scholarly journals The role of antigen-presenting cells in triggering graft-versus-host disease and graft-versus-leukemia

Blood ◽  
2007 ◽  
Vol 110 (1) ◽  
pp. 9-17 ◽  
Author(s):  
Ronjon Chakraverty ◽  
Megan Sykes

After allogeneic blood or bone marrow transplantation, donor T cells interact with a distorted antigen-presenting cell (APC) environment in which some, but not all, host APCs are replaced by APCs from the donor. Significantly, host APCs are required for the priming of acute graft-versus-host disease (GVHD). Donor APCs play a lesser role in the induction of acute GVHD despite their predicted capacity to cross-present host antigens. In contrast, donor APCs may play a role in perpetuating the tissue injury observed in chronic GVHD. Host APCs are also required for maximal graft-versus-leukemia responses. Recent studies have suggested potential strategies by which the continued presence of host APCs can be exploited to prime strong donor immunity to tumors without the induction of GVHD.

Blood ◽  
2005 ◽  
Vol 105 (5) ◽  
pp. 2227-2234 ◽  
Author(s):  
Britt E. Anderson ◽  
Jennifer M. McNiff ◽  
Dhanpat Jain ◽  
Bruce R. Blazar ◽  
Warren D. Shlomchik ◽  
...  

AbstractThe application of allogeneic stem cell transplantation (alloSCT) is limited by graft-versus-host disease (GVHD). GVHD can be divided into acute and chronic forms that likely have different requirements for initiation and pathogenesis mechanisms. In prior studies we demonstrated that residual host antigen-presenting cells (APCs) were required to initiate acute GVHD (aGVHD) mediated by CD8 T cells. In contrast, here we demonstrate that either donor or host APCs can initiate CD4-mediated GVHD in a model that has features of chronic GVHD (cGVHD). Both donor and host APCs must provide CD80/86-dependent costimulation to elicit maximal cGVHD, and there is no GVHD when both donor and host lack CD80/86. Finally, we were surprised to find that, although either donor or host APCs are sufficient to stimulate skin cGVHD, donor APCs play a dominant role in intestinal cGVHD. Both CD40 and CD80/86 are critical for donor APC function in intestinal cGVHD, but only CD80/86 is required for skin cGVHD. Thus, there are target-tissue–specific differences in APC requirements. These results identify differences in APC requirements between CD8-mediated aGVHD and CD4-mediated cGVHD. They further highlight donor APCs as additional targets for GVHD therapy.


Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 843-843
Author(s):  
Natalie Stickel ◽  
Gabriele Prinz ◽  
Dietmar Pfeifer ◽  
Annette Schmitt-Graeff ◽  
Marie Follo ◽  
...  

Abstract Introduction: Acute graft-versus-host disease (GvHD) arises from the attack of recipient tissues by donor allogeneic T cells and represents one of the major limitations of allogeneic hematopoietic cell transplantation (allo-HCT). In spite of many clinical trials, the standard immunosuppressive regimens for prevention of acute GvHD have improved little in the last two decades. Hence, a better understanding of the biology of acute GvHD may improve therapeutic options. MicroRNA-146a (miR-146a) was found to be increased in the sera of patients with GvHD. Therefore, we aimed to decipher the role of miR-146a in allogeneic donor T cells during GvHD by functional studies and in patients undergoing allo-HCT by single nucleotide polymorphism (SNP) analysis. Methods: We used two different murine major MHC mismatch models for acute GvHD. Recipient mice were conditioned with irradiation before transplantation of bone marrow and either wildtype or miR-146a deficient T cells from allogeneic donor mice. Furthermore, genomic DNA from 289 patients that underwent allo-HCT and their respective hematopoietic stem cell donors was isolated in order to determine their miR-146a rs2910164genotype. Results: We observed miR-146a upregulation in T cells of mice developing acute GvHD compared to untreated mice in a major MHC and a minor histocompatibility antigen mismatch model. Transfer of miR-146a deficient T cells caused increased GvHD severity, elevated TNF serum levels and reduced survival. Conversely, the phytochemical induction of miR-146a or its overexpression in donor T cells using a specific miR-146a mimic reduced GvHD severity. TNF receptor-associated factor 6 (TRAF6), a verified target of miR-146a, was upregulated in miR-146a-/- T cells following alloantigen stimulation. Higher TRAF6 levels translated into increased NF-κB activity and TNF production in miR-146a-/- T cells, while other pro-inflammatory cytokine levels were unaffected. The detrimental effect of miR-146a deficiency in T cells could be antagonized by TNF blockade in vivo. Moreover, in contrast to WT T cells, over expression of miR-146a in Tnf deficient T cells had no effect on their alloreactivity. In the human system, the minor genotype of the SNP rs2910164, which causes reduced miR-146a expression, was more frequent in patients developing acute GvHD grade III/IV compared to all other allo-HCT recipients (n=289). Conclusions: Taken together we show that miR-146a functions as a negative regulator of the TRAF6/TNF-axis in allogeneic donor T cells during GvHD, leading to reduced TNF transcription. Given our observation on the predictive role of the SNP leading to decreased miR-146a expression in acute GvHD in patients and the possibility to exogenously enhance miR-146a expression, we provide a novel and targeted molecular approach to mitigate GvHD. Disclosures No relevant conflicts of interest to declare.


