Increased migration of antigen presenting cells to newly-formed lymphatic vessels in transplanted kidneys by glycol-split heparin

Abstract

Background: Chronic renal transplant dysfunction is characterized by loss of renal function and tissue remodeling, including chronic inflammation and lymph vessel formation. Proteoglycans are known for their chemokine presenting capacity. We hypothesize that interruption of the lymphatic chemokine-proteoglycan interaction interferes with the lymphatic outflow of leukocytes from the renal graft and might decrease the anti-graft allo-immune response.

Methods: In a rat renal chronic transplant dysfunction model (female Dark-Agouti to male Wistar Furth), chemokines were profiled by qRT-PCR in microdissected tubulo-interstitial tissue. Disruption of lymphatic chemokine-proteoglycan interaction was studied by (non-anticoagulant) heparin-derived polysaccharides in vitro and in renal allografts. The renal allograft function was assessed by rise in plasma creatinine and urea.

Results: Within newly-formed lymph vessels of transplanted kidneys, numerous CD45+ leukocytes were found, mainly MHCII+, ED-1-, IDO-, HIS14-, CD103- antigen presenting cells, most likely representing a subset of dendritic cells. Treatment of transplanted rats with regular heparin and two different (non-)anticoagulant heparin derivatives revealed worsening of kidney function only in the glycol-split heparin treated group despite a two-fold reduction of tubulo-interstitial leukocytes (p<0.02). Quantitative digital image analysis however revealed increased numbers of intra-lymphatic antigen-presenting cells only in the glycol-split heparin group (p<0.01). The number of intra-lymphatic leukocytes significantly correlates with plasma creatinine and urea, and inversely with creatinine clearance.

Conclusions: Treatment of transplanted rats with glycol-split heparin significantly increases the number of intra-lymphatic antigen presenting cells, by increased renal diffusion of lymphatic chemokines, thereby increasing the activation and recruitment of antigen presenting cells towards the lymph vessel. This effect is unwanted in the transplantation setting, but might be advantageous in e.g., dendritic cell vaccination.

Fig 1. CTD in allografted kidneys is accompanied by lymph vessel formation.

(A) Counting of VEGFR3+ lymph vessels revealed a threefold increase in allografted kidneys compared to the non-transplanted contralateral kidneys from the same donor rats (p<0.01). qRT-PCR of podoplanin and perlecan in dissected tubulo-interstitium revealed an increase in the expression of the (B) lymphatic endothelial marker podoplanin and (C) the vascular basement membrane marker perlecan in allografted kidneys (N = 5) compared to non-transplanted control kidneys (N = 5)(both p<0,05). Data are presented as mean ± SEM. DA-to-DA isografted kidneys (N = 5) did not show an increase in the expression of podoplanin and perlecan. Double staining for perlecan (D) and lymphatic marker LYVE-1 (E) revealed colocalization of perlecan and lymphatic endothelium (F).

Fig 2. Tubulo-interstitial upregulation of lymphatic chemokines.

qRT-PCR for lymphatic chemokines in laser dissected tubulo-interstitium reveals a strong increase of CCL21, CCL19, CCL2 and CCL5 in allografted kidneys (N = 5) compared to healthy control kidneys (N = 5) (P<0,01, P<0,001,P<0,01 and P<0,05 respectively). No upregulation was seen in CX3CL1 and CXCL12 expression in allografts. Isografted kidneys (N = 5) showed upregulation of only CX3CL1. Experiments were done in triplicate. Data are presented as mean ± SEM.

Fig 3. Intra-lymphatic leukocytes in CTD are mainly HLA-II⁺, ED-1^-, IDO^-, CD103^- and HIS14^- APCs.

