Role of type I interferon signaling in human ... - Journal of Virology

2 downloads 0 Views 3MB Size Report
Feb 4, 2015 - Pavlo Gilchuk*, Sebastian Joyce*, John V. Williams. *,†,‡ ...... Schuster JE, Cox RG, Hastings AK, Boyd KL, Wadia J, Chen Z, Burton DR,. 556.
JVI Accepted Manuscript Posted Online 4 February 2015 J. Virol. doi:10.1128/JVI.03275-14 Copyright © 2015, American Society for Microbiology. All Rights Reserved.

Type I IFN affects early and late immunity to HMPV

Hastings et al

Role of type I interferon signaling in human metapneumovirus pathogenesis and control of viral replication. Running Title: Type I IFN affects early and late immunity to HMPV

Andrew K. Hastings*, John J. Erickson*, Jennifer E. Schuster†, Kelli L. Boyd*, Sharon J. Tollefson†, Monika Johnson†, Pavlo Gilchuk*, Sebastian Joyce*, John V. Williams*,†,‡

Departments of *Pathology, Microbiology & Immunology and †Pediatrics, Vanderbilt University School of Medicine, Nashville, TN 37232 Funding Sources: ‡

Corresponding Author:

John V. Williams 1161 21st Avenue South, D-7235 MCN Nashville, TN 37232-2581 Telephone:

615-936-2186

Facsimile:

615-343-9723

Electronic:

[email protected]

Abstract word count: 218/250 Manuscript character count: 5339

Type I IFN affects early and late immunity to HMPV

Hastings et al

1

Abstract

2

Type I interferon receptor (IFNAR) signaling regulates the expression of proteins that

3

are crucial contributors to immune responses. Paramyxoviruses including human

4

metapneumovirus (HMPV) have evolved mechanisms to inhibit IFNAR signaling, but the

5

specific contribution of IFNAR signaling to control of HMPV replication, pathogenesis,

6

and adaptive immunity is unknown. We used IFNAR-deficient (IFNAR-/-) mice to assess

7

the effect of IFNAR signaling on HMPV replication and CD8+ T cell response. HMPV-

8

infected IFNAR-/- mice had a higher peak of early viral replication, but cleared virus with

9

similar kinetics to wild-type (WT) mice. However, IFNAR-/- mice infected with HMPV

10

displayed less airway dysfunction and lung inflammation. CD8+ T cells in IFNAR-/- mice

11

post-HMPV expressed similar levels of the inhibitory receptor programmed death-1 (PD-

12

1) as WT. However, despite lower expression of the inhibitory ligand PD-L1, HMPV-

13

specific CD8+ T cells in IFNAR-/- mice were more functionally impaired than WT and

14

upregulated the inhibitory receptor TIM-3. Analysis of the antigen presenting cell

15

subsets in the lungs revealed that the expansion of PD-L1lo dendritic cells (DCs), but not

16

PD-L1high alveolar macrophages, was dependent on IFNAR signaling. Collectively, our

17

results indicate a role for IFNAR signaling in the early control of HMPV replication,

18

disease progression, and the development of an optimal adaptive immune response.

19

Moreover, our findings suggest an IFNAR-independent mechanism of lung CD8+ T cell

20

impairment.

Type I IFN affects early and late immunity to HMPV

Hastings et al

21

Importance

22

Human metapneumovirus (HMPV) is a leading cause of acute respiratory illness. CD8+

23

T cells are critical for clearing viral infection, yet recent evidence shows that HMPV and

24

other respiratory viruses induce CD8+ T cell impairment via PD-1:PD-L1 signaling. We

25

sought to understand the role of type I interferon (IFN) in the innate and adaptive

26

immune response to HMPV using a mouse model lacking IFN signaling. Although

27

HMPV titers were higher in the absence of type I IFN, virus was nonetheless cleared

28

and mice were less ill, indicating that type I IFN is not required to resolve HMPV

29

infection but contributes to pathogenesis. Further, despite lower levels of the inhibitory

30

ligand PD-L1 in mice lacking type I IFN, CD8+ T cells were more impaired in these mice

31

than in WT mice. Our data suggest that specific antigen-presenting cell subsets and the

32

inhibitory receptor TIM-3 may contribute to CD8+ T cell impairment.

