Modulation of vascular remodeling : a role for the immune system, growth factors, and transcriptional regulation Seghers, L.

(1)

Seghers, L.

Citation

Seghers, L. (2011, November 30). Modulation of vascular remodeling : a role for the immune system, growth factors, and transcriptional regulation.

Retrieved from https://hdl.handle.net/1887/18166

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden

Downloaded from: https://hdl.handle.net/1887/18166

Note: To cite this publication please use the final published version (if applicable).

(2)

5

Differences in NK gene complex between C57BL/6 and BALB/c mice determine post

ischemic blood flow recovery

Leonard Seghers^1,2,3; Antonius J.N.M. Bastiaansen¹, Vincent van Weel¹, René E.M. Toes⁴, Margreet R. de Vries^1,3, Alwine A. Hellingman^1,3, Jeroen van Bergen⁵,

Victor W.M. van Hinsbergh², Paul H.A. Quax^1,3

1 Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands

2 Department of Physiology, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam, The Netherlands

3 Einthoven Laboratory for Experimental Vascular Medicine, Leiden University Medical Center, Leiden, The Netherlands

4 Department of Rheumatology, Leiden University Medical Center, Leiden, The Netherlands

5 Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands

Submitted for publication

(3)

R1R2 R3R4 R5R6 R7R8 R10R9 R12R11 R13R14 R15R16 R17R18 R19R20 R21R22 R23R24 R25R26 R27R28 R29R30 R31R32 R33R34 R35R36 R37R38 R39

Abstract

Aims: Inflammatory cells such as monocytes and T-lymphocytes contribute to collateral artery development (arteriogenesis). Recently, Natural Killer (NK) cells were shown to stimulate arteriogenesis. After femoral artery ligation, C57BL/6 mice display rapid and NK-cell dependent induction of arteriogenesis and blood flow recovery, whereas BALB/c mice recover poorly. Here, we studied the consequences of genetic differences in NK cell gene locus between C57BL/6 and BALB/c on blood flow recovery in a murine model of hind limb ischemia.

Methods and Results: To test whether differences in the NK gene complex between C57BL/6 and BALB/c contribute to this difference, congenic mice carrying the C57BL/6 NK gene complex (NKC) on a BALB/c genetic background (BALB.B6-CMV1^r) were used. 28 days after femoral artery ligation, blood flow recovery in BALB.B6-CMV1^r mice (95 %) was similar to the recovery in C57BL/6 mice (102%) and significantly improved compared to BALB/c (maximally 47%). Furthermore, in BALB.B6-CMV1^r mice an increase in collateral arteries in the adductor muscle and in capillaries in the ischemic calf muscle was observed as determined by immunohistochemistry. In line with a role for NK cells, NK cell depletion in BALB.B6-CMV1^r significantly impaired blood flow recovery. Additionally, in vitro NK cell stimulation assays revealed increased NK cell responsiveness in C57BL/6 and BALB.B6-CMV1^r mice compared to BALB/c mice.

Conclusions: These data underscore the importance of NK cells in arteriogenesis, and in particular show that genes located in the C57BL/6 NKC on chromosome 6 contribute to profound arteriogenesis, possibly by increasing NK cell responsiveness.

(4)

A role for the C57BL/6 NKC in blood flow recovery

Introduction

Two important mechanisms for postnatal blood vessel growth are angiogenesis, i.e.

the sprouting of new capillaries in ischemic tissue, a process mainly driven by hypoxia induced growth factors as VEGF, and arteriogenesis, the formation of new collateral arteries from smaller pre-existing collateral arteries, a process mainly driven by shear stress and inflammatory factors [1, 2]. Evidence for involvement of the immune system in arteriogenesis is increasing. Next to the extensively described role of monocytes [3-7]

also a role for CD4⁺ T-cells [8, 9] CD8⁺ T-cells [10] and even regulatory T-cells [11, 12] in arteriogenesis is demonstrated. In addition, recently NK cells [9] were demonstrated to be modulators of arteriogenesis.

Much of the research in this area is based on the differences in vascular remodeling response in between C57BL/6 and BALB/c mice [9, 13-15] two mouse strains with a striking differences in immune response. Differences in lymphocyte accumulation around collateral arteries were observed between C57BL/6 and BALB/c mice. C57BL/6 mice accumulated significantly more lymphocytes than BALB/c mice at different time points after femoral artery ligation [9]. Interestingly, these two mouse strains also differ in their immune bias, i.e. C57BL/6 mice are Th1 responders and BALB/c mice are T helper 2 (Th2) responders. A Th1 response is associated with more accelerated vascular remodelling, whereas a Th2 response is known to be more cultivated and is therefore less contributing to vascular remodelling [16].

Not only the T-lymphocyte response is different between these two mouse strains, also important differences in a genetic locus for NK cell receptors exist between these C57BL/6 and BALB/c strains [17]. And although NK cells [9] were recently proven to be modulators of arteriogenesis, the exact mechanism of how NK cell contribute to arteriogenesis is not clear yet. We hypothesize that these difference in NK cell receptors especially those in the NKC locus may explain at least part of the observed differences in the arteriogenic response.

