A limited role for regulatory T cells in post-ischemic neovascularization

Abstract Recently, it was demonstrated that arteriogenesis is enhanced in mice deficient in regulatory T cells (CD4+CD25+FoxP3+ T cell), which can suppress effector T cell responses. The present study investigates the effects of these regulatory T cells on arteriogenesis in more detail by either specific expanding or depleting regulatory T cells. Hind limb ischemia was induced by electro-coagulation of the femoral artery in mice. Regulatory T cells were either expanded by injecting mice with a complex of interleukin (IL)-2 with the IL-2 monoclonal antibody JES6–1, or depleted by anti-CD25 antibody or diphtheria toxin injections in DEREG mice (depletion of regulatory T cells). Blood flow restoration was monitored using laser Doppler perfusion imaging. Collateral arteries were visualized by immunohistochemistry. Regulatory T cell expansion led to a moderate though significant suppression of blood flow restoration after ischemia induction. Surprisingly, depletion of regulatory T cells resulted in minor increase on blood flow recovery. However, collateral and capillary densities in the post-ischemic skeletal muscle were significantly increased in DEREG mice depleted for regulatory T cells. The presence of regulatory T cells after ischemia induction when analysed in non-depleted DEREG mice could be demonstrated by green fluorescent protein staining only in lymph nodes in the ischemic area, and not in the ischemic muscle tissue. The current study demonstrates that, even under conditions of major changes in regulatory T cell content, the contribution of regulatory T cells to the regulation of the arteriogenic response is only moderate.


Introduction
Numerous studies have implicated that the immune system plays a major role in arteriogenesis [1,2]. A specific cell of the immune system, the regulatory T cell (CD4 ϩ CD25 ϩ FoxP3 ϩ T cell), is specialized in the suppression of effector T-cell pathogenic immune responses. Regulatory T cells control T-cell homeostasis [3,4] and are indispensable for balancing immune responses [5,6]. Accumulating evidence suggests an important role of regulatory T cells in the control of atherosclerotic lesion development and progression [7][8][9]. Recently, Zouggari et al. [10] assessed for the first time the role of regulatory T cells on blood flow recovery after ischemia induction in mouse models with a reduced regulatory T cell number and function [8,[10][11][12][13]. They reported a key role of regulatory T cells in post-ischemic neovascularization. Post-ischemic blood flow recovery was significantly enhanced in these mice up to 70% 21 days after ischemia induction as compared to slow recovery in control mice [10]. Because the mouse models used for depletion of regulatory T cells may also show effect on other T cell subtypes besides the regulatory T cell [14], a more detailed and specific analysis of the role of the regulatory T cell in arteriogenesis is required. The question remains how regulatory T cells regulate arteriogenesis and moreover whether regulatory T cell expansion has effects on arteriogenesis. The effect of expanding regulatory T cell number on arteriogenesis can be studied using the recently described method for in vivo expansion of regulatory T cells by Webster et al. [15] Injecting mice with a complex of interleukin (IL)-2 mixed with JES6-1 IL-2 monoclonal antibody (mAB) led to a marked increase in regulatory T cell number. This approach caused selective expansion of regulatory T cells with little or no change in other cells, making it superior to other attempts [10,15] to increase regulatory T cells [15]. Next to expansion of regulatory T cell number, depletion of regulatory T cells is highly informative. The standard approach is depletion of CD25 ϩ cells. But this approach has its shortcomings because anti-CD25 treatment does not result in a selectively depletion of regulatory T cells only [7,16], but also affects activated T cells [17], dendritic cells (DCs) [18] and natural killer (NK) cells [19]. Recently, a more powerful approach to fully deplete regulatory T cells was described by Lahl et al. [20]. They generated DEREG mice (depletion of regulatory T cells), expressing a diphtheria toxin (DT) receptor/green fluorescent protein (GFP) fusion protein under the control of the FoxP3 gene locus. DT injections in DEREG mice allowed for highly specific and up to 95-98% depletion of FoxP3 ϩ regulatory T cells at any desired time-point without disease development [20].
In the present study, the contribution of regulatory T cells in the arteriogenic response was studied by major modulations in regulatory T cell number in vivo. Regulatory T cell numbers in mice were either increased by injecting mice with IL-2 mixed with an IL-2 mAB or decreased by anti-CD25 antibody or DT injections in DEREG mice. Furthermore, using the GFP-labelled regulatory T cells in DEREG mice, the presence of regulatory T cells in the ischemic muscle tissue will be analysed in detail.

