Abstract
Introduction
Phosphatase and tensin homolog (PTEN) is the principal negative regulator of the PI3-kinase pathway, members of which are essential for regulating natural killer (NK) cell activation and effector functions (Tassi et al Immunity 2007, Kim et al Blood 2007). However, the role of PTEN in NK cell biology remains unknown, in part hampered by the embryonic lethality of global PTEN loss. Thus, we hypothesized that disruption of PTEN would uniquely impact NK cell developmental and functional pathways. To evaluate whether PTEN was required for normal NK cell functions, we generated and evaluated a mouse model of NK cell-specific PTEN deficiency (Ncr1iCre knockin x PTENflox; PTENΔ/Δ).
Results
In contrast to T and B lymphocytes with conditional PTEN loss, we discovered that PTEN primarily acts to regulate NK cell distribution, but not their development. PTEN deletion resulted in a significant loss of NK cells (Fig. 1) in the bone marrow (48% reduction, p<0.001), spleen (56% reduction, p<0.001), and other lymphoid tissues, but markedly increased numbers within the peripheral blood (3.4-fold increase, p<0.001) and lung (3.8-fold increase, p<0.05). Surprisingly, we observed unaltered NK cell maturation (defined by CD27 and CD11b expression) within the peripheral organs of PTENΔ/Δ mice, indicating that PTEN operates to re-distribute NK cells without effects on terminal maturation.
To determine whether this aberrant localization could be attributed to dysregulated migration, we examined NK cell trafficking between lymphoid organs. PTEN-deficient NK cells egress more efficiently from the bone marrow and preferentially reside in sinusoidal compartments (mean sinusoidal fraction: 19% [control] vs. 34% [PTENΔ/Δ], p<0.01). Following combined in vivo inhibition of CXCL12 and VLA-4, the peripheral blood of control mice phenocopied the blood NK cell expansion observed in PTENΔ/Δ mice, suggesting that insensitivity to these retention signals is one contributing mechanism. Short-term, i.v. adoptive transfer of control and PTENΔ/Δ NK cells revealed that PTENΔ/Δ NK cells are strongly retained once in the blood (2.8-fold greater retention of blood PTENΔ/Δ vs. control NK cells, p<0.05) (Fig. 2). Furthermore, the loss of PTEN results in increased NK cell homeostatic proliferation in vivo (mean 3-day bone marrow NK cell BrdU incorporation: 20% [control] vs. 52% [PTENΔ/Δ], p=0.0008). Thus, PTEN regulates multiple factors that contribute to the normal steady-state distribution in the naïve mouse.
Given the inappropriate localization without PTEN, we further evaluated the NK cell requirement for PTEN during an anti-lymphoma response. PTENΔ/Δ mice challenged i.p. with the NK-sensitive RMA/S lymphoma had defective expansion of the peritoneal compartment (absolute peritoneal NK cells at 48 hours: 1.9x105 vs. 7.2x104, p=0.04). Furthermore, using an adoptive transfer model that requires NK cell trafficking to distal sites of lymphoma challenge, we found that PTEN-deficient NK cells had significant defects in their recruitment to localized tumors (18.4-fold vs 1.5-fold increase in recruited NK cells, p<0.001) (Fig. 3).
Conclusions
In this study, we describe the first report of NK-cell intrinsic PTEN loss in vivo. Collectively, our data suggests that unopposed PI3K signaling in NK cells dominantly affects key events responsible for appropriate cell trafficking and distribution, which is distinct from the role of PTEN in related lymphocyte lineages. These data implicate PTEN as a critical mediator of NK cell recruitment to sites of lymphoma and suggest that PTEN dysregulation, as in the case of PTEN loss-of-function mutations and hamartoma tumor syndromes, may result in defective NK cell-mediated immunity.
Figure 1 NK-specific PTEN-deficient mice re-distribute NK cells among NK cell compartments. Figure 1. NK-specific PTEN-deficient mice re-distribute NK cells among NK cell compartments.
Figure 2 Inappropriate NK cell retention in the blood contributes to the NK cell re-distribution observed in PTENΔ/Δ mice. Blood mononuclear cells were isolated, i.v. transferred into WT recipients and sacrificed after 16 hours. Figure 2. Inappropriate NK cell retention in the blood contributes to the NK cell re-distribution observed in PTENΔ/Δ mice. Blood mononuclear cells were isolated, i.v. transferred into WT recipients and sacrificed after 16 hours.
Figure 3 Intravenous adoptively transferred PTENΔ/Δ NK cells are unable to migrate to peritoneal lymphoma. Control or PTENΔ/Δ NK cells were i.v. transferred with RMA/S challenge i.p. into RAG2-/-γc-/- hosts. Peritoneal exudate cells (PECs) were isolated after 48 hours. Figure 3. Intravenous adoptively transferred PTENΔ/Δ NK cells are unable to migrate to peritoneal lymphoma. Control or PTENΔ/Δ NK cells were i.v. transferred with RMA/S challenge i.p. into RAG2-/-γc-/- hosts. Peritoneal exudate cells (PECs) were isolated after 48 hours.
Disclosures
No relevant conflicts of interest to declare.
Thrombin inhibits osteoclast differentiation through a non-proteolytic mechanism