The Natural Killer gene Complex (NKC) [17] is a gene locus on chromosome 6 encoding multiple activating and inhibitory NK cell receptors and holds genetic differences between C57BL/6 and BALB/c mice. The NKC codes for receptors such as NKG2D, Nkrp1c (NK1.1), CD94/NGK2, and the highly polymorphic Ly49 receptor family [18, 19]. Compared to C57BL/6, BALB/c mice lack a 200kb region containing encoding members of the Ly49 receptor family. The C57BL/6 NKC region possesses 12 Ly49 genes, whereas the BALB/c strain has only 7 Ly49 genes [17]. Scalzo and colleagues studied the genes of the NKC in relation to the resistance or sensitivity to the murine cytomegalovirus (mCMV) in C57BL/6 and BALB/c mice, respectively. They generated a congenic BALB/c mouse strain that contains the entire C57BL/6 NKC [20]. These mice

(5)

therefore express the entire C57BL/6 NK receptor repertoire on their NK cells and since they exhibit a level of resistance to mCMV comparable to that of C57BL/6 mice, they were called BALB.B6-CMV1^r (CMV1^r) [21].

In the present study we use these CMV1^r mice to investigate the contribution of the C57BL/6 NKC to arteriogenesis and evaluate whether the presence of the C57BL/6 NKC alleles in BALB/c mice could improve the arteriogenic phenotype of BALB/c.

Methods Mice

All animal experiments were approved by the committee on animal welfare of the Netherlands Organization for Applied Scientific Research (TNO, Leiden, The Netherlands) and LUMC (Leiden, The Netherlands). All animal experiments conform to the directive 86/609/EU of the European Parliament. For all experiments male mice were used, aged 12-16 weeks. C57BL/6 mice (Charles River, L’Arbresle Cedex, France) and BALB/c mice (Harlan, Horst, The Netherlands) were purchased. BALBc.B6-CMV1^r breeding couples were kindly provided by Dr. W. M. Yokoyama (St. Louis, USA). All animals were fed regular chow diet (Sniff Spezialdiäten GMBH, Soest, Germany).

Surgical procedure

Before surgery, mice were anesthetized by an intra-peritoneal injection with Midazolam (5 mg/kg, Roche, Woerden, The Netherlands), Medetomidine (0.5 mg/kg, Orion, Espoo, Finland) and Fentanyl (0.05 mg/kg, Janssen Pharmaceutics, Beerse, Belgium). After the procedure mice were antagonized with a subcutaneous injection of Atipamezol (2.5 mg/kg, Orion) and Flumazenil (0.5 mg/kg Roche). Surgical tolerance and adequacy of anesthesia were both assessed by the pedal withdrawal reflex.

Hindlimb ischemia was induced by coagulation of the left femoral artery proximal to the bifurcation of the deep and superficial femoral artery [22]. Repeated blood flow measurements over the paws were obtained at baseline, immediately after femoral artery coagulation, and serially over 4 weeks by laser-doppler-perfusion-imaging (LDPI; Moor Ltd, Axminster, United Kingdom). Perfusion is expressed as a ratio of left (ischemic) to right (non-ischemic) limb. During LDPI follow-up measurements mice were anesthetized by an intra-peritoneal injection with Midazolam (5 mg/kg, Roche) and Medetomidine (0.5 mg/kg, Orion) and were afterwards antagonized with a subcutaneous injection of Atipamezol (2.5 mg/kg, Orion) and Flumazenil (0.5 mg/kg Roche).

Directly after the last LDPI measurement mice were sacrificed by cervical dislocation when still under general anesthesia; Midazolam (5 mg/kg, Roche), Medetomidine

(6)

(0.5 mg/kg, Orion) and Fentanyl (0.05 mg/kg, Janssen Pharmaceutics). Subsequently adductor thigh muscles and gastrocnemius muscles were dissected from both hind limbs and fixed overnight in 4% (v/v) formaldehyde, dehydrated and paraffin embedded. Body temperature of mice was maintained during surgery and LDPI follow-up measurements by placing the animals on a heating pad.

NK cell depletion

For depletion of NK1.1+ cells, CMV1^r mice received intraperitoneal injections of 200µg of anti-NK1.1 antibody (PK136), an isotype-matched mouse IgG2a control antibody or PBS control vehicle 5 days, 3 days and 1 day before surgery, and twice a week after surgery. Depletion was confirmed in peripheral blood by fluorescence-activated cell sorting (FACS) -analysis using antibodies against NK1.1 and CD3 (BD Biosciences, San Jose, USA) just before surgery and 14 days after surgery.