Experimental animals
Ten-week-old C57Bl6 mice (The Jackson Laboratory, Bar Harbor, ME, USA) were used (10 mice per group). Furthermore, DEREG mice (Leiden University Medical Center, Leiden, The Netherlands), aged 9-16 weeks, were used. Experiments were approved by the committee on animal welfare of our institute.

Regulatory T cell expansion
To increase regulatory T-cell number, C57Bl6 mice were injected with1 g IL-2 (PeproTech, Inc., Rocky Hill, NJ, USA) mixed with 5 g IL-2 mAB (clone JES6-1A12; R&D Systems, Minneapolis, MN, USA) [15]. Mice were injected at days 3, 2 and 1 before ischemia induction and two times a week after the surgical procedure. Expansion of regulatory T cell level was monitored by fluorescence-activated cell sorting (FACS) (mouse regulatory T-cell staining kit; eBioscience Inc., San Diego, CA, USA).

Regulatory T-cell depletion
Depletion of regulatory T cells is accomplished by treating C57Bl6 mice with CD25-depleting PC61 antibody [21]. A total of 100 g per mice was injected at day 1 before ischemia induction and repeated once per week after surgery.
DEREG mice express a DT receptor/GFP fusion protein under control of the FoxP3 gene locus, leading to selective depletion of FoxP3 ϩ regulatory T cells after injection of 1 g DT per mouse [22]. Depletion of regulatory T cells was monitored by FACS analyses (mouse regulatory T cell staining kit; eBioscience).

Surgical procedure
In the regulatory T-cell expansion experiments, blood flow was expected to be suppressed. In order to distillate an attenuated blood flow recovery compared to well-known fast recovery of control mice, single electro-coagulation of the femoral artery was performed. Mice depleted for regulatory T cells were expected to have accelerated blood flow recovery compared to controls that are well-known good responders. Therefore, in this situation, double electrocoagulation of both the femoral artery and iliac artery was performed because the double electro-coagulation resulted in a larger therapeutic window [23]. This enables a better analysis of enhancing effects on arteriogenesis.

Laser Doppler perfusion imaging (LDPI)
Blood flow was measured before ischemia induction, immediately after ischemia induction and 3, 7, 14, 21 and 28 days after surgery in the ischemic and non-ischemic paws, using LDPI (Moor Instruments, Millwey Axminster, Devon, UK). LDPI is performed with a beam from a 2 mW helium laser that sequentially scans tissue surface to a depth of a few hundred micrometres. The penetration depth of the laser beam thus allows measurements of superficial skin perfusion only. Therefore, perfusion in both paws (from the ankle to the base of toes) is obtained at baseline and serially over 4 weeks after hind limb ischemia induction. Paw perfusion functions as a resultant of limb perfusion recovery. Furthermore, to avoid measuring increased perfusion near the wound, it is preferable to exclusively select the paws, more distally, as the region of interest for analyses. Each animal served as its own control. Perfusion was expressed as a ratio of the ischemic to non-ischemic limb, as described previously [23].

Histological analysis
Animals were killed 14 and 28 days after ischemia induction and calf and adductor muscles (from ischemic and non-ischemic paw) were removed and fixed with 4% formaldehyde and paraffin embedded. Serial  For SMA labelling, tissue sections were incubated overnight with anti-␣SMA (mouse anti-human, dilution 1:800; DAKO, Glostrup, Denmark) without antigen retrieval. Labelling was followed by an HRP-conjugated secondary antibody (rabbit antimouse, dilution 1:300; DAKO).
For tracing of GFP ϩ regulatory T cells in DEREG mice, adductor muscle slides were incubated with anti-GFP (rabbit antimouse, dilution 1:4000; Invitrogen, Breda, the Netherlands) without antigen retrieval. After overnight incubation, labelling was followed by a biotin-conjugated secondary antibody (donkey anti-rabbit, dilution 1:300). As a positive control, a slide of GFP ϩ cardiac muscle tissue was used.
All sections were counterstained with haematoxylin. Isotype control antibodies were used as controls. Quantification of labelled tissue sections was performed with ImageJ (nine sections per mouse were analysed to obtain the mean per animal, 10 animals per group were measured).