Blood ◽  
2004 ◽  
Vol 104 (11) ◽  
pp. 3059-3059
Author(s):  
Britt Anderson ◽  
Jennifer McNiff ◽  
Dhanpat Jain ◽  
Bruce R. Blazar ◽  
Warren D. Shlomchik ◽  
...  

Abstract The application of allogeneic stem cell transplantation is limited by graft-vs.-host disease (GVHD). GVHD can be divided into acute and chronic forms that likely have different initiation requirements and pathogenic mechanisms. Previously we demonstrated that residual host antigen presenting cells (APC) were required for CD8-mediated acute GVHD (aGVHD). In contrast, here we show that either donor or host APCs can initiate CD4-mediated GVHD in the B10.D2(H-2d)→BALB/c (H-2d) model of chronic GVHD (cGVHD). We selectively impaired donor APCs, host APCs or both by using donors and/or hosts deficient in both costimulatory molecules CD80 and CD86 (CD80/86−/−). When all CD28:CD80/86 signaling was prevented by using CD80/86−/− donors and hosts, wild type (WT) CD4 cells were unable to induce cGVHD. Therefore donor CD4 cells absolutely require signals from CD80/86 to mediate cGVHD. Next we compared cGVHD in CD80/86−/− BM + CD4→WT vs. WT BM + CD4→CD80/86−/− hosts, to debilitate antigen presentation by donor or host APCs, respectively. Cutaneous cGVHD developed in both groups, demonstrating that donor or recipient APCs are sufficient to initiate disease. However, the incidence of cutaneous cGVHD in CD80/86−/− recipients was less than that in WT→WT recipients (P<0.01), suggesting that recipient APCs are more important than donor APCs for skin cGVHD. In support of this, cGVHD was not reduced inCD80/86−/− BM + CD4 →WT recipients. Therefore, reconstitution with defective donor APCs does not affect disease when host APCs are intact. To address whether host APCs were dispensable, we compared cGVHD in donor→host (APCs >98% donor) and host→host chimeric recipients. Similar cGVHD developed in both chimeras; thus donor APCs alone are sufficient, consistent with the data in CD80/86−/− hosts. Interestingly, cGVHD was initiated in the absence of significant host hematopoietic antigen. Thus, exogenously acquired parenchymal antigen was sufficient. The above experiments focused on cutaneous GVHD. However, recipients of WT BM versus non-WT BM consistently developed diarrhea and increased weight loss. While CD4 recipients of CD80/86−/− or CD40−/− BM regained and maintained their original weights, CD4 recipients of WT BM never returned to their pre-transplant weight (P<0.01 after day 15). Consistent with the clinical finding, recipients of CD4 cells and WT BM had higher bowel pathology scores than equivalent mice that received CD80/86−/− BM (P<0.01). Recipients of CD80/86−/− BM + WT CD4 cells had statistically indistinguishable path scores from recipients of either type of BM without CD4 cells (P=0.19 and 0.73); therefore, without CD80/86 expression on donor BM, there was no statistical evidence that donor T cells caused bowel GVHD. These results provide new insights into the role of donor APCs in GVHD and demonstrate that APC requirements differ depending on the site of disease, both novel findings. They also identify differences in APC requirements between CD8-mediated aGVHD and CD4-mediated cGVHD and further highlight APCs as additional targets for GVHD prevention and therapy.


