated with IL-6 (40 ng/ml) or IL-10 (80 ng/ml) for 15 min. Fixed and permeabilized cellswere stained for phenotypic markers, and the relative abundances of pSTAT1 and pSTAT3 in B cells werequantified by flow cytometric analysis. *P < 0.05 and **P < 0.001, two-sided t test. (B) Infected mice weretreated with PBS as a control or were treated with 1 mg of anti–Ly-6G (clone IA8) or anti-Gr1 (clone RBC-8C5)antibody intraperitoneally daily for 3 days and then were sacrificed on the fourth day. Data are means ± SD ofthreemicepergroup. Thesplenocyteswere left untreated or were treatedwith IL-6 (40 ng/ml) for 15min. Fixedand permeabilized cells were stained for phenotypic markers, and the relative abundances of pSTAT1 andpSTAT3 in B cells were quantified by flow cytometric analysis. *P < 0.05 and **P < 0.001, two-sided t test.




http://stke.sciD

ownloaded from

uninfected animals supports the sustained recruitment ofmemoryCD4+T cellsto infection sites during persistent infection. Although the clonality of theCD4+ T cell populations was not investigated here, a steady increase in thenumber of antigen-specific, IFN-g–producing memory CD4+ T cells wasobserved after infection (fig. S1D).

In summary, markers of the TH1 response, antibody response, cytokinesignaling, and innate immunity were used to evaluate how infection per-turbed the immune system network. Infection induced innate immune re-sponses in infected animals, which were greatest in mice harboring highfecal bacterial loads. These mice had systemic neutrophilia and high serumconcentrations of IL-6, as well as fewer plasma cells and TH1 cells andgreater numbers of Treg, compared with infected mice with lower fecal bac-terial loads. Splenic neutrophils were negatively correlated with B cellsignaling only in infected mice, and direct interventions revealed that neu-trophils regulated B cell signaling potential.

Common approaches to understanding heterogeneity in biologicalsystems involve analyses of individual correlations to uncover testable re-lationships, but often fail to take full advantage of natural variation withinthe data. Our approach to this unmet need was to gain additional insights byclustering natural variation as correlations that revealed a series of interact-ing cassettes (which further served to model a system-wide view of the hostimmune response). When perturbations (such as infection) induced the ap-pearance of new cassettes or new relationships between preexisting cas-settes, key drivers of the immune response were identified. Additionally,elements of the host immune response that were preserved or amplifiedthrough infection could be distinguished from those that were altered orreversed because they acted in concert.We believe that such approacheswillbe useful in understanding and structuring so-called heterogeneity in awiderange of biological systems.

on August 21, 2020

encemag.org/

MATERIALS AND METHODS

MiceFemale 129X1/SvJ mice (The Jackson Laboratory) were 8 to 10 weeks oldat the time of infection. All mice were handled in accordancewith StanfordUniversity animal care guidelines. Neutrophil intervention experimentswere performed as previously described (14). For neutrophil inductionexperiments, mice were injected intraperitoneally with 1 mg of PEGylatedG-CSF (GenScript, Z00393-50) on each of three consecutive days andweresacrificed on the fourth day. For neutrophil depletion experiments, micewere injected intraperitoneally with 1 mg of anti–Ly-6G (clone IA8, BioX-cell, BE0075-1) or anti-Gr1 (clone RB6-8C5, BioXcell, BE0075-1) onthree consecutive days and then were sacrificed on the fourth day.

Bacterial infectionThe S. Typhimurium strain SL1344 was grown overnight in LB. Bacteriawerewashed and diluted. Inoculationswere performed either by oral gavageof 108 CFU in 200 ml of phosphate-buffered saline (PBS) or by feedingbread containing 108 CFU.