Cells were labelled with fluorescein isothiocyanate (FITC) anti-CD4 (eBioscience) and Allophycocyanin (APC) anti-CD25. An intracellular FoxP3 staining was performed with PE anti-FoxP3 (eBioscience). Regulatory T cell labelling was analysed by flow cytometry on a LSRII FACS apparatus (Becton Dickinson B.V., Breda, the Netherlands) and
analysed with BD FACS DIVA software.

In vitro suppression assay
Peripheral lymph nodes were removed from three DEREG mice after repetitive DT injections during 3 weeks and in three DEREG mice without DT injections. The functional suppression assay was performed as described by Rausch et al. [24].

Statistical analysis
Results are expressed as mean Ϯ S.E.M. Comparisons between means were performed with an independent t-test. P-values Ͻ0.05 were considered statistically significant. All calculations were performed in SPSS 16.0. (Fig. 1A). Mice treated with injections of IL-2-mAB complex (n ϭ 10) showed significant attenuated blood flow recovery up to 20% at days 7 and 14 after ischemia compared to controls (n ϭ 10), indicating that an increase in regulatory T cell number was responsible for a suppression in blood flow recovery after hind limb ischemia (Fig. 1B). Furthermore, mice treated with IL-2-mAB complex showed a significant lower number of ␣-SMAexpressing collaterals in the post-ischemic adductor muscle as compared to controls (Fig. 1C). [25,26]. No effect of anti-CD25 antibody treatment (n ϭ 10) on blood flow recovery was observed as compared to controls (n ϭ 10) after single electro-coagulation of the artery (Fig. 2A). The moderate effects of the depletion on arteriogenesis could be obscured by the rapid response of these mice. Although the therapeutic window was increased after double electrocoagulation of both the femoral artery and iliac artery, CD25 depleting antibody-injections (n ϭ 10) did not result in any alteration of blood flow recovery as compared to controls (n ϭ 10) (Fig. 2B). FACS analyses showed 87% depletion of regulatory T cells directly after the surgical procedure. However, regulatory T cell levels returned to half of normal levels 1 week after ischemia, but at that moment, again anti-CD25 was injected into these mice (Fig. 2C). Thus with anti-CD25 antibody treatment, a minimal CD25 level was present all the time. So, a better approach to deplete regulatory T cells was needed.

Selective depletion of regulatory T cells led to moderate effects on arteriogenesis Depletion of FoxP3 Ϯ T cells in DEREG mice.
DEREG mice allowed depletion of FoxP3 ϩ regulatory T cells upon DT injections. Because two consecutive DT injections led to almost complete [20], but short (Ͻ6 days) depletion of regulatory T cells (Fig. 3A), DEREG mice were treated with repetitive DT injections. FACS analyses showed up to 98% depletion of FoxP3 ϩ regulatory T cells for 15 days after repetitive injections with DT (Fig. 3B). Analyses based on the GFP expression of regulatory T cells in DEREG mice, showed outgrowth of GFP -FoxP3 regulatory T cells within week 2 (Fig. 3B) as also was described before [24]. However, these GFP -FoxP3 ϩ cells were non-functional regulatory T cells. To demonstrate this also in the current study, an in vitro suppression assay was performed after 3 weeks of repetitive DT injections in DEREG mice as a functional analysis of these GFP -FoxP3 regulatory T cells. After isolation of the cells from the lymph nodes they were incubated with effector T cells and the effect on effector T cell proliferation was monitored. The GFP -FoxP3 ϩ cells had a marked reduced inhibitory effect on T cell proliferation (Fig. 3C).
Regulatory T cell-depleted DEREG mice (n ϭ 22) showed moderate improvement of blood flow recovery after ischemia induction as compared to control mice (n ϭ 20) (Fig. 3D). Nine days after ischemia induction, DEREG mice depleted for regulatory T cells showed 23% improvement of blood flow recovery. However, probably due to high variance, this was not statistically significant.
Repetitive DT injections did not have any effect on post-ischemic flow recovery, because flow recovery between C57Bl6 mice with (n ϭ 10) and without DT injections (n ϭ 10) showed no differences (Fig. S1). On the tissue level, regulatory T-cell-depleted DEREG mice showed a significant increase of SMA-expressing collateral density in the post-ischemic adductor muscle (Fig. 3E). Furthermore, these mice showed a significantly increase of the CD31 ϩ capillary density of the post-ischemic calf muscles as compared to controls (Fig. 3F). Selective and full depletion of regulatory T cells led to significant increases in capillary and collateral artery formation in the skeletal muscle tissue, but minor improvements of postischemic blood flow recovery. Fig. 1(A