Blood ◽  
2021 ◽  
Author(s):  
Qingxiao Song ◽  
Xiaoning Wang ◽  
Xiwei Wu ◽  
Hanjun Qin ◽  
Yingfei Li ◽  
...  

Donor T cells mediate both graft-versus-leukemia (GVL) activity and graft-versus-host disease (GVHD) after allogeneic hematopoietic cell transplantation (Allo-HCT). Development of methods that preserve GVL activity while preventing GVHD remains a long-sought goal. Tolerogenic anti-IL-2 monoclonal antibody (mAb) (JES6-1) forms anti-IL-2/IL-2 complexes that block IL-2 binding to IL-2Rb and IL-2Rg on Tcon cells that have low expression of IL-2Rα. Here we show that administration of JES6 early after Allo-HCT in mice markedly attenuates acute GVHD while preserving GVL activity that is dramatically stronger than observed with tacrolimus (TAC) treatment. The anti-IL-2 treatment down-regulated activation of IL-2-Stat5 pathway and reduced production of GM-CSF. In GVHD target tissues, enhanced T cell PD-1 interaction with tissue-PD-L1 led to reduced activation of AKT-mTOR pathway and increased expression of Eomes and Blimp-1, increased T cell anergy/exhaustion, expansion of Foxp3-IL-10-producing Tr1 cells, and depletion of GM-CSF-producing Th1/Tc1 cells. In recipient lymphoid tissues, lack of donor T cell PD-1 interaction with tissue-PD-L1 preserved donor PD-1+TCF-1+Ly108+CD8+ T memory progenitors (Tmp) and functional effectors that have strong GVL activity. Anti-IL-2 and TAC treatments have qualitatively distinct effects on donor T cells in the lymphoid tissues, and CD8+ Tmp cells are enriched with the anti-IL-2 treatment compared to TAC treatment. We conclude that administration of tolerogenic anti-IL-2 mAb early after Allo-HCT represents a novel approach for preventing acute GVHD while preserving GVL activity.


Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 522-522
Author(s):  
A. Mario Q. Marcondes ◽  
Laura Tabellini ◽  
John A. Hansen ◽  
Charles A. Dinarello ◽  
H. Joachim Deeg

Abstract Abstract 522 Graft-versus-host disease (GVHD) is an important complication of allogeneic hematopoietic cell transplantation (HCT). The role of various cell populations, cytokines and chemokines, present pre and post-transplantation, in the development of GVHD has been studied extensively.We investigated the potential role of Interleukin (IL)-32 in alloreactivity and GVHD. IL-32 is the protein product of the NK4 transcript first reported in IL-2 activated T lymphocytes and natural killer cells. IL-32 has pro-inflammatory and pro-apoptotic properties and induces expression of TNF-α in several cell targets. We used one-way mixed lymphocyte cultures (MLC) as a simple in vitro model of GVHD to determine IL-32 expression upon alloactivation. The α and γ isforms of (IL)-32 protein were 2-fold upregulated in allogeneic MLC compared to autologous controls (n=4, p=0.037). Concurrently, the concentrations of TNF-α, IL-6 and IL-8 in the allogeneic MLC supernatants were, as expected, upregulated significantly. This finding led us to evaluate IL-32 expression as a potential marker for GVHD after allogeneic HCT in 45 patients with either active acute GVHD or chronic GVHD, and 16 patients who did not show clinical evidence of GVHD. IL-32 mRNA levels in peripheral blood unsorted white blood cells, correlated with acute GVHD (RT-PCR values expressed as mean +/− SEM of the ratio of expression of IL-32/GUS-B = 1.01, p=0.011, vs control values IL-32/GUS-B = 0.23) but not with chronic GVHD, (IL-32/GUS-B = 0.26, p=0.16) (Figure 1). As the serine protease neutrophil proteinase 3 (PR3) has been shown to serve as activator of IL-32, and to process several inflammatory cytokines, including IL-32, TNF-α and IL-8, we postulated that the addition of the serine protease inhibitor α-1 anti-trypsin (AAT), would interfere with the processing of IL-32 by PR3, and as a result would lead to decreased proliferation of cells in MLC, and reduced cytokine production. To test this hypothesis, MLCs were treated with AAT at concentrations of 1–20 ug/ml and expression of IL-32 and PR3 were determined. In MLCs treated with AAT at the optimal concentration of 5 ug/ml, added to cultures on alternate days for a period of 7 days T lymphocyte proliferation was suppressed when compared to vehicle-treated MLC, (mean CPM=33.000 versus CPM=67.000; p=0.012). Concurrently there was a 2.5 fold decrease in IL-32 and PR3 protein levels (n=4, p=0.023). CONCLUSION: IL-32 is upregulated in patients with acute GVHD. Determination of IL-32 in patients with suspected GVHD may support the diagnosis and the decision to treat.AAT interferred with T cell activation and the release of cytokines, including IL-32, suggesting that administration of AAT may have therapeutic potential in patients with acute GVHD. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 728-728 ◽  
Author(s):  
Yu Yu ◽  
Dapeng Wang ◽  
Kenrick Semple ◽  
Claudio Anasetti ◽  
Xue-Zhong Yu

