Modifiers of inflammatory angiogenesis in a murine model 3

3.5 Effects of neutrophil depletion on angiogenesis in the corneal injury model A Corneal neovascularization at day 3 and day 5 after injury in the control and RB6-8C5 treatment groups.

CHAPTER III RESULTS

3.1 Introduction of corneal and skin injury models

3.1.1 Angiogenesis in the corneal injury model

The corneal injury model has been used to investigate angiogenesis (Azar 2006) In this study, we have been able to standardize the extent of injury to allow for a fairly consistent angiogenic response on the cornea The corneal angiogenesis in BALB/c mice is shown in Figure 3.1 After 24 hours, the pre-existing vessels vasodilated and corneas became cloudy

By 72 hours, the sprouts lengthened and multiplied to produce a rich anastomosing plexus

72 hours to 5 days vascular sprouts elongated and matured, and the cloudiness of cornea diminished The angiogenesis reached a peak at day 5 after injury and the vascular sprouts reached the injured sites From day 7 regression started, and new vessels began to fade Our results are consistent with those reported by Burger et al (1983) and Sunderkötter et al (1991) Based on the current data, we have been able to establish that the angiogenesis response is the highest at day 5 after injury

3.1.2 Sexual dimorphism in the corneal and skin injury model

To reduce potential variation in experiments, the differences in angiogenesis and skin wound healing were observed between BALB/c male and female mice The present study shows

male mice (Fig 3.2) In contrast there was no significant difference in skin wound healing between male and female mice (Fig 3.3) However, skin wound healing in female mice tended to close faster than male mice at days 3, 5 and 7 after injury As such, only female mice were used in all the subsequent experiments to establish the corneal and skin injury models

Fig 3.1 Angiogenesis response in the corneal injury model

The time course of angiogenesis response in the cornea of BALB/c mouse before injury (a), 0.5 hours (b), 1 day (c), 3 days (d), 5 days (e) and 7 days (f) after injury

Fig 3.2 Comparisons of corneal angiogenesis between BALB/c female and male mice

Quantitive analysis of neovascularized area (A) and vessel length (B) at day 5 after corneal

Fig 3.3 Skin wound healing in BALB/c female and male mice

Wound closure was evaluated by morphometrical analysis of the wound areas Each wound region was digitally photographed at the indicated time points, and the percentage of wound areas to the initial areas was calculated from the photographs The bars show the mean ± SEM of independent experiments (n = 6)

3.2 Key roles of neutrophils in angiogenesis in the corneal injury model

3.2.1 Lymphocytes have little impacts on angiogenesis in the corneal injury model

Although lymphocytes has been indicated to produce angiogenic factors (Lingen, 2001), their roles in angiogenesis induced by injury are still unclear We established the corneal injury model in Rag1KO mice and control mice to study the roles of lymphocytes in inflammatory angiogenesis As shown in Figure 3.4, there was no difference in corneal angiogenesis between BALB/c and Rag1 KO mice indicating that lymphocytes may not play important roles in natural inflammatory angiogenesis

Fig 3.4 Comparison of corneal angiogenesis between female BALB/c and Rag1 KO mice

A Corneal angiogenesis at day 3 and day 5 after injury in BALB/c and Rag1KO mice B and C: Quantitative analysis of neovascularization at day 5 after injury The bars show the

mean ± SEM of independent experiments (n = 6)

3.2.2 Key roles of neutrophils in angiogenesis in the corneal injury model

Neutrophils are the dominant infiltrating leukocyte population in the early stages of inflammation Neutrophils have been shown to play an important role in the inflammatory response, acting as a first line of self-defense against invading microorganisms (Kasama et al., 2005; Scapini et al., 2000) However, recent studies have indicated that neutrophils may also play an important role in angiogenesis Isolated human neutrophils are known to release VEGF from preformed stores upon activation (Gaudry et al 1997) Neutrophils are

also associated with angiogenesis in a Matrigel sponge model in vivo induced (Kibbey et

al., 1994) The induction of angiogenesis by IL-8/CXCL8 is also neutrophil dependent, the angiogenic response being completely abrogated upon neutrophil depletion (Benelli et al.,

2002) However, no direct in vivo evidence relates the neutrophil to natural inflammatory

angiogenesis Thus, we have used a corneal injury model to study the role of neutrophils in the process of natural inflammatory angiogenesis as well as the mechanisms underlying the observed response