Type I IFN affects early and late immunity to HMPV

33 34

Hastings et al

Introduction Human metapneumovirus (HMPV) is a leading cause of acute lower respiratory

35

infection (LRI), with infants, elderly, and immunocompromised at the highest risk for

36

severe complications from viral infection (1-9). No licensed therapeutics or vaccines

37

exist to combat or prevent HMPV infection. Nearly all individuals have been exposed to

38

HMPV by the age of five years (10, 11). Infection with this virus results in a neutralizing

39

antibody response in almost all healthy individuals, but data show that neutralizing

40

antibody titers present in a large percentage of previously infected people are

41

insufficient to prevent reinfection (12-14). This indicates that humoral immunity alone is

42

insufficient for complete protection from HMPV in humans. The mechanism by which

43

HMPV evades the adaptive immune system is still unknown, but recent evidence

44

suggests that impairment of the lung CD8+ T cell response following HMPV infection is

45

a contributing factor (15). In contrast to humans, infection of immunocompetent mice

46

with HMPV results in sterilizing immunity, preventing reinfection (16, 17).

47

HMPV, like other members of the Paramyxoviridae family such as respiratory

48

syncytial virus (RSV) and parainfluenza viruses (PIV), can subvert the innate immune

49

response through modulation of the type I interferon (IFN) signaling pathway (18, 19).

50

Type I IFN signaling, which is initiated through activation of the interferon α receptor

51

(IFNAR), is thought to be integral to the early immune response through the induction of

52

anti-viral effector molecules (20-22). In addition, this pathway can modulate the adaptive

53

immune response by contributing to both clonal expansion and maintenance of memory

54

T cells as well as priming and differentiation of antigen presenting cells (APCs) (23-26).

Type I IFN affects early and late immunity to HMPV

Hastings et al

55

Recent data indicate that HMPV infection generates functionally impaired virus-

56

specific CD8+ T cells in the lungs as a result of signaling through the inhibitory receptor

57

programmed death 1 (PD-1) (15). PD-1, along with other inhibitory receptors, has been

58

shown to be highly upregulated in both cancer and chronic viral infections (27-29), but

59

little is known about the role of PD-1 in acute respiratory viral infections. The ligand for

60

PD-1, programmed death ligand 1 (PD-L1), is expressed on professional APCs as well

61

as primary infected lung epithelial cells and is thought to be induced in an IFN-

62

dependent manner (30, 31).

63

In this study, we used an established model of HMPV infection to demonstrate

64

that genetic ablation of the IFN α receptor (IFNAR-/- mice) diminished the HMPV-specific

65

CD8+ T cell response. We found that although IFNAR-deficient animals were able to

66

clear the virus after infection, and developed significantly higher antibody titers, they

67

displayed lower overall disease and lung inflammation than wild-type animals. Despite

68

similar PD-1 expression and lower PD-L1 expression levels in IFNAR-/- mice during

69

HMPV infection compared to wild-type mice, HMPV-specific CD8+ T cells were more

70

impaired in IFNAR-/- than WT. T cell Ig- and mucin-domain–containing molecule–3 (TIM-

71

3) was significantly upregulated on HMPV-specific CD8+ T cells in the IFNAR-/- animals.

72

Further investigation into the specific APC subsets in the lung during HMPV infection

73

showed that alveolar macrophages do not rely on IFNAR signaling for expansion or the

74

expression of PD-L1, but significantly fewer dendritic cells (DCs) were found in IFNAR-/-

75

mice during HMPV infection. Both DCs and interstitial macrophages upregulated PD-L1

76

in an IFNAR-dependent manner, while alveolar macrophages expressed high levels of

Type I IFN affects early and late immunity to HMPV

Hastings et al

77

the inhibitory ligand, compared to other lung APC subsets, in both infected and

78

uninfected WT and IFNAR-/- mice.

Type I IFN affects early and late immunity to HMPV

79

Hastings et al

MATERIALS AND METHODS

80 81

Mice and Viruses. C57BL/6 (B6) mice were purchased from The Jackson Laboratory.