Cytokine stimulationmLNs and spleenswere harvested andmechanically dissociated into single-cell suspensions at a concentration of 1 × 107 cells/ml in RPMI-1640containing 10% fetal bovine serum (FBS) and penicillin, streptomycin,and glutamine (RPMI-10). Bone marrow was isolated from mouse femursand tibias by flushing the bones with RPMI-10. Dissociated cells wereallowed to recover for 15 min at 37°C. Cells were then left unstimulatedor stimulatedwith IFN-g, IL-2, IL-4, IL-6, IL-12 (all fromBDBiosciences),or IL-21 (R&D) (all at 40 ng/ml) or with IL-10 (BDBiosciences; 80 ng/ml).

www

Exposure to cytokines was performed at 37°C for 15 min. Cells were thenfixed with 1.6% paraformaldehyde (PFA) at room temperature for 10 min,washed, resuspended in methanol at 4°C, and stored at −80°C. Blood wascollected by cardiac puncture with needles preloaded with heparin (Sigma-Aldrich) and was left unstimulated or was treated with either IL-6 or IL-12(40 ng/ml) for 15 min. Cells were then fixed with Lyse/Fix Buffer (BDBio-sciences) according to the manufacturer’s guidelines. The cells were thenwashed and resuspended in cold methanol and stored at −80°C.

Bacterial countsFecal pelletswereweighed and dissociated in 1ml of sterile PBS.mLNs andspleens were weighed and dissociated in RPMI-10. Serial dilutions wereplatedonLBagar plates containing streptomycin.TheCFUof S.Typhimuriumper weight of organ or feces was then calculated.

T cell activationBonemarrow–derivedmacrophageswere plated at 2.5×105 cells in a 24-welldish (Corning) and infected overnight with S.Typhimurium at amultiplicityof infection of 5 at 37°C in RPMI-1640 with 10% FBS and gentamicin.After 18 hours, 5 × 106 splenocytes from infectedmice in RPMI-10withoutantibiotics were added to the culture. After 1 hour, brefeldin A (10 mg/ml)was added, and cellswere cultured for an additional 3hours. For the last 10minof culture, Pacific Blue maleimide (0.01 ng/ml) was added to the culture tostain membrane-permeabilized dead cells. Cells were then fixed with 1.6%PFA and collected for intracellular cytokine staining.

Flow cytometryFor staining of cell surface markers and transcription factors, fixed cellswerewashed twicewith stainingmedium (PBS containing 0.5%bovine ser-um albumin and 0.05% sodium azide). Cells were first stained with a panelof antibodies specific for cell surface markers for 30 min in staining medi-um. The antibodies used were Cy7-phycoerythrin (PE)–conjugated anti-CD8, Texas Red (TR)–PE–conjugated anti-B220, Cy7-allophycocyanin(APC)–conjugated anti-CD4, Pacific Orange–conjugated anti-CD44, bioti-nylated anti–CD62 ligand (CD62L), Cy5.5–peridinin chlorophyll protein(PerCP)–conjugated anti-CD25, fluorescein isothiocyanate (FITC)–conjugated anti–Ly-6C, and PE-conjugated CD49d (BD Biosciences).Cells were permeabilized with staining medium containing 0.3% saponinand subsequently stained with the following antibodies: Alexa fluorophore(Ax) 647–anti–T-bet, Ax700–anti-FoxP3 (both from eBioscience), PacificBlue–conjugated anti–Ki-67 (BD Biosciences), and streptavidin quantumdot (QD) 605 (Invitrogen). A macrophage antibody cocktail consisting ofTR-PE–conjugated anti-B220, Cy5.5-PerCP–conjugated anti-CD11b, APC-conjugated anti-CD11c, Cy7-APC–conjugated anti-Gr1, FITC-conjugatedanti-CD86 (all from BD Biosciences), and Cy7-PE–conjugated anti-F4/80(eBioscience) was also used in some experiments. After staining, cells werewashed with staining medium containing saponin and then were analyzedon an LSR II flow cytometer equipped with 407-, 488-, and 633-nm lasers.Digital datawere acquiredwith BDFACSDiva software, with >50,000 size-gated cells collected per sample. Data were analyzed with FlowJo software(Tree Star). For intracellular cytokine staining, cells were first stained withantibodies against surface markers: Ax700–anti-CD4, Cy7-PE–anti-CD8,FITC–anti–Ly-6C, Pacific Orange–anti-CD44, biotinylated anti-CD62L, andCy5.5-PerCP–anti-CD11b (all from BD Biosciences). Anti-CD44 wasconjugated to PacificOrange succinimidyl ester (Invitrogen) in house. Cellswere washed and permeabilized with saponin staining medium. Cells werethen stainedwithAPC–anti–IFN-g, PE–anti–IL-2, and streptavidinQD605.For staining of phospho-proteins, cells stored in methanol were washedtwice with FACS staining buffer and then stained with the following anti-bodies: Ax700–anti-CD4, Pacific Blue–anti-CD8, TR-PE–anti-B220,