Abstract Abstract 728 Background: Allogeneic hemopoietic stem cell transplantation (HCT) is an effective therapy with potential cure of hematological malignancies through T cell-mediated graft-versus-leukemia (GVL) effects. However, beneficial GVL effects are frequently offset by the development of destructive graft-versus-host disease (GVHD) also induced by donor T cells. Recent studies including ours have demonstrated that donor T cells differentiated into type 1 or type 17-subset contribute to GVHD. Thus, we hypothesize that blocking both Th1 and Th17 lineage via disrupting Th1-specific (T-bet) and Th17-specific (RORγt) transcription factors can significantly reduce GVHD after allo-HCT. Method: Two murine models of bone marrow transplantation (BMT) that represent clinical GVHD and GVL were used: C57BL/6 (B6)→BALB/c and B6→(B6 × DBA2)F1. To mimic clinical residual hematological malignant disease, B cell lymphoma (A20) and mastocytoma (p815) were infused into BALB/c and (B6 × DBA2)F1 mice, respectively. Results: We first compared the ability of WT, T-bet−/−, RORγt−/−, and T-bet−/−/RORγt−/− T cells in the induction of GVHD, and found that RORγt−/− T cells had a comparable ability to cause GVHD as WT T cells, whereas T-bet−/− T cells were less pathogenic. The T-bet−/−/RORγt−/− T cells failed to induce acute GVHD but caused minor to modest chronic GVHD in some of recipients at the doses tested. To investigate whether recipients of T-bet−/−/RORγt−/− T cells had less severe target organ GVHD damage, we analyzed GVHD associated organ damage in liver, lung and bowel. Fourteen days after adoptive transfer of WT, T-bet−/−, RORγt−/−, and T-bet−/−/RORγt−/− T cells, recipients which received T-bet−/−/RORγt−/− donor T cells showed markedly reduced T cell infiltration and tissue damage in liver, lung, and bowel. Mechanistic studies revealed that T-bet−/−/RORγt−/− T cells produced significantly less IFNγ (Th1 cytokine) and IL-17 (Th17-cytokine) but significantly more IL-4 and IL-5 (Th2-cytokines) as compared to WT T cells. In addition, T-bet−/−/RORγt −/− donor T-cells express significantly less CXCR3 and CCR6, chemokine receptors required for infiltration of alloreactive T cells into GVHD targeted organ, which could be the reason that significantly fewer T-bet−/−/RORγt−/− T cells were accumulated in recipient liver and lung than WT T cells. Furthermore, we tested the ability of WT and T-bet−/−/RORγt−/− T cells in mediating GVL effect. Although T-bet−/−/RORγt−/− T cells failed to induce acute GVHD, their ability to reject A20 or p815 cells was comparable to that of the WT T cells at the dose tested. Conclusions: These results indicate that blocking T-bet and RORγt prevents acute GVHD by suppressing donor T cell differentiation towards Th1 and Th17 and promoting differentiation towards Th2, and inhibiting donor T cell expansion and infiltration into GVHD target organs. Furthermore, blocking T-bet and RORγt could preserve GVL effect. Thus, the current study validates new targets for the separation of donor T cell–mediated GVHD and GVL activity, which could eventually be beneficial to patients with hematological malignancies. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2013 ◽  
Vol 122 (21) ◽  
pp. 4463-4463
Author(s):  
Sarah Morin-Zorman ◽  
Christian Wysocki ◽  
Catherine Matte-Martone ◽  
Kathryn W Juchem ◽  
Hung Sheng Tan ◽  
...  