I Efficacy of neutrophil depletion by RB6-8C5 treatment

To determine the role of neutrophils in inflammatory angiogenesis, neutrophils were systemically depleted by intraperitoneal administration of RB6-8C5 Neutrophil depletion

blood counts are shown in Table 3.1 Generally, very few neutrophils were found in the peripheral blood of RB6-8C5-treated mice RB6-8C5 treatment resulted in more than 97% reduction in the number of circulating neutrophils during the course of experiment (Table 3.1) This result indicated that RB6-8C5 antibody could efficiently deplete circulating neutrophils, which provides a reproducible systemic neutropenia model for the study of neutrophil function Neutrophil depletion had minimal effects on the differential count of monocytes

II Effects of neutrophil depletion on corneal angiogenesis

The effect of neutrophil depletion on the corneal angiogenesis was evaluated by the biomicroscopic observations, microvessel counts and measurements in new vessel length and neovascularized area (Fig 3.5)

Figure 3.6 shows corneas at days 3 and 5 after injury in the control and RB6-8C5 treatment groups In the control group, the limbal vessels began sprouting into the corneas at day 3 after the injury Subsequently, vascular sprouts matured and elongated to the injury site, reaching it at day 5, with the new vessel length and neovascularized area at 0.68±0.04 mm and 2.08±0.15 mm2 respectively (Fig 3.5B, 3.5C) Neutrophil depletion severely inhibited corneal angiogenesis: fewer new buds of vessels were present in the corneas of

RB6-8C5-treated mice by day 3, and no obvious progression of new vessel growth during the process of angiogenesis was visible (Fig 3.5A) By quantitative image analysis, neutrophil depletion resulted in more than 90% reduction in new blood vessel length and

neovascularized area compared with the control mice at day 5 after injury (Fig 3.5B, 3.5C)

We next assessed the role of neutrophils in angiogenesis by determining the microvessel density Figure 3.6 demonstrates that neutrophil depletion decreased the microvessel density In the control group, a 1.5-fold increase occurred in the microvessel density from day 3 to day 5 after injury Compared with the control group, 75% and 90% reduction in the microvessel density were observed in the RB6-8C5 treatment group at day 3 and day 5 after injury, respectively Moreover, the microvessel density remained unchanged from day 3 to day 5 in the RB6-8C5 treatment group

III Effects of neutrophil depletion on corneal inflammation response

24 hours after injury the corneas became cloudy in both control and RB6-8C5-treated groups and the degree of cloudiness was graded We found that RB6-8C5-treated mice showed significantly less corneal opacity than control mice from day 1 to day 3 (Fig 3.7)

At day 1 after injury the mean scale of control group was 2.2, while it was 1.4 in RB6-8C5 -treated group At day 5 after injury, the mean scale of control group increased to 2.4; however, in RB6-8C5 treated group, it decreased to 1.1 Current data suggest that neutrophil depletion significantly reduced the opacity in the cornea following injury, which is consistent with the previous work (Oshima et al., 2006)

IV Effects of neutrophil depletion on neutrophil infiltration

We then examined whether the observed inhibition of the angiogenic response in the RB6-8C5 treatment group was due to the impaired neutrophil recruitment Corneal

neutrophil counts were performed 1, 3, and 5 days after the injury by

immunohistochemical stain (Fig 3.8) In the control group, the neutrophil infiltration was highest at day 1 after injury and subsequently decreased at day 3 and day 5 (Fig.3.8 B control 1 and control 2) H&E stains (Fig 3.8 B control 2) were performed to confirm the results of the immunohistochemistry stain In the RB6-8C5 treatment group, there were few neutrophils detected in the cornea during the 5 days after the injury Uninjured corneas did not possess any neutrophils in the control and RB6-8C5 treatment groups (data not shown)

In the normal cornea there are a few residential macrophages After injury, the number of macrophage in the cornea had no increase both in the control and RB6-8C5 treated mice There were no T or B cells detected in the uninjured cornea At day 5 after injury, few T cells were detected in the corneas of control and RB6-8C5 treated mice We could not detect B cells in the corneas of control and RB6-8C5-treated mice during 5 days following the injury (data not shown)