Immunohistochemistry

5 µm thick paraffin-embedded cross sections of gastrocnemius or adductor muscle were stained using antibodies for PECAM-1 (CD31) (BD Biosciences, San Jose, USA) or smooth muscle α-actin (clone 1A4; Dako, Glostrup, Denmark), respectively. Stainings for both CD31 and smooth muscle α-actin were quantified from sections photographed randomly (3 images per section) using image analysis (Qwin, Leica, Wetzlar, Germany).

Angiography

To study collateral vessel development, post-mortem angiography of both hind limbs were made, using polyacrylamide-bismuth contrast mixture (Sigma Aldrich Chemie, Zwijndrecht, The Netherlands) at 7 days after femoral artery occlusion, as described [23]. Briefly, mice were anesthetized by an intra-peritoneal injection with Midazolam (5 mg/kg, Roche), Medetomidine (0.5 mg/kg, Orion) and Fentanyl (0.05 mg/kg). Surgical tolerance and adequacy of anesthesia were both assessed by the pedal withdrawal reflex. After a median laparotomy papaverin (2 mg/ml, Centrafarm, Etten-Leur, The Netherlands) was injected into the abdominal aorta to induce vasodilation, followed by contrast medium at constant pressure (100 mmHg). Subsequent to contrast medium infusion, mice were sacrificed by cervical dislocation. Post-mortem angiographic images were acquired by röntgenographic exposure using a Faxitron X-ray machine (Faxitron, Lincolnshire, United States of America). Detailed photographs of the hindlimb region on the radiographs were taken with a Nikon digital camera. Grading of collateral filling was performed in a single blinded fashion by two independent observers and was based on the Rentrop score. Grading was as follows: 0 = no filling of collaterals, 1 = filling of collaterals only, 2 = partial filling of distal femoral artery, 3 = complete filling of distal femoral artery.

(7)

NK cell activation assay

1 x 10⁷ splenocytes were cultured in 24-wells plates coated with either anti-NKp46 antibody (R&D Systems Europe, Oxon, United Kingdom), isotype-matched control (goat IgG, R&D Systems Europe, Oxon, United Kingdom) antibody (both at 5µg / well), or in medium containing 0.1 µg/ ml Phorbol 12-Myristate 13-Acetate (PMA) and 0.5 µg/ ml ionomycin (Sigma-Aldrich, St. Louis, United States of America). Intracellular staining for interferon-gamma (IFN-γ, BD Biosciences, San Jose, USA) was performed after a 4-h incubation at 37ºC in the presence of Brefeldin A. Percentages of IFN-γ producing NK cells were obtained by gating on CD3ε-, NK1.1+ or CD3ε-, DX5+ cells during FACS analysis on BD FACS LSR II using BD FACS Diva software v6.0 (BD Biosciences, San Jose, USA). For each mouse, the percentage of NK cells producing IFN-γ in response to NKp46 incubation was normalized for the response to PMA/ionomycin as follows: (NKp46 – control antibody)/(PMA/iono – control antibody) x 100%.

Statistical analysis

Results are expressed as mean±SEM. Comparisons between medians were performed using the Mann Whitney or Wilcoxon rank test as appropriate. Ordinal scores (e.g.

angiography) were compared using Pearson Chi-Square test. A p-value <0.05 was considered statistically significant.

Results

The C57BL/6 NK gene complex rescues the poor arteriogenic response of BALB/c mice To determine whether the presence of C57BL/6 NKC alleles in BALB/c could improve the arteriogenic phenotype of BALB/c, we studied arteriogenesis in BALB.B6-CMV1^r mice, which express the entire C57BL/6 NKC on a BALB/c background [20].

After unilateral femoral artery ligation [23] CMV1^r mice showed a strong improvement and a near complete rescue of the poorly responding BALB/c phenotype towards the rapid C57BL/6 response (Figure 1). CMV1^r mice closely followed the C57BL/6 blood flow recovery up to 7 days after ligation, reaching a maximum paw perfusion recovery ratio of 0.95±0.08 at 28 days post ligation versus 1.02±0.05 in C57BL/6 mice (difference not significant). The profound blood flow recovery in CMV1^r mice was significantly different from BALB/c mice (CMV1^r vs BALB/c p<0.01 at all time points after ligation).

In BALB/c mice a maximum paw perfusion recovery ratio of only 0.47±0.06 was reached after 28 days. This was also significantly different from C57BL/6 (p-values; C57BL/6 vs BALB/c p<0.01 at all time points after ligation). Taken into consideration that three BALB/c mice had to be terminated because of severe necrosis, this difference may even be larger.

(8)

Figure 1. Ischemic / non-ischemic paw perfusion ratios of C57BL/6 ♦, (n=14), CMV1^r ■, (n=14;

originally 16, but 2 mice died due to an unrelated cause of death and were therefore not included in the analysis) and BALB/c mice ▲, (n=13; originally 16 but three mice were terminated because of severe necrosis during follow-up and were therefore not included in the analysis). * C57BL/6 vs.

BALB/c (* p<0.01) and † CMV1^r compared to BALB/c († p<0.01).