Graft-versus-host disease (GVHD) limits the broader application of allogeneic hematopoietic stem cell transplantation. In prior studies we defined roles for both host and donor-derived antigen presenting cells (APCs) in the activation of alloreactive donor T cells and in promotion of GVHD. While initial T cell activation in GVHD occurs predominantly in secondary lymphoid organs, we have consistently observed MHCII+ donor-derived APCs, including dendritic cells (DCs), in histopathologic GVHD lesions, frequently adjacent to infiltrating T cells, suggesting they have a role in local GVHD reactions. Donor-derived tissue APCs (t-APCs), including tissue-DCs (t-DCs) could activate donor T cells through indirect or cross-presentation of host antigens, produce chemokines that recruit other effectors, and elaborate inflammatory mediators or suppressors of inflammation. We first characterized t-DC subsets in the skin and bowel of GVHD-affected mice. 129 (H-2b) hosts were irradiated and reconstituted with B6 (H-2b) BM with or without CD4+ and CD8+ T cells to induce GVHD and analyzed mononuclear cells from skin and bowel approximately 4 weeks post transplant. In skin, both main dermal DC populations (CD11b+ and CD103+) were significantly increased in GVHD mice as compared to BM alone controls, though the ratios of CD11b+: CD103+ DCs were similar. In the bowel lamina propria, the ratios of CD11b+CD103- to CD11b+CD103+ were increased in GVHD mice in the colon but were similar to that in BM alone controls in the small bowel. We next studied the roles of CCR6 and CCR2 in the recruitment of donor-derived APCs to skin and bowel. We transplanted mice with CCR6-/- BM in competition with wild type (wt) BM and found that the contribution of each to skin and bowel APCs matched their contributions to myeloid hematopoiesis in BM, spleen and blood, indicating that CCR6 is not required. To study the role of CCR2 we first compared mice transplanted with either wt or CCR2-/- BM with wt T cells. Despite having a profound reduction in blood monocytes, all skin and bowel t-APC subsets were present in CCR2-/- recipients, indicating that CCR2 is not required for t-APC recruitment in contrast to its role in many other models of inflammation. However, CD103+ DCs were more prevalent relative to CD11b+ DCs, consistent with a pre-cDC origin. Despite monocytopenia, recipients of CCR2-/- BM developed clinical GVHD; histology data is being analyzed and will be presented. To better define the contributions of CCR2 to t-APC recruitment and to determine monocyte versus pre-cDC origin of t-DCs, we transplanted mice with CCR2-/- BM in competition with wt BM and compared ratios of BM and blood precursors (pre-cDCs and monocytes) to t-DC ratios. For CD103+ DCs, wt/KO ratios matched the ratios of general myeloid hematopoiesis and pre-cDCs, indicating a pre-cDC origin. For CD11b+CD103- DCs, the ratio of wt/KO matched that in blood monocytes. We further subsetted CD11b+ t-DCs based on the expression of Ly6C, MAR1, CD64 and CD24, used to differentiate pre-cDC from mono-derived DCs in other organs, and did not identify any population with wt/KO ratios that did not match that of the general CD11b+ DC population, suggesting that most if not all CD11b+ t-DCs are of monocyte origin. Experiments are underway examining the role of CX3CR1 in t-APC recruitment and these data will be presented. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2010 ◽  
Vol 116 (2) ◽  
pp. 287-296 ◽  
Author(s):  
Geoffrey R. Hill ◽  
Rachel D. Kuns ◽  
Neil C. Raffelt ◽  
Alistair L. J. Don ◽  
Stuart D. Olver ◽  
...  