Table 3.1 Peripheral blood counts in the control and RB6-8C5 treated mice

Groups % of total counts( days after injury)

Mice were given 0.1 mg of control IgG or RB6-8C5 at 3 days interval beginning on day -1

Blood smears were stained with Giemsa stain, and at least 250 cells per slide were counted

Data are presented as means± standard deviation The data are representative of 3 separate

experiments Each experiment included five mice

0 0.5 1 1.5 2 2.5

***

Fig 3.5 Effects of neutrophil depletion on angiogenesis in the corneal injury model

A Corneal neovascularization at day 3 and day 5 after injury in the control and RB6-8C5 treatment groups B and C Quantitative analysis of neovascularization at day 5 after injury

Bars show the mean ± SEM of independent experiments (n = 10) *** P < 0.001 compared

with control group

Fig 3.6 Effects of neutrophil depletion on microvessel density in the cornea

A Representative photographs of CD31 immunohistochemistry stain in the corneas of

control and RB6-8C5 treatment groups at day 5 after injury (original magnification x 100,

*

Fig 3.7 Effects of neutrophil depletion on the degree of corneal opacity

Scale of cornea opacity was evaluated at days 1and 3 after injury in the control and RB6-8C5 treated groups Bars show the mean ± SEM of independent experiments (n = 10)

* P < 0.05 compared with control group

A

0 1000

Days after injury

Fig 3.8 Infiltration of neutrophils in the corneal angiogenesis

show the mean ± SEM of independent experiments (n = 6) The neutrophil counts in RB6-8C5 treated group were extremely low

B

Fig 3.8 Infiltration of neutrophils in the corneal angiogenesis

B Representative micrographs of immunohistochemistry and hematoxylin and eosin (H&E)

staining of corneal sections in the control and RB6-8C5 treatment groups RB6-8C5

Immunostaining for neutrophils in the cornea of the RB6-8C5 treatment group after injury

Bar 100 μm (original magnification:×100) Control 1 Immunostaining for neutrophils in

the cornea of the control group after injury Bar 100 μm (original magnification: ×100)

Control 2 H&E staining of neutrophils in the cornea of the control group after injury Bar

30 μm (original magnification: ×400, oil)

V Induction and localization of VEGF in cornea

VEGF is a potent stimulator of the endothelial cell growth and can stimulate both the physiological and pathological angiogenesis (Kim et al., 1992) In the present study, we compared VEGF proteins in the cornea at 0, 1, 3 and 5 days after injury by the ELISA method In the control group, the protein level of VEGF increased nearly 2 fold from day 0

to day 1 after injury, decreased at day 3, and returned to near control levels by day 5 (8.42±2.37 pg/cornea) Neutrophil depletion decreased VEGF protein levels by 63% and 54% at day 1 and day 3 after injury, respectively (Fig 3.9A)

To determine the cellular source of VEGF in the injured cornea, serial sections of eyeballs taken 1 day after injury were immunostained for VEGF in the control and RB6-8C5 treatment groups Both corneal epithelial cells and the infiltrating neutrophils were

positively stained with VEGF in the cornea of the control group (Fig 3.9B) No positive staining was observed in the negative control for VEGF (Fig 3.9B) In contrast, in the RB6-8C5 treatment group, no VEGF positive signals were detected in the corneas (Fig 3.9B) No VEGF-positive stain was observed in the normal corneas (data not shown)

Day 0 Day 1 Day 3 Day 5

Days after injury

***

***

Fig 3.9 Detection of VEGF in the cornea during angiogenesis

A Time kinetics for protein levels of VEGF after injury in the control and RB6-8C5

treatment groups The bars show the mean ± SEM (n=6) *** P < 0.005 compared with

control group

B

Fig 3.9 Detection of VEGF in the cornea during angiogenesis

B Representative photographs of VEGF immunohistochemistry staining in the cornea at day 1 after injury a Negative control for VEGF in the cornea at day 1 after injury (original magnification: x 100) b Immunostaining for VEGF in the cornea of the control group at day 1 after injury (original magnification: x 100) c Immunostaining for VEGF in the

Immunostaining for VEGF in the cornea of the RB6-8C5 treatment group (original

magnification: x 100, bar 100 µm) f Immunostaining for VEGF in the cornea of the

RB6-8C5 treatment group (original magnification: x 400, bar 30 µm)