.SCIENCESIGNALING.org 12 January 2016 Vol 9 Issue 410 ra4 10



on August 21, 2020


Dow

nloaded from

Cy5.5-PerCP–anti-CD11b, Cy7-APC–anti-Gr1, Pacific Orange–anti-CD44, Cy7-PE–anti-CD62L, and Cy7-PE–anti-F4/80. In addition, cellswere stained with a pair of phospho-specific antibodies, either Ax488–anti-pSTAT1 and Ax647–anti-pSTAT3 or Ax647–anti-pSTAT4 andAx488–anti-pSTAT5 (all from BD Biosciences), or with PE–anti–totalSTAT1 (BD Biosciences) or Ax647–anti–total IkBa (inhibitor of NF-kBa) (Cell Signaling Technologies). Cells were washed and then analyzedon an LSR II flow cytometer.

Serum cytokine and antibody quantificationBlood was collected by cardiac puncture and centrifuged; the serum su-pernatant was collected and stored at −80°C. To determine the concentra-tions of IL-6, IL-10, and IFN-g, the Cytometric Bead Array Flex Set (BDBioscience) was used according to the manufacturer’s recommendations.For Salmonella-specific antibody detection, serum was incubated withplate-bound Salmonella lysate. Plates were washed, and horseradishperoxidase–conjugated goat anti-mouse immunoglobulin G antibodywas applied. Plates were analyzed with a plate reader for relative quan-tification of antibody.

VisualizationsSPADE clusters flow cytometry events on the basis of marker staining in-tensity and displays the clusters in a minimum spanning tree, as previouslydescribed (15, 27). To create a SPADE tree, each time point and an equalnumber of events per flow cytometry standard (FCS) file of each of fivemice were concatenated with the flowCore package (www.bioconductor.org/packages/2.3/bioc/html/flowCore.html). User-defined cell subsets wereidentified on the basis of the surface marker intensity within a set of cellclusters. Figure 1was generatedwith a beta version of SPADE, and the pub-lic version is now available at (http://cytospade.org). The flow cytometryplots in fig. S1C were generated with MATLAB (MathWorks), whereasother flow cytometry plots were generated with FlowJo (Tree Star). Figures2B and 6were madewith Prism (GraphPad). The cell signaling heat map inFig. 2A was generated with Cytobank (www.cytobank.org). The networkrepresentations in fig. S2 were generated with Cytoscape (www.cytoscape.org). R software (www.r-project.org) was used for clustering using Ward’smethod and generating matrices. The matrices in Figs. 4 and 5 were adjustedto a red-blue color scheme in Adobe Illustrator.

Statistical analysisTheK-S test was used to test for statistically significant differences in STATphosphorylation between uninfected and infected mice (Fig. 2A), and theLevene test was used to calculate significant differences in variance com-pared to the median. The c2 test was used to test for the enrichment of cor-relationswithin cassettes in uninfectedmice. Pairwise comparisons betweengroups were performed with a t test, with P < 0.05 considered a statisticallysignificant difference. For analyses of correlations, Spearman’s rank corre-lation test was performed on batches. Data from two groups of uninfectedmice (10 mice each) and two groups of infected mice (8 and 11 mice each)were processed. To establish a random distribution of correlation values foreach group for each measurement, the value was assigned to a randommouse, and then the correlation values between each of the other mice weredetermined. This was repeated 50 times to generate a bell-shaped distri-bution of correlation values. Next, the probability of every correlationoccurring was determined on the basis of the random distribution curve.A correlation was deemed significant if it met two criteria: (i) the productof the probabilities was ≤0.025, and (ii) the probability was in the top thirdin each group. This process was also repeated with two experiments of 12and 16 infected mice each, and the correlation map was very similar to theone presented.

www

SUPPLEMENTARY MATERIALSwww.sciencesignaling.org/cgi/content/full/9/410/ra4/DC1Fig. S1. The adaptive immune response to Salmonella increases through the first monthof infection.Fig. S2. The connectivity of the immune network is increased during Salmonella infection.

REFERENCES AND NOTES1. O. D. Perez, G. P. Nolan, Simultaneous measurement of multiple active kinase states

using polychromatic flow cytometry. Nat. Biotechnol. 20, 155–162 (2002).2. P. O. Krutzik, M. R. Clutter, G. P. Nolan, Coordinate analysis of murine immune cell

surface markers and intracellular phosphoproteins by flow cytometry. J. Immunol.175, 2357–2365 (2005).