82

IFN alpha/beta receptor deficient (IFNAR-/-) B6 mice were kindly provided by Dr. Herbert

83

W. Virgin. All animals were bred and maintained in specific pathogen-free conditions

84

under guidelines approved by the AAALAC and the Vanderbilt Institutional Animal Care

85

and Use Committee. Six to twelve-week-old age- and gender-matched animals were

86

used in all experiments. HMPV (pathogenic clinical strain TN/94-49, genotype A2) was

87

grown and titered in LLC-MK2 cells as previously described (32). For all experiments,

88

mice were anesthetized with intraperitoneal ketamine-xylazine and infected intranasally

89

with 1.5x106 PFU of HMPV. Serum was collected from WT and IFNAR-/- mice by

90

submandibular venipuncture and was used in a plaque reduction assay to determine

91

HMPV neutralizing antibody titers as previously described (32). The median of triplicate

92

values for % HMPV neutralization for each animal of each genotype were plotted as a

93

function of log2-transformed serum dilution factor and analyzed in a sigmoidal dose

94

response curve to determine IC50. Nasal turbinates (NT) and lungs were collected for

95

analysis and tissue viral titers measured by plaque titration as previously described (32).

96

For histopathology, the left lung was removed and inflated with 10% buffered formalin,

97

paraffin-embedded, stained, and analyzed using a formal scoring system in a group-

98

blinded fashion by an experienced lung pathologist as previously described (15).

99 100

Pulse Oximetry. To measure breath distension, mice were anesthetized using an

101

isofluorane:oxygen (2%:98%) (2 liters/min) mixture (VetEquip). Mice were secured on

Type I IFN affects early and late immunity to HMPV

Hastings et al

102

their backs and were given constant anesthesia during the procedure via a nosecone

103

attachment. The right leg and thigh were shaved, and a thigh sensor was secured to the

104

right thigh of each mouse and covered with a dark cloth to reduce ambient light. A pulse

105

oximeter (MouseOx; Starr Life Sciences Corp.) was used to measure arterial O2

106

saturation, heart rate, pulse rate, pulse distension, and breath distension every 0.1

107

seconds (MouseOx software, version 4.0). Each mouse was monitored until sufficient

108

data was collected in which all parameters were successfully measured, and only these

109

data were used in analyses (1-2 minutes per mouse). Breath distension for each animal

110

was calculated by averaging all measurements for each mouse per condition per

111

genotype.

112 113

Flow Cytometry. Cells were isolated from lungs of infected animals as previously

114

described (15). Briefly, lungs were rinsed in R10 medium (RPMI-1640 [Mediatech] plus

115

10% FBS, 2 mM glutamine, 50 μg/ml gentamicin, 2.5 μg/ml amphotericin B, and 50 μM

116

β-mercaptoethanol [Gibco, Invitrogen]), and then minced with a scalpel and incubated

117

with 2 mg/ml collagenase A (Roche) and 20 μg/ml DNase (Roche) for 1 hour at 37°C.

118

Single-cell suspensions of digested lungs were obtained by pressing through a steel

119

screen (80 mesh) and then passing over a nylon cell strainer (70-μm pore size).

120

Erythrocytes were lysed using Red Blood Cell Lysis Buffer (Sigma-Aldrich). For labeling

121

of HMPV-specific CD8+ T cells, single cell suspensions of mouse lungs were incubated

122

with APC-labeled H2-Db tetramers (0.1-1 µg/mL), anti-CD8 (clone 53-6.7, BD

123

Biosciences), and anti-CD19 (clone 1D3, eBioscience) (15). Surface/tetramer staining

124

was performed for 1 hour at room temperature in PBS containing 2% FBS and 50 nM

Type I IFN affects early and late immunity to HMPV

Hastings et al

125

dasatinib (LC Laboratories) (33). For assessment of PD-L1 expression, cells were

126

stained with anti-EPCAM (clone G8.8, Biolegend), anti-CD11c (clone HL3, BD), anti-

127

CD11b (clone M1/70, Tonbo), and anti-MHC Class II (clone 2G9, eBioscience). Cells