VI Effects of neutrophil depletion on induction of MIP-1α, MIP-2, TNF-α and MCP-1 in cornea

MIP-2 and TNF-α have been reported to have critical roles in angiogenesis (Szekanecz and Koch 2001; Kim et al 2001; Fahey et al 1990) MIP-1α is able to promote neutrophil infiltration (Wolpe et al 1988) Therefore, we determined the protein levels of MIP-1α,

MIP-2, and TNF-α using ELISA in the in vivo murine corneal injury model (Fig

3.10-3.12)

MIP-2 and MIP-1α protein levels were increased significantly at day 1 after injury and decreased markedly by day 5 In the uninjured cornea, the levels of MIP-1α and MIP-2 proteins were not detected In contrast, TNF-α (28.77±2.97 pg/cornea) was observed in the uninjured cornea and the level remained unchanged after the injury

Neutrophil depletion was shown to inhibit the protein levels of MIP-1α and MIP-2 in the cornea (Fig 3.10; 3.11) MIP-1α and MIP-2 proteins in the RB6-8C5 treatment group were found to be only 27% and 24% of the levels in the control group at day 1 after injury By day 5, the proteins were undetectable in the RB6-8C5-treated cornea Neutrophil depletion had no effects on the levels of TNF-α in the cornea compared with the control group (Fig 3.12)

The expression of MCP-1 protein was also observed (Fig.3.13) The protein level of MCP-1 was increased at day 1, and decreased at day 3 after injury in both control and RB6-8C5 treated mice In contrast to the level of MIP-1α and MIP-2, the protein level of MCP-1 was significantly higher in RB6-8C5-treated group as compared with control group (Fig 3.13)

VII PMA- induced VEGF, MIP-1α and MIP-2 release from murine neutrophils in a

in vitro study

Human neutrophils have been shown to have an intracellular pool of VEGF and release VEGF upon PMA stimulation (Gaudry et al 1997) In this study, we investigated whether murine neutrophils have preformed stores of VEGF and could release them upon

stimulation As shown in Figure 3.14, VEGF was undetectable in the supernatant, whereas,

a considerable amount of VEGF was found inside the neutrophils (0 and 16.88 pg/106 cells, respectively) After PMA treatment, a small amount of intracellular VEGF were detected while a considerable amount of VEGF was detected in the supernatant (0.3 and 12.14 pg/106 cells, respectively) Upon PMA stimulation, the maximal level of VEGF secreted was similar to the intracellular content of unstimulated neutrophils The data suggests the possible existence of an intracellular pool of VEGF in mouse neutrophils Pretreatment of neutrophils with cycloheximide did not affect PMA-induced VEGF release, reinforcing the

hypothesis that VEGF may be released from a pre-existing intracellular pool instead of de novo synthesis (Fig 3.14)

We investigated the contents of MIP-1α and MIP-2 in the culture supernatant and inside the neutrophils with or without PMA activation In the unstimulated neutrophils, no MIP-1α and MIP-2 was found intracellularly However, after stimulation, the amount of MIP-1α and MIP-2 in the cell supernatant reached 222 pg and 276 pg, respectively,

suggesting that neutrophils produced MIP-1α and MIP-2 from de novo synthesis (Fig

3.15) The hypothesis was further supported by the finding that PMA-induced MIP-1α and MIP-2 production was abrogated by the pretreatment with cycloheximide

Days after injury

***

*

Fig 3.10 Time kinetics for protein levels of MIP-1α in the corneal injury model

Corneas were removed at the indicated time after injury Corneal lysate was prepared and individually assayed by enzyme-linked immunosorbent assay Bars show the mean ± SEM

(n=6) *P < 0.05, **P < 0.01, ***P < 0.005 compared with control group

Days after injury

*

Fig 3.11 Time kinetics for protein levels of MIP-2 in the corneal injury model

Corneas were removed at the indicated time after injury Corneal lysate was prepared and individually assayed by enzyme-linked immunosorbent assay The bars show the mean ±

SEM (n=6) *P < 0.05 compared with control group

Days after injury

Control RB6-8C5 treatment

Fig.3.12 Time kinetics for protein levels of TNF-α in the corneal injury model

Corneas were removed at the indicated time after injury Corneal lysate was prepared and individually assayed by enzyme-linked immunosorbent assay The bars show the mean ± SEM (n=6)