To examine the presence and size of collateral arteries, post-mortem angiograms were made and collateral artery diameters were measured 7 days after femoral artery ligation.

Quantification of these post-mortem angiograms, using Rentrop scoring, revealed that CMV1^r mice very much resembled C57BL/6 mice in the presence of collateral arteries (Rentrop scores 2.0±0.35 and 2.75±0.16 respectively). BALB/c mice had fewer collaterals (Rentrop score 1.0±0.35, p<0.05), which was significantly different from C57BL/6 mice (Figure 2A-D). In addition, collateral artery diameter was analyzed by smooth muscle α-actin staining on adductor thigh muscles. CMV1^r mice had an average collateral artery diameter of 41.3±3.7 µm, similar to C57BL/6 mice (45±5.0 µm), while in BALB/c mice significantly smaller collateral artery diameters were found when compared to both C57BL/6 and CMV1^r mice (30.5±2.8 µm; p-values <0.05) (Figure 2E).

In summary, CMV1^r mice, which are congenic for the C57BL/6 NKC, demonstrate a strong improvement of blood flow recovery from the poor response of BALB/c mice towards the rapid response of C57BL/6 mice, which correlates with improved collateral formation. This suggests that genes located in the C57BL/6 NKC contribute to arteriogenesis.

Figure 1

0 3 7 14 21 28

days after femoral artery ligation 0

0.2 0.4 0.6 0.8 1 1.2 1.4

ischemia/ non-ischemicpawperfusion ratio

C57BL/6 CMV1r BALB/c

*

†

*

†

*

†

*

^†

*

†

*

^†

(9)

Figure 2. Post mortem angiographic images of upper hindlimb 7 days after femoral artery ligation:

A: C57BL/6, B: CMV1^r, C: BALB/c mice. White arrows indicate collateral arteries. D. Quantification of angiographic images (rentrop score) of C57BL/6 (n=8), CMV1^r (n=9) and BALB/c mice (n=5). (*

p<0.05) E. Diameters (µm) of collateral arteries in left adductor thigh muscles (mean±SEM) 7 days after ligation (in n=5 to 9). (* p<0.05)

Improved angiogenesis in CMV1^rmice

To investigate the effects of the C57BL/6 NKC on angiogenesis, the capillary density in calf muscles was quantified by an endothelial specific PECAM-1 (CD31) staining 7 days after femoral artery ligation. In C57BL/6 mice a 1.71±0.3 fold increase in capillary density in challenged (left) calf muscles versus unchallenged calf (right) muscles was observed (Fig. 3A). CMV1^r mice displayed a 1.34±0.2 fold increase in capillary density. Both these increases were significantly greater than those in BALB/c mice (p<0.05), which rather



 















 ^





^



























 ^





  ^^

 





 





















 

^



(10)

showed a decrease in capillary density (0.57±0.1 fold) in left versus right calf muscles (Figure 3B-C).

In conclusion, the angiogenic response in CMV1^r mice also mimicked that of C57BL/6 mice, and was significantly better than that of BALB/c mice.

Figure 3. Capillary density of calf muscles after femoral artery ligation (mean±SEM).

A. Ratio of capillary number per mm² (PECAM-1 staining 7 days after femoral artery ligation) of left (challenged) versus right calf muscles (n=5 to 9). (* p<0.05) B. Quantification of area per capillary in left versus right calf muscles. C. Representative microscopic images (magnification 200x) of PECAM-1 staining in left and right calf muscles from C57BL/6, CMV1^r and BALB/c mice.

Please note that the capillary structures in left calf muscles of BALB/c mice were altered and were therefore difficult to quantify because of the more heterogeneous morphology and size differences.

See color figure on page 225.

0 10 20 30 40 50 60

area / capillary(um2)

LC RC

C57BL/6CMV1^rBALB/c

B

0C57BL/6CMV1^rBALB/c 0.5

1 1.5 2 2.5

LeftvsRight ratio of # capillaries/mm2

*

* A Figure 3

L R

C

C57BL/6

CMV1^r

BALB/c

(11)

NK Cell depletion in CMV1^r mice impairs the arteriogenic response

To test whether the rescue of the arteriogenic phenotype in CMV1^r mice, which carry the C57BL/6 NKC, was indeed mediated by NK cells, NK cells were depleted in these mice. Therefore, CMV1^r mice were injected intraperitoneally with anti-NK1.1, isotype- matched control antibody (IgG) or PBS 5, 3 and 1 days before surgery. Flowcytometric (FACS) analysis of peripheral blood samples on the day of surgery showed that NK- cell depletion had been successful (p<0.01; Figure 4A). The depletion was sustained as revealed by FACS analysis 14 days after surgery (data not shown).

NK cell depletion in CMV1^r mice resulted in significantly impaired blood flow recovery from day 7 until day 28 when compared to controls (Figure 4B). Blood flow recovery was inhibited by approximately 35% at day 7, approximately 37% on day 14, approximately 26% at day 21 and reached a maximum paw perfusion ratio of 0.64±0.08 at 28 days.