128

were also stained for PD-1 (clone J43, BD Biosciences), TIM-3 (clone RMT3-23), LAG-3

129

(clone C9B7W), 2B4 (clone m2B4 (B6)458.1, Biolegend), PD-L1 (clone MIH5, BD), or

130

an isotype control antibody (hamster IgG2κ). Staining for HMPV-specific CD8+ T cells

131

was normalized to the binding of a cognate APC-labeled H2-Db tetramer loaded with

132

NP366 flu peptide to CD8+ T cells (typically 0.05-0.2% of CD8+ T cells). For all cell

133

populations FSC and SSC gating were used to obtain cells of appropriate size and

134

shape.

135 136

To identify HMPV-infected cell populations, homogenized lung cell suspensions were

137

stained with polyclonal anti-HMPV guinea pig sera (32) for 1 hour at room temperature

138

(RT), washed, and then stained with an anti-guinea pig FITC secondary antibody (Life

139

Technologies). Lung epithelial cells and CD11c+ high dendritic cells were identified

140

using forward/side scatter and stained for EPCAM and CD11c as described above.

141

Gates for HMPV-positive cell populations were set using uninfected mice and isotype

142

controls (BioLegend and BD). All flow cytometric data were collected using an LSRII or

143

Fortessa (BD Biosciences) and analyzed with FlowJo software (Tree Star).

144 145

Intracellular Cytokine Staining (ICS). Lung or spleen lymphocytes were isolated and

146

re-stimulated in vitro for 6 hours at 37°C with a non-specific peptide or the indicated

147

synthetic peptide (10µM final concentration) in the presence of an anti-CD107a antibody

Type I IFN affects early and late immunity to HMPV

Hastings et al

148

(clone 1D4B, BD Bioscience). The protein transport inhibitors brefeldin A and monensin

149

(BD Biosciences) were added for the final 4 hours of re-stimulation. Stimulation with

150

PMA/ionomycin (50ng/mL PMA and 2µg/mL ionomycin, Sigma) served as a positive

151

control. After re-stimulation, cells were stained for surface expression of CD3 (clone

152

145-2C11), CD8, and CD19, followed by fixation/permeabilization and staining for

153

intracellular IFN (clone XMG1.2) (BD Biosciences) and were analyzed using flow

154

cytometry. Background cytokine levels following re-stimulation were normalized to a

155

non-specific peptide.

156 157

Real-time RT-PCR. 200 µL of undiluted lung homogenate from infected or uninfected

158

IFNAR+/+ and IFNAR-/- mice was lysed with an equal volume of RLT lysis buffer

159

(Qiagen) and frozen at -20°C. Samples were thawed and RNA extracted using the

160

MagNA Pure LC Total Nucleic Acid Isolation Kit (Roche Applied Sciences) on a MagNA

161

Pure LC using the Total NA External Lysis protocol and stored at -80°C. Real-time RT-

162

PCR was performed in 25μL reaction mixtures containing 5μL of extracted RNA on an

163

ABI StepOnePlus Real-Time PCR System (Life Technologies/Applied Biosystems)

164

using the AgPath-ID One-Step RT-PCR kit (Life Technologies/Applied Biosystems). For

165

PD-L1, IFNγ, IL-2, IL-12, IL-4, and IL-10 gene expression, exon-spanning primers and

166

probes were used according to the manufacturer’s instructions (Applied

167

Biosystems/Ambion). All values were normalized to the housekeeping gene HPRT.

168

Cytokine transcript levels were low or undetectable in uninfected IFNAR-/- and WT mice.

169

Therefore, experimental samples are reported as fold change in HMPV-infected WT

Type I IFN affects early and late immunity to HMPV

Hastings et al

170

mice compared to HMPV-infected IFNAR-/- using the ΔΔCt method (34). Samples with

171

cycle threshold (Ct) values less than 40 were considered positive.

172 173

Statistical Analyses. Data analysis was performed using Prism v4.0 (GraphPad

174

Software). Groups were compared using unpaired t-test or one-way ANOVA with post-

175

hoc Tukey test for multiple comparisons. P