Days after injury

*

Fig 3.13 Time kinetics for protein levels of MCP-1 in the corneal injury model

Corneas were removed at the indicated time after injury Corneal lysate was prepared and individually assayed by enzyme-linked immunosorbent assay Bars show the mean ± SEM

(n=6) *P < 0.05, compared with control group

Fig 3.14 Effects of PMA on the release of VEGF from murine neutrophils

Neutrophils were cultured for 2 hours with or without 100 ng/ml Phorbol-12-myristate 13-acetate (PMA) after preincubation with or without 10 µg/ml of cycloheximide (CHX) for 30 minutes at 37ºC Culture supernatants were separated from the cells by centrifugation The bars show the mean values ±SEM of the VEGF protein between the cellular and extracellular compartments at unstimulated and stimulated neutrophils, depicted as cell-associated and released The data are representative of 4 separate experiments Each experiment included 3 samples

3.3 Important roles of neutrophils and lymphocytes in skin wound

healing

3.3.1 Important roles of lymphocytes in wound healing

We found that the wound healing in Rag1KO mice was significantly delayed from day 7 following the injury compared with control mice 9 (Fig.3.17) In Rag1KO mice there was

no T cell detected in the normal skin and injured skin across all the time points There was also no T cell infiltration during wound healing in Rag1KO mice Furthermore, there were

no significant differences in inflammatory cell infiltration (neutrophil and macrophage) (Fig.3.18, 3.19), angiogenesis, and the protein levels of VEGF, MIP-1a, MCP-1, and TNF-α between control and Rag1KO mice (Fig.3.18-3.23, 3.25, 3.26) Only MIP-2 levels

at days 5 and 9 and TGF-β1 levels at day 9 after the injury were significantly lower then control mice (Fig 3.24, 3.27) The scar formation in Rag1KO mice was not different from control mice significantly (data not shown)

3.3.2 Important roles of neutrophils in skin wound healing

I Efficacy of neutrophil depletion by RB6-8C5 treatment in skin injury model

To determine the role of neutrophils in skin wound healing, neutrophils were systemically depleted by intraperitoneal administration of RB6-8C5 Neutrophil depletion was

confirmed by Giemsa staining of the blood smear Peripheral blood was collected from mice at days -1, 0, 2, 5, 7 and 8 after injury to determine the extent of neutropenia

Peripheral blood counts are shown in Figure 3.16 Generally, very few neutrophils were found in the peripheral blood of RB6-8C5-treated mice during 5 days following the injury From day 5 after the injury, the differential neutrophil count began to increase and reached the pre-RB6-treatment level in RB6 treated Rag1+/- and RB6 treated Rag1KO mice at day

7 after injury

This result indicated that RB6-8C5 antibody can efficiently deplete circulating neutrophils, which provides a reproducible systemic neutropenia model for the study of neutrophil function

Fig 3.16 Effects of RB6-8C5 on the neutrophil differential count in control, Rag1KO, RB6-control and RB6-Rag1KO mice

Mice were given 0.1 mg of control IgG or RB6-8C5 at 3 days interval beginning on day -1 Blood smears were stained with Giemsa stain to detect neutrophils At least 150 cells per slide were counted Data are presented as means± SEM The data are representative of 3 separate experiments Each experiment included 4 mice

II Effects of neutrophil depletion on skin wound healing

Full-thickness 8-mm punch biopsies were produced on the shaved back of control,

Rag1KO, RB6-control and RB6-Rag1KO mice Wound sizes were monitored up to 20 days, and the data is shown in Figure 3.17

RB6-8C5 treatment in control and Rag1KO mice delayed the wound closure from day 3 after injury compared with control and Rag1 KO mice, respectively, even after neutrophil count in peripheral blood returned to the pre-RB6-treatement level This demonstrates that neutrophil depletion severely inhibited the skin wound healing process Compared with Rag1KO mice, wound closure was also significantly delayed in RB6-8C5-treated control mice at days 3, 7, and 11 after the skin injury Similarly, RB6-Rag1KO mice showed significant delayed wound closure compared with RB6-control mice at days 7, 9 and 11 after injury Taken together, our current data suggest that neutrophils play important roles

in skin wound healing

A

Fig 3.17 Skin wound healing in control, Rag1KO, RB6-control and RB6-Rag1KO mice

A Representative macroscopic views of skin wounds on days 0, 5 and 9 after injury