Both the IgG and PBS control group showed a similarly rapid blood flow recovery by reaching paw perfusion ratios of 0.75±0.07 and 0.87±0.08, respectively at 28 days. These data demonstrate that NK cells contribute to the improved blood flow recovery in CMV1^r mice.

NK cell responsiveness is improved in BALB.B6-CMV1r compared to BALB/c

The presence of the C57BL/6 NKC in CMV1^r mice results in the development of NK cells expressing a different receptor repertoire compared to BALB/c. In recent years, it has become clear that the number and strength of inhibitory interactions between Ly49 receptors and MHC class I determine NK cell responsiveness [24, 25]. We therefore hypothesized that the differences in responsiveness between CMV1^r and BALB/c NK cells might correlate with the different arteriogenic phenotypes of these mice.

To investigate this, we employed a commonly used assay to measure NK cell responsiveness, measuring intracellular IFN-γ in NK cells upon crosslinking of activating receptors [24, 25]. To this end, splenocytes of C57BL/6 (n=6), BALB/c (n=6) and CMV1^r mice (n=6) were stimulated in vitro with plate-bound antibodies to NKp46, an activating NK receptor present in equal densities on all splenic NK cells in the three mouse strains.

An isotype-matched control antibody and Phorbol-12-Myristate-13-Acetate (PMA) and ionomycin served as negative and positive control stimuli, respectively. NK cell responses to the different stimuli were analyzed by FACS, measuring the frequencies of IFN-γ positive cells within the NK cell population after 5 hours of stimulation.

The response of CMV1^r NK cells to NKp46 stimulation was quite similar to C57BL/6 NK cells, reaching an average 24.2±3.1% and 31.1±10.7% of the maximum response respectively (Figure 5). This IFN-γ response was significantly different from BALB/c NK cells (p<0.01) that responded poorly reaching only 2.7±0.9% of the maximum IFN-γ response after NKp46 stimulation. NK cell IFN-γ response to PMA / ionomycin stimulation was not different between strains.

(12)

95 These data demonstrate that CMV1^r NK cells have a significantly improved responsiveness when compared to BALB/c NK cells in reaction to a NK receptor specific activating stimulus, and that this difference is likely explained by NK cell receptors encoded by the C57BL/6 NKC.

Figure 4. A. Representative of flow cytometric analysis for NK cells in peripheral blood from NK1.1 depleted CMV1^r mice, IgG treated controls and PBS treated controls at the moment of surgery;

average NK cell percentages of lymphocyte population: NK1.1 0.05±0.00%, IgG 2.75±0.27% and PBS 2.39±0.32%. Gated for CD3ε- NK1.1+ cells.

B. Ischemic / non-ischemic paw perfusion ratios of NK1.1 depleted CMV1^r mice ♦, (n=9; originally 10, but 1 mice died due to an unrelated cause of death and was therefore not included in the analysis), IgG treated controls ■, (n=9; originally 10, but 1 mice died due to an unrelated cause of death and was therefore not included in the analysis) and PBS treated controls ▲, (n=8; originally 10, but 2 mice died due to an unrelated cause of death and were therefore not included in the analysis).

* NK1.1 vs. PBS p-value <0.05, # NK1.1 vs. IgG p-value <0.05

Figure 4

NK1.1 PBSIgG 0

0.2 0.4 0.6 0.8 1.0 1.2

ischemic/ non-ischemic pawperfusionratio

* *

^#

B

NK1.1

IgG ctrl

PBS

CD3ε

NK1.1

A

NK cells NK cells NK cells

Figure 4

NK1.1 PBSIgG 0

0.2 0.4 0.6 0.8 1.0 1.2

ischemic/ non-ischemic pawperfusionratio

days after femoral artery ligation

0 3 7 14 21 28

* *

^#

B

NK1.1

IgG ctrl

PBS

CD3ε

NK1.1

A

NK cells NK cells NK cells

(13)

Figure 5. A. NK cells expressing DX5 (CD49b) and intracellular IFN-γ (previously gated on CD3ε-).

Flow cytometry of IFN-γ expression in NK cells from C57BL/6, BALB/c and CMV1^r mice after stimulation with plate bound NKp46, goat IgG isotype control or with PMA / ionomycin.

Representative of two independent experiments (n=3 mice per strain per experiment). B. NK cell responsiveness to specific NK receptor mediated activation by NKp46 in C57BL/6, BALB/c and CMV1^r NK cells expressed as average percentage of the maximum IFN-γ response that was achieved by non-specific PMA/ionomycin stimulation (data pooled from two independent experiments n=3 mice per strain per experiment).

Responsiveness to PMA/ionomycin non-specific stimulation was not different between the tested strains. Induction of NK cell IFN-γ production by NKp46 was significant when compared to

‘background’ IFN-γ production of NK cells incubated with isotype matched control (goat IgG) in all tested mousestrains (data not shown).