Abstract Suppressor of cytokine signaling-3 (SOCS3) is the main intracellular regulator of signaling by granulocyte colony-stimulating factor, an immune-modulatory cytokine used to mobilize stem cells for transplantation. We have therefore studied the contribution of SOCS3 to the spectrum of graft-versus-host disease (GVHD) after allogeneic stem cell transplantation (SCT). Grafts from SOCS3−/Δvav donor mice in which SOCS3 deficiency is restricted to the hematopoietic compartment had an augmented capacity to induce acute GVHD. With the use of SOCS3−/ΔLysM and SOCS3−/Δlck donors in which SOCS3 deficiency was restricted to the myeloid or T-cell lineage, respectively, we confirmed SOCS3 deficiency promoted acute GVHD mortality and histopathology within the gastrointestinal tract by effects solely within the donor T cell. SOCS3−/Δlck donor T cells underwent enhanced alloantigen-dependent proliferation and generation of interleukin-10 (IL-10), IL-17, and interferon-γ (IFNγ) after SCT. The enhanced capacity of the SOCS3−/Δlck donor T cell to induce acute GVHD was dependent on IFNγ but independent of IL-10 or IL-17. Surprisingly, SOCS3−/Δlck donor T cells also induced severe, transforming growth factor β– and IFNγ-dependent, sclerodermatous GVHD. Thus, the delivery of small molecule SOCS3 mimetics may prove to be useful for the inhibition of both acute and chronic GVHD.


JBMTCT ◽  
2020 ◽  
Vol 1 (1) ◽  
pp. 53-66
Author(s):  
Vaneuza A. M. Funke ◽  
Maria Claudia Rodrigues Moreira ◽  
Afonso Celso Vigorito

Graft versus host disease is one of the main complications of Hematopoietic stem cell, in­volving about 50% to 80% of the patients. Acute GVHD clinical manifestations and therapy is discussed, as well as new NIH criteria for the diagnosis and classification of chronic GVHD. Therapy for both refractory chronic and acute GVHD is an important field of discussion once there is no superiority for the majority of the agents after primary therapy has failed. Hence, this review is meant to be a useful tool of consultation for clinicians who are dealing with this complex complication.


Blood ◽  
1989 ◽  
Vol 73 (6) ◽  
pp. 1720-1728 ◽  
Author(s):  
KM Sullivan ◽  
PL Weiden ◽  
R Storb ◽  
RP Witherspoon ◽  
A Fefer ◽  
...  

Abstract To assess the influence of graft-versus-host disease (GVHD) on recurrent leukemia and survival after allogeneic marrow transplantation, we studied 1,202 patients with acute nonlymphocytic leukemia (ANL), acute lymphocytic leukemia (ALL), and chronic myelogenous leukemia (CML) given unmodified marrow grafts from HLA- identical siblings. Proportional hazards regression models using acute GVHD and chronic GVHD as time-dependent covariates demonstrated a significant association of GVHD with a decreased relative risk (RR, 0.33 to 0.42) of relapse in patients with ANL, ALL, and CML transplanted in advanced disease. Among patients developing either acute or chronic GVHD, treatment failure (that is, mortality or relapse) was decreased in patients with ALL transplanted in relapse (RR = 0.70, P less than .033) and CML in blast crisis (RR = 0.37, P less than .009). This effect was independent of age, sex, preparative regimen, GVHD prophylaxis, or length of follow-up. Five-year actuarial estimates were derived for the subset of 657 patients who survived in remission 150 days after transplant and were at risk for development of chronic GVHD. Among patients with ANL in first remission or CML in chronic phase, GVHD had an adverse effect on survival and no apparent influence on relapse. Among patients with ANL and ALL transplanted in relapse, the probability of relapse after day 150 was 74% without [corrected] GVHD, 45% with acute and chronic GVHD, 35% with [corrected] only acute GVHD, and 34% with only chronic GVHD (P less than .001). Actuarial survival in these four GVHD groups was 25%, 34%, 59%, and 62%, respectively (P less than .009). Among patients with CML in acceleration or blast crisis, the probability of relapse after day 150 was 65% without GVHD and 36% with acute and/or chronic GVHD (P less than .017). We conclude that acute and chronic GVHD were associated with a durable antileukemic effect and improved survival in patients transplanted in advanced stages of ALL and CML.


Sign in / Sign up

Export Citation Format

Share Document