Full-thickness wounds (8 mm in diameter) were made with a punch biopsy instrument and wound healing was monitored by taking digital photographs All the digital photograph was processed in the same procedure

B

Fig 3.17 Skin wound healing in control, Rag1KO, RB6-control and RB6-Rag1KO mice

B: Evaluation of wound closure by morphometrical analysis of the wound areas Each wound region was digitally photographed at the

III Effects of neutrophil depletion on neutrophil infiltration in skin wound

We next examined whether the observed inhibition of wound healing in Rag1KO,

RB6-control and RB6-Rag1KO mice were due to the impaired inflammatory cell

recruitment in the skin wounds

Neutrophils normally appear at the skin wound within minutes of injury (Park and Barbul, 2004) Skin neutrophil counts were assessed 1, 5, and 9 days after the injury by

immunohistochemical study using anti-neutrophil antibody (Fig 3.18) H&E stains were performed to confirm the results of the immunohistochemistry staining Uninjured skin did not possess any neutrophils in the control mice In the control group, neutrophil infiltration peaked at day 1 after injury and subsequently decreased at day 5 At day 9, there were few neutrophils detected in the wound bed There was no significant difference in neutrophil infiltration between control and Rag1KO group during 9 days following injury (Fig 3.18)

In RB6-8C5- treated control and RB6-8C5- treated Rag1KO mice, neutrophil infiltration was severely inhibited during 5 days following the skin injury At day 9, neutrophil

infiltration into the skin wounds increased, but the level was still lower than the highest level observed in the control and Rag1KO mice

Days after injury

*

^^

Fig 3.18 Neutrophil infiltration in skin wounds in control, Rag1KO, RB6-control

and RB6-Rag1KO mice

Neutrophils were identified by immunohistochemistry with anti-neutrophil antibody

Immunoreactive inflammatory cells per high power field (HPF) were determined by

counting immunostained cells in 5 high power fields at x 400 magnification in the wounds

at days 1, 5, and 9 after skin injury (n= 6) Bars show the mean ± SEM * P<0.05, ***

P<0.001 compared with control group ^ P<0.05, ^^ P<0.01, ^^^ P<0.001 compared with

Rag1KO group

IV Effects of neutrophil depletion on macrophage infiltration in skin wound

In the inflammatory response monocytes/macrophages migrate into the wound 48-96 hours after injury and become the predominant cell population before fibroblast migration and replication (Park and Barbul, 2004) Monocyte/macrophage influx is widely regarded to play an essential role for effective wound healing, as wound healing is severely impaired when monocyte infiltration is prevented (Leibovitch & Ross, 1975) To determine whether the initial neutrophil impairment results in subsequent alterations in monocyte recruitment

to the wound site, macrophages were immunostained with a rat monoclonal antibody (IgG)

to F4/80+ and were manually counted As shown in Figure 3.19, there were no significant differences in macrophage recruitment among control, Rag1KO, RB6-control and

RB6-Rag1KO mice In the uninjured skin, macrophages were also detected and there were

no differences between control and Rag1KO mice (data not shown)

V Effects of neutrophil depletion on T cell infiltration in skin wound

It has been shown that T cell plays an important part in wound healing, infiltrating the injured site 3 days after injury (Park and Barbul, 2004) In the present study, T cell

recruitment was evaluated using immunostaining antibody CD3ε in skin wounds of control, Rag1KO, RB6-control and RB6-Rag1KO mice (Fig.3.20) T cell staining was also

evaluated in the uninjured skin of control mice In the normal skin of control mice, T cells were mainly detected in the epidermal layer (7/HPF) At day 5 after the injury, there were

three T cells per x400 high power field in the wound beds of control mice Subsequently, T cells recruited in skin wounds increased to seven cells per high power field at day 9 after injury (Fig.3.20) RB6-8C5 treatment in control mice did not affect T cell recruitment to skin wounds during the 9 days following injury In Rag1KO and RB6-Rag1KO group, no

T cells were detected in the uninjured skin and injured skin across all time points