* p-value <0.01.

Discussion

In the present study we demonstrate that differences in blood flow recovery between C57BL/6 and BALB/c can in part be explained by differences in Natural Killer gene complex between these two strains. Presence of the C57BL/6 NK gene complex (NKC) in congenic BALB/c mice (CMV1^r) led to a substantial improvement of their naturally poor arteriogenic phenotype towards the rapid arteriogenic response of C57BL/6 mice.

Moreover, depletion of NK cells in CMV1^r mice resulted in significantly impaired blood flow recovery. Furthermore, in vitro NK cell stimulation demonstrated significantly greater responsiveness of NK cells from CMV1^r and C57BL/6 mice compared to BALB/c mice, thereby suggesting a possible mechanism for how the NKC differences contribute

IFN-γ

isotype

BALB/c

C57BL/6 CMV1^r

DX5 (CD49b) NKp46PMA / iono

A. B.

0 10 20 30 40 50

C57BL/6 BALB/c CMV1^r

% of maximum IFN-γ response

1.17%

8.56%

22.55%

0.52%

0.82%

5.04%

1.15%

2.48%

7.45%

*

(14)

to different blood flow recovery. Together, these data indicate that genes located in the C57BL/6 NKC, that are normally absent in BALB/c mice, can contribute to a rapid arteriogenic response in a NK-cell dependent fashion.

A number of studies have indicated a modulating role for several leukocyte subsets in arteriogenesis. A role for the immune system in arteriogenesis was thereby established and differences in immune bias result in differences in arteriogenesis, such as observed between C57BL/6 and BALB/c mice. Between these mice, T lymphocyte response is different, but also differences in the Natural Killer gene complex have been reported.

Based on these NKC differences, NK cells in C57BL/6 mice might react differently when compared to NK cells from BALB/c mice. Recently, C57BL/6 NK cells were indicated to play a contributory role in arteriogenesis. In the present study we expand these observations and demonstrate that the differences in Natural Killer cell gene locus between C57BL/6 and BALB/c mice can explain differences in blood flow recovery between these two mouse strains. For this we used a BALB/c mouse congenic for the entire C57BL/6 NKC.

The improved blood flow recovery in the congenic CMV1^r mice was accompanied by a hallmark of arteriogenesis, the diameter of the collaterals were increased in CMV1^r mice compared to BALB/c mice. Moreover, angiogenesis in the calf muscles was augmented.

Although a direct effect on the capillary endothelium resulting in increased angiogenesis cannot be excluded, more likely this is mediated via indirect effects of improved perfusion and as a consequence reduced deterioration of the muscular tissue. In BALB/c mice the attenuated blood flow recovery resulted in impaired angiogenesis as these mice experienced more ischemia by poor arteriogenesis, a phenomenon that was also observed by Chalothorn et al [13].

The improved arteriogenesis in CMV1^r mice is modulated by a NK cell related factor, since NK cell depletion in CMV1^r mice resulted in significant impaired blood flow recovery, similar to what has been shown in C57BL/6 mice. However, the NK cell depletion did never result in a drop in arteriogenesis to the poor level of BALB/c mice. This suggests that not only NK cell related factors regulate arteriogenesis, but that also non-NK cell related factors contributed to the profound blood flow recovery in C57BL/6 and CMV1^r mice. One such factor could be the better developed pre-existing collateral artery network in C57BL/6 mice as compared to BALB/c[13, 15]. However, pre-existing collateral artery networks in CMV1^r and BALB/c mice are reportedly similar [26] and therefore this does not explain the observed differences in blood flow recovery between these strains.

Another important factor, that was recently reported, could be the quantitative trait locus on chromosome 7 (Lsq-1) that was associated with reduced tissue necrosis and better perfusion recovery after hind limb ischemia in C57BL/6 compared to BALB/c mice [14].

We found that the associated C57BL/6 SNP (marker rs13479513) variant was present in

(15)

our CMV1^r mice. Therefore, part of the non-NK cell related improvement in blood flow recovery in CMV1^r mice may be explained by the presence of this C57BL/6 SNP.

NK cell depletion in CMV1^r mice resulted in impaired blood flow recovery indicating that at least part of the improved arteriogenesis in these mice can be explained by a NK cell related factor that is present within the C57BL/6 NKC, but absent from the BALB/c NKC. Compared to the C57BL/6 NKC, the BALB/c NKC contains fewer functional genes encoding activating and inhibitory NK cell receptors[17, 27] and this will affect NK cell functionality. NK cell responsiveness is tuned by the signaling input from activating [28]

and inhibitory [24, 25] NK cell receptors under steady-state conditions, as a result of processes referred to as NK cell ‘arming’, ‘licensing’ or ‘education’ [29].