Intra-lymphatic leukocytes were scored by counting intra luminal nucleated cells in a podoplanin/PAS double staining (A). Intra lymphatic MHCII⁺ cells (B)(green), macrophages (ED-1)(C)(green),dendritic cells (CD103)(D)(green), B-cells (HIS14)(E)(red) and tolerogenic cells (IDO)(F)(red) were scored in a double staining with the lymphatic marker VEGFR3 or Lyve-1. Closed arrows point out intralymphatic accumulation of cells positive for the respective staining. Open arrows indicate presence of cells positive for the respective staining outside the lymphatic system. Comparing the intra-lymphatic amount of different leukocyte species in untreated renal allografts (N = 7) revealed that the main cell accumulating in the tubulo-interstitium is MHCII⁺, ED-1^-, IDO^-, CD103^- and HIS14^- (G), likely representing antigen presenting cells. Scale bars represent 50μm. Data are presented as mean ± SEM.

Fig 4. Effect of heparin/non-anticoagulant heparin derivative treatment on leukocyte influx in rat renal transplantation.

Adhesion (white arrows) of leukocytes (CD45 in red) to blood vessel endothelium (vWF in green) in kidney allografts is reduced in the glycol-split heparin treated rats (N = 7) compared to the vehicle treated rats (N = 7) (p<0.05) (A-C). Moreover, glycol-split heparin treatment (N = 7) also reduces leukocyte count in the interstitium of transplanted rat kidneys (CD45 in red), compared to vehicle treated animals (N = 7) (p<0.05) (D-F). No significant differences were seen between heparin (N = 4) or N-acetylated heparin treated animals (N = 5) compared to vehicle treated animals (N = 7). Dotted lines represent vessel lumen. Scale bars represent 50μm. Data are presented as mean ± SEM.

Fig 5. Glycol-split heparin treatment increased intra-lymphatic leukocyte accumulation.

Nucleated cells on the luminal side of lymph vessels were identified and counted to assess the intralymphatic accumulation of leukocytes in heparin/non-anticoagulant heparin derivative treated allograft kidneys. Glycol-split heparin treatment resulted in an increased lymphatic leukocyte accumulation compared to vehicle treated animals (p<0,01)(A). Data is expressed as the number of cells per 100 μm of endothelium. Staining for both lymphatic endothelium and MHCII revealed a borderline increase of MHCII positive cells in the lymphatic lumen in glycol-split heparin treated (N = 7) compared to vehicle treated animals (N = 7) (p = 0,08) (B). Intralymphatic scoring of CD103⁺ dendritic cells, HIS14⁺ B-cells and ED-1⁺ macrophages did not reveal any effect of heparin/non-anticoagulant heparin derivative treatment on the intraluminal presence of these cells (C-E). No significant differences were seen between heparin (N = 4) or N-acetylated heparin treated animals (N = 5) compared to vehicle treated animals (N = 7) in either of the intralymphatic cell populations. Scoring was performed as in Fig 3. Data are presented as mean ± SEM.

Fig 6. Glycol-split heparin revealed a high binding affinity for chemokines resulting in reduced (peri-)lymphatic CCL21 presence in glycol-split heparin treated kidneys.

To determine whether heparin/non-anticoagulant heparin derivatives are capable of competing with the CCL2 and CCL21 interaction with perlecan, CCL2 or CCL21 were incubated together with a concentration range of heparin/non-anticoagulant heparin derivatives on a perlecan coated plate. Residual CCL2 and CCL21 binding to the immobilized perlecan was measured and revealed that glycol-split heparin has the highest affinity for CCL2 and CCL21 (A+B). Significant differences are expressed as glycol-split heparin compared to heparin (†) and N-acetylated heparin (*). Experiments were done 3 times in duplicate. In vivo treatment with glycol-split heparin (N = 7) resulted in a reduced CCL21 staining of (peri-)lymphatic areas compared to vehicle treated animals (N = 7) (C-E). Lymph vessels are pointed with white arrows. Scale bars represent 50μm. * = p<0,05, ††/** = p<0,01, †††/*** = p<0,001. Data are presented as mean ± SEM.

Fig 7. Kidney function correlated with lymphatic leukocytes.

Correlation analysis between kidney function (plasma creatinine, plasma urea and creatinine clearance) and histological parameters (interstitial inflammation, interstitial fibrosis and lymphatic leukocytes) revealed that there is no correlation between kidney function and interstitial inflammation or interstitial fibrosis. However, there was a strong correlation between kidney function and the total number of intra lymphatic leukocytes (C: p = 0.001, F: p<0.001, I: p<0.01).