Distribution of regulatory T cells to ischemic area.
The presence of regulatory T cells in the ischemic hind limb was assessed in the ischemic adductor muscle and draining lymph nodes after 3, 7 and 14 days after surgery. Non-depleted DEREG mice had GFP ϩ regulatory T cells, which we traced using immunohistochemistry and flow cytometry analyses. Regulatory T cells were not observed in the skeletal muscle tissue at any time-point examined with immunohistochemical stainings (Fig. 4A)

. Detailed views of vessel structures in the ischemic adductor muscle did not show any regulatory T cells near collaterals. Moreover, regulatory T cells were present in lymph nodes in the ischemic hind limb.
Compared to lymph nodes of the non-operated hind limb, significant more regulatory T cells were present in the lymph nodes of the ischemic paw (Fig. 4B).  [10]. For modulation of regulatory T cell content, more specific approaches were used for a major expansion or complete depletion of regulatory T cells. [15]. In the present study, expansion of regulatory T cells resulted in an attenuated blood flow recovery after ischemia induction. Up to 20% suppression of blood flow at day 14 after ischemia induction was observed. These results are in contrast to the major effects of application of this IL-2-mAB complex in various autoimmune disease mouse models, which resulted in the expansion of regulatory T cells in mice leading to a strong resistance to T cell mediated autoimmune disease [15].

Expansion of regulatory T cells was induced by injecting mice with IL-2 mixed with the JES6-1 IL-2 mAB [15]. Unlike most IL-2 mABs, injections of this IL-2-mAB complex led to selective proliferation of regulatory T cells with little or no change in other inflammatory cells making the IL-2-mAB complex superior to other interventions
In addition to regulatory T cell expansion, we studied the effects of depletion of regulatory T cells on arteriogenesis. Depletion of CD25 ϩ cells with injections of anti-CD25 antibody did not result in any alteration of post-ischemic blood flow recovery as compared to controls. Zouggari et al. [10] did report significant improved blood   [8,[10][11][12][13]. However, CD28 is the major B7-binding costimulatory  [14]. CD4 ϩ T-cell and CD8 ϩ T cell responses seemed to be disturbed in these mice too. The interpretation that the observed effects in CD28 -/mice and B7-1/2 -/mice are due to regulatory T cell modulation should be challenged.

receptor on T cells and is critical for T cell responses in vivo
To disentangle the mechanism of how regulatory T cells influence the arteriogenic response, the presence of regulatory T cells in the ischemic tissue was studied. No GFP ϩ regulatory T cells could be detected in post-ischemic adductor muscle in nondepleted DEREG mice with anti-GFP immunohistochemical staining 3, 7 and 14 days after surgery. The absence of regulatory T cells in the ischemic tissue can be related to the only moderate effects of regulatory T cells in arteriogenesis, especially because regulatory T cells are present in the lymph nodes in the ischemic hind limb area. These results suggest that regulatory T cells contribute to the arteriogenic response not directly by local induction of collateral artery formation, but rather in a paracrine manner by affecting other inflammatory cells in the ischemic muscle tissue from a distance.
An interesting observation in the current study is that even full depletion of regulatory T cells showed only moderate effects on arteriogenesis. This is in contrast to other studies on vascular remodelling as in atherosclerosis. There, the involvement of regulatory T cells in atherosclerotic lesion development was convincingly demonstrated using only selective depletion. Ait-Oufella et al. [8] reported significant effects on atherosclerosis already with minor modulations in regulatory T cell content. Moreover, vaccination of mice with FoxP3 transfected DCs resulted in a 27-30% decrease in regulatory T cells in blood and organs [7]. Although depletion is limited and far from complete, mice vaccinated against FoxP3 showed a significant 34% increase in plaque size as compared to controls [7].
In conclusion, the present study demonstrates that regulatory T cells do contribute to the regulation of post-ischemic neovascularization. However, only moderate effects on post-ischemic neovasculzarization were observed after major modulations of regulatory T cell number.