Little is known about the exact role of the NK cell in arteriogenesis, but it is tempting to suggest that the NKC differences induce a different NK cell responsiveness. Especially since responsiveness is determined largely by interactions of inhibitory Ly49 and CD94/

NKG2A receptors, encoded by the NKC on chromosome 6, with autologous Major Histocompatibility Complex (MHC) class I molecules, encoded by the H-2 locus on chromosome 17. Rather counterintuitively, strong steady-state inhibition translates into high NK cell responsiveness, while NK cells lacking inhibitory receptors binding self- MHC class I are hyporesponsive [24, 25, 29]. The main difference between the C57BL/6 and the BALB/c NKC is the presence of a 200 kb Ly49 cluster encoding 4 additional functional Ly49 genes in the C57BL/6 NKC. We speculate that the additional receptors present in CMV1^r mice explain the greater NK cell responsiveness, as we demonstrated in the in vitro NK cell stimulation experiment, and therefore contribute to the improved arteriogenesis in CMV1^r compared to BALB/c mice, which are H-2 identical. We are currently investigating this issue using multiple intra-NKC recombinant and congenic mouse strains [30].

Humans do not possess functional Ly49 genes, but use instead the killer Ig-like receptor family (KIR) [31]. Like the Ly49 receptor family, the KIR family has a great genetic diversity [32], and impact of this divergence on collateral formation in humans is not known yet. This could provide information about the extent of the arteriogenic response in different patients. For example patients with coronary artery disease demonstrated a different composition of their NK cell compartment in peripheral blood when compared to age-matched healthy controls [33].

In summary, we showed that genetic differences in the Natural Killer gene complex between C57BL/6 and BALB/c relate with a difference in the arteriogenic response after induction of hind limb ischemia. Our data show that a crucial NK cell related factor within the C57BL/6 NKC contributes to profound arteriogenesis and thus further underline the importance of NK cells in arteriogenesis. Furthermore, the reduced NK responsiveness in BALB/c mice, but not in C57BL/6 and CMV1^r mice, provides a mechanism for how

(16)

the C57BL/6 NKC contributes to the differences in blood flow recovery via altered NK activation capacities.

Sources of funding

This study was sponsored by the EU European Vascular Genomics Network (LSHM- CT-2003-503254), the Dutch Program for Tissue Engineering (DPTE) grant VGT6747, and by the Netherlands Initiative for Regenerative Medicine. JvB was supported by the Landsteiner Foundation for Blood transfusion Research (grant 0515).

Disclosures

Conflict of Interest: none declared.

(17)

References

1. Carmeliet P. Mechanisms of angiogenesis and arteriogenesis. Nat Med 2000;6:389-395.

2. Heil M, Schaper W. Insights into pathways of arteriogenesis. Curr Pharm Biotechnol 2007;8:35- 42.

3. Bergmann CE, Hoefer IE, Meder B, Roth H, van Royen N, Breit SM, et al. Arteriogenesis depends on circulating monocytes and macrophage accumulation and is severely depressed in op/op mice. J Leukoc Biol 2006;80:59-65.

4. Cochain C, Rodero MP, Vilar J, Recalde A, Richart AL, Loinard C, et al. Regulation of monocyte subset systemic levels by distinct chemokine receptors controls post-ischaemic neovascularization. Cardiovasc Res 2010;88:186-195.

5. Hellingman AA, Zwaginga JJ, van Beem RT, Hamming JF, Fibbe WE, Quax PH, et al. T-cell- pre-stimulated monocytes promote neovascularisation in a murine hind limb ischaemia model. Eur J Vasc Endovasc Surg 2011;41:418-428.

6. Ito WD, Arras M, Winkler B, Scholz D, Schaper J, Schaper W. Monocyte chemotactic protein-1 increases collateral and peripheral conductance after femoral artery occlusion.

Circ Res 1997;80:829-837.

7. Schirmer SH, van Royen N. Stimulation of collateral artery growth: a potential treatment for peripheral artery disease. Expert Rev Cardiovasc Ther 2004;2:581-588.

8. Stabile E, Burnett MS, Watkins C, Kinnaird T, Bachis A, la Sala A., et al. Impaired arteriogenic response to acute hindlimb ischemia in CD4-knockout mice. Circulation 2003;108:205-210.

9. van Weel V, Toes RE, Seghers L, Deckers MM, de Vries MR, Eilers PH, et al. Natural killer cells and CD4+ T-cells modulate collateral artery development. Arterioscler Thromb Vasc Biol 2007;27:2310-2318.

10. Stabile E, Kinnaird T, la Sala A., Hanson SK, Watkins C, Campia U, et al. CD8+ T lymphocytes regulate the arteriogenic response to ischemia by infiltrating the site of collateral vessel development and recruiting CD4+ mononuclear cells through the expression of interleukin-16. Circulation 2006;113:118-124.

11. Zouggari Y, Ait-Oufella H, Waeckel L, Vilar J, Loinard C, Cochain C, et al. Regulatory T cells modulate postischemic neovascularization. Circulation 2009;120:1415-1425.