3. A. N. Hotson, J. W. Hardy, M. B. Hale, C. H. Contag, G. P. Nolan, The T cell STATsignaling network is reprogrammed within hours of bacteremia via secondary signals.J. Immunol. 182, 7558–7568 (2009).

4. W. E. O’Gorman, P. Sampath, E. F. Simonds, R. Sikorski, M. O’Malley, P. O. Krutzik,H. Chen, V. Panchanathan, G. Chaudhri, G. Karupiah, D. B. Lewis, S. H. Thorne,G. P. Nolan, Alternate mechanisms of initial pattern recognition drive differentialimmune responses to related poxviruses. Cell Host Microbe 8, 174–185 (2010).

5. D. M. Longo, B. Louie, S. Putta, E. Evensen, J. Ptacek, J. Cordeiro, E. Wang, Z. Pos,R. E. Hawtin, F. M. Marincola, A. Cesano, Single-cell network profiling of peripheralblood mononuclear cells from healthy donors reveals age- and race-associateddifferences in immune signaling pathway activation. J. Immunol. 188, 1717–1725(2012).

6. Typhoid Mary, Nervous lesions in Buerger’s disease. Lancet 232, 1188 (1938).7. S. K. Hoiseth, B. A. D. Stocker, Aromatic-dependent Salmonella typhimurium are non-

virulent and effective as live vaccines. Nature 291, 238–239 (1981).8. D. M. Monack, D. M. Bouley, S. Falkow, Salmonella typhimurium persists within

macrophages in the mesenteric lymph nodes of chronically infected Nramp1+/+ miceand can be reactivated by IFNg neutralization. J. Exp. Med. 199, 231–241 (2004).

9. Y. Valdez, G. A. Grassl, J. A. Guttman, B. Coburn, P. Gros, B. A. Vallance, B. B. Finlay,Nramp1 drives an accelerated inflammatory response during Salmonella-induced colitisin mice. Cell. Microbiol. 11, 351–362 (2009).

10. T. D. Lawley, D. M. Bouley, Y. E. Hoy, C. Gerke, D. A. Relman, D. M. Monack, Hosttransmission of Salmonella enterica serovar Typhimurium is controlled by virulencefactors and indigenous intestinal microbiota. Infect. Immun. 76, 403–416 (2008).

11. J. W. Conlan, Critical roles of neutrophils in host defense against experimental sys-temic infections of mice by Listeria monocytogenes, Salmonella typhimurium, andYersinia enterocolitica. Infect. Immun. 65, 630–635 (1997).

12. T. M. Johanns, J. M. Ertelt, J. H. Rowe, S. S. Way, Regulatory T cell suppressivepotency dictates the balance between bacterial proliferation and clearance duringpersistent Salmonella infection. PLOS Pathog. 6, e1001043 (2010).

13. S. J. McSorley, M. K. Jenkins, Antibody is required for protection against virulent butnot attenuated Salmonella enterica serovar typhimurium. Infect. Immun. 68, 3344–3348(2000).

14. S. Gopinath, A. Hotson, J. Johns, G. Nolan, D. Monack, The systemic immune stateof super-shedder mice is characterized by a unique neutrophil-dependent blunting ofTH1 responses. PLOS Pathog. 9, e1003408 (2013).

15. P. Qiu, E. F. Simonds, S. C. Bendall, K. D. Gibbs Jr., R. V. Bruggner, M. D. Linderman,K. Sachs, G. P. Nolan, S. K. Plevritis, Extracting a cellular hierarchy from high-dimensionalcytometry data with SPADE. Nat. Biotechnol. 29, 886–891 (2011).

16. S. C. Bendall, E. F. Simonds, P. Qiu, E.-a. D. Amir, P. O. Krutzik, R. Finck, R. V. Bruggner,R. Melamed, A. Trejo, O. I. Ornatsky, R. S. Balderas, S. K. Plevritis, K. Sachs, D. Pe’er,S. D. Tanner, G. P. Nolan, Single-cell mass cytometry of differential immune and drugresponses across a human hematopoietic continuum. Science 332, 687–696 (2011).

17. J. Wrammert, E. Källberg, W. W. Agace, T. Leanderson, Ly6C expression differenti-ates plasma cells from other B cell subsets in mice. Eur. J. Immunol. 32, 97–103(2002).