12. Hellingman A, van der Vlugt L, Lijkwan M, Bastiaansen A, Sparwasser T, Smits H, et al. A limited role for regulatory T cells in post-ischemic neovascularization. J Cell Mol Med 2011.

13. Chalothorn D, Clayton JA, Zhang H, Pomp D, Faber JE. Collateral density, remodeling, and VEGF-A expression differ widely between mouse strains. Physiol Genomics 2007;30:179-191.

14. Dokun AO, Keum S, Hazarika S, Li Y, Lamonte GM, Wheeler F, et al. A quantitative trait locus (LSq-1) on mouse chromosome 7 is linked to the absence of tissue loss after surgical hindlimb ischemia. Circulation 2008;117:1207-1215.

15. Helisch A, Wagner S, Khan N, Drinane M, Wolfram S, Heil M, et al. Impact of mouse strain differences in innate hindlimb collateral vasculature. Arterioscler Thromb Vasc Biol 2006;26:520-526.

16. Schulte S, Sukhova GK, Libby P. Genetically programmed biases in Th1 and Th2 immune responses modulate atherogenesis. Am J Pathol 2008;172:1500-1508.

17. Anderson SK, Dewar K, Goulet ML, Leveque G, Makrigiannis AP. Complete elucidation of a minimal class I MHC natural killer cell receptor haplotype. Genes Immun 2005;6:481-492.

(18)

18. Ryan JC, Naper C, Hayashi S, Daws MR. Physiologic functions of activating natural killer (NK) complex-encoded receptors on NK cells. Immunol Rev 2001;181:126-137.

19. Yokoyama WM, Plougastel BF. Immune functions encoded by the natural killer gene complex. Nat Rev Immunol 2003;3:304-316.

20. Scalzo AA, Lyons PA, Fitzgerald NA, Forbes CA, Shellam GR. The BALB.B6-Cmv1r mouse:

a strain congenic for Cmv1 and the NK gene complex. Immunogenetics 1995;41:148-151.

21. Cheng TP, French AR, Plougastel BF, Pingel JT, Orihuela MM, Buller ML, et al. Ly49h is necessary for genetic resistance to murine cytomegalovirus. Immunogenetics 2008;60:565- 573.

22. Hellingman AA, Bastiaansen AJ, de Vries MR, Seghers L, Lijkwan MA, Lowik CW, et al.

Variations in surgical procedures for hind limb ischaemia mouse models result in differences in collateral formation. Eur J Vasc Endovasc Surg 2010;40:796-803.

23. van Weel V, Deckers MM, Grimbergen JM, van Leuven KJ, Lardenoye JH, Schlingemann RO, et al. Vascular endothelial growth factor overexpression in ischemic skeletal muscle enhances myoglobin expression in vivo. Circ Res 2004;95:58-66.

24. Fernandez NC, Treiner E, Vance RE, Jamieson AM, Lemieux S, Raulet DH. A subset of natural killer cells achieves self-tolerance without expressing inhibitory receptors specific for self-MHC molecules. Blood 2005;105:4416-4423.

25. Kim S, Poursine-Laurent J, Truscott SM, Lybarger L, Song YJ, Yang L, et al. Licensing of natural killer cells by host major histocompatibility complex class I molecules. Nature 2005;436:709-713.

26. Taherzadeh Z, VanBavel E, de VJ, Matlung HL, van MG, Brewster LM, et al. Strain- dependent susceptibility for hypertension in mice resides in the natural killer gene complex.

Am J Physiol Heart Circ Physiol 2010;298:H1273-H1282.

27. Proteau MF, Rousselle E, Makrigiannis AP. Mapping of the BALB/c Ly49 cluster defines a minimal natural killer cell receptor gene repertoire. Genomics 2004;84:669-677.

28. Sun JC, Lanier LL. Tolerance of NK cells encountering their viral ligand during development.

J Exp Med 2008;205:1819-1828.

29. Hoglund P, Brodin P. Current perspectives of natural killer cell education by MHC class I molecules. Nat Rev Immunol 2010;10:724-734.

30. Scalzo AA, Brown MG, Chu DT, Heusel JW, Yokoyama WM, Forbes CA. Development of intra-natural killer complex (NKC) recombinant and congenic mouse strains for mapping and functional analysis of NK cell regulatory loci. Immunogenetics 1999;49:238-241.

31. Barten R, Torkar M, Haude A, Trowsdale J, Wilson MJ. Divergent and convergent evolution of NK-cell receptors. Trends Immunol 2001;22:52-57.

32. Wilson MJ, Torkar M, Haude A, Milne S, Jones T, Sheer D, et al. Plasticity in the organization and sequences of human KIR/ILT gene families. Proc Natl Acad Sci U S A 2000;97:4778-4783.

33. Hak L, Mysliwska J, Wieckiewicz J, Szyndler K, Trzonkowski P, Siebert J, et al. NK cell compartment in patients with coronary heart disease. Immun Ageing 2007;4:3.