18. S. Lacotte, M. Decossas, C. Le Coz, S. Brun, S. Muller, H. Dumortier, Early differen-tiated CD138highMHCII+IgG+ plasma cells express CXCR3 and localize into inflamedkidneys of lupus mice. PLOS One 8, e58140 (2013).

19. K. Tokoyoda, S. Zehentmeier, A. N. Hegazy, I. Albrecht, J. R. Grün, M. Löhning,A. Radbruch, Professional memory CD4+ T lymphocytes preferentially reside and rest inthe bone marrow. Immunity 30, 721–730 (2009).

20. E. Román, E. Miller, A. Harmsen, J. Wiley, U. H. von Andrian, G. Huston, S. L. Swain,CD4 effector T cell subsets in the response to influenza: Heterogeneity, migration,and function. J. Exp. Med. 196, 957–968 (2002).

21. D. S. McDermott, S. M. Varga, Quantifying antigen-specific CD4 T cells during a viralinfection: CD4 T cell responses are larger than we think. J. Immunol. 187, 5568–5576(2011).

22. J. D. Miller, R. G. van der Most, R. S. Akondy, J. T. Glidewell, S. Albott, D. Masopust,K. Murali-Krishna, P. L. Mahar, S. Edupuganti, S. Lalor, S. Germon, C. Del Rio,M. J. Mulligan, S. I. Staprans, J. D. Altman, M. B. Feinberg, R. Ahmed, Human effector




D

and memory CD8+ T cell responses to smallpox and yellow fever vaccines. Immunity 28,710–722 (2008).

23. M. Dominguez-Villar, C. M. Baecher-Allan, D. A. Hafler, Identification of T helper type1–like, Foxp3+ regulatory T cells in human autoimmune disease. Nat. Med. 17, 673–675(2011).

24. I. Puga, M. Cols, C. M. Barra, B. He, L. Cassis, M. Gentile, L. Comerma, A. Chorny,M. Shan, W. Xu, G. Magri, D. M. Knowles, W. Tam, A. Chiu, J. B. Bussel, S. Serrano,J. A. Lorente, B. Bellosillo, J. Lloreta, N. Juanpere, F. Alameda, T. Baró, C. Díaz de Heredia,N. Torán, A. Català, M. Torrebadell, C. Fortuny, V. Cusí, C. Carreras, G. A. Diaz,J. M. Blander, C.-M. Farber, G. Silvestri, C. Cunningham-Rundles, M. Calvillo, C. Dufour,L. D. Notarangelo, V. Lougaris, A. Plebani, J.-L. Casanova, S. C. Ganal, A. Diefenbach,J. I. Aróstegui, M. Juan, J. Yagüe, N. Mahlaoui, J. Donadieu, K. Chen, A. Cerutti, B cell–helper neutrophils stimulate the diversification and production of immunoglobulin in themarginal zone of the spleen. Nat. Immunol. 13, 170–180 (2012).

25. R.Ravindran, J. Foley, T. Stoklasek, L. H.Glimcher, S. J.Mcsorley, Expressionof T-bet byCD4T cells is essential for resistance to Salmonella infection. J. Immunol. 175, 4603–4610 (2005).

26. H.-W. Mittrücker, A. Köhler, S. H. E. Kaufmann, Characterization of the murine T-lymphocyteresponse to Salmonella enterica serovar Typhimurium infection. Infect. Immun. 70, 199–203(2002).

27. B. Bodenmiller, E. R. Zunder, R. Finck, T. J. Chen, E. S. Savig, R. V. Bruggner,E. F. Simonds, S. C. Bendall, K. Sachs, P. O. Krutzik, G. P. Nolan, Multiplexed masscytometry profiling of cellular states perturbed by small-molecule regulators. Nat. Biotechnol.30, 858–867 (2012).

www

Acknowledgments: We thank K. Sachs for guidance in determining correlations,R. Bruggner for aid in creating computational figures, A. Trejo and A. Jager for assistingin laboratory execution, and W. O’Gorman for critical reading of the manuscript. Funding:This work was supported by NIH grant HHSN272200700038C and National Heart, Lung,and Blood Institute grant N01-HV-00242 (to G.P.N.) and a Burroughs Wellcome FundInvestigator in the Pathogenesis of Infectious Diseases (to D.M.). A.N.H. was funded byMolecular and Cellular Immunobiology training grant 5 T32 AI07290. S.G. was funded bythe Stanford Graduate Fellowship and supported by NIH grant RO1 A1095396. The fundershad no role in study design, data collection and analysis, decision to publish, or prepa-ration of the manuscript. Author contributions: A.N.H., S.G., G.P.N., and D.M. concep-tualized the experiments and wrote the manuscript; A.N.H. and S.G. performed theexperiments; M.N. and A.K. provided statistical support; and R.F. assisted in data inter-pretation and figure generation. Competing interests: The authors declare that theyhave no competing interests.

Submitted 4 April 2015Accepted 17 December 2015Final Publication 12 January 201610.1126/scisignal.aaa9303Citation: A. N. Hotson, S. Gopinath, M. Nicolau, A. Khasanova, R. Finck, D. Monack,G. P. Nolan, Coordinate actions of innate immune responses oppose those of theadaptive immune system during Salmonella infection of mice. Sci. Signal. 9, ra4 (2016).

ow


on August 21, 2020


nloaded from


infection of miceSalmonelladuring Coordinate actions of innate immune responses oppose those of the adaptive immune system

Andrew N. Hotson, Smita Gopinath, Monica Nicolau, Anna Khasanova, Rachel Finck, Denise Monack and Garry P. Nolan

DOI: 10.1126/scisignal.aaa9303 (410), ra4.9Sci. Signal.

the immune system puzzle.during the course of infection. Thus, this systems biology approach enables the assembly of interconnected sections of components of the immune system interacted with each other. For example, neutrophils inhibited the functions of B cellsmultiparametric data set revealed response elements that clustered together into patterns that showed how different of serum cytokine concentrations, antibody responses, and bacterial burden. Mathematical analysis of thismeasuring the numbers and signaling states of multiple immune cell types and combining these data with measurements

-infected mice over time bySalmonella. analyzed the variable immune responses of et alsystem puzzle. Hotson Analysis of individual components of the immune response to infection solves only small parts of the immune

Assembling the immune system jigsaw

ARTICLE TOOLS http://stke.sciencemag.org/content/9/410/ra4

MATERIALSSUPPLEMENTARY http://stke.sciencemag.org/content/suppl/2016/01/08/9.410.ra4.DC1

CONTENTRELATED

http://stke.sciencemag.org/content/sigtrans/10/494/eaan6284.fullhttp://stke.sciencemag.org/content/sigtrans/10/471/eaan2405.fullhttp://science.sciencemag.org/content/sci/355/6330/1211.fullhttp://stke.sciencemag.org/content/sigtrans/9/421/eg4.fullhttp://science.sciencemag.org/content/sci/346/6213/1250689.fullhttp://science.sciencemag.org/content/sci/349/6244/1259425.fullhttp://stke.sciencemag.org/content/sigtrans/8/399/ra104.fullhttp://stke.sciencemag.org/content/sigtrans/8/407/ra129.full

REFERENCES

http://stke.sciencemag.org/content/9/410/ra4#BIBLThis article cites 27 articles, 12 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.Science SignalingYork Avenue NW, Washington, DC 20005. The title (ISSN 1937-9145) is published by the American Association for the Advancement of Science, 1200 NewScience Signaling

Copyright © 2016, American Association for the Advancement of Science

on August 21, 2020


Dow

nloaded from

http://stke.sciencemag.org/content/9/410/ra4

http://stke.sciencemag.org/content/suppl/2016/01/08/9.410.ra4.DC1

http://stke.sciencemag.org/content/sigtrans/8/407/ra129.full

http://stke.sciencemag.org/content/sigtrans/8/399/ra104.full

http://science.sciencemag.org/content/sci/349/6244/1259425.full


http://stke.sciencemag.org/content/sigtrans/9/421/eg4.full


http://stke.sciencemag.org/content/sigtrans/10/471/eaan2405.full

http://stke.sciencemag.org/content/sigtrans/10/494/eaan6284.full

http://stke.sciencemag.org/content/9/410/ra4#BIBL

http://www.sciencemag.org/help/reprints-and-permissions

http://www.sciencemag.org/about/terms-service


Coordinate actions of innate immune responses …...SYSTEMS IMMUNOLOGY Coordinate actions of innate immune responses oppose those of the adaptive immune system during Salmonella infection

Documents