Induction and Scoring of Graft-Versus-Host Disease in a Xenogeneic Murine Model and Quantification of Human T Cells in Mouse Tissues using Digital PCR

Summary

Here, we present a protocol to induce and score disease in a xenogeneic graft-versus-host disease (xenoGVHD) model. xenoGVHD provides an in vivo model to study immunosuppression of human T cells. Additionally, we describe how to detect human T cells in tissues with digital PCR as a tool to quantify immunosuppression.

Abstract

Acute graft-versus-host disease (GVHD) is a significant limitation for patients receiving hematopoietic stem cell transplant as therapy for hematological deficiencies and malignancies. Acute GVHD occurs when donor T cells recognize host tissues as a foreign antigen and mount an immune response to the host. Current treatments involve toxic immunosuppressive drugs that render patients susceptible to infection and recurrence. Thus, there is ongoing research to provide an acute GVHD therapy that can effectively target donor T cells and reduce side effects. Much of this pre-clinical work uses the xenogenic GVHD (xenoGVHD) murine model that allows for testing of immunosuppressive therapies on human cells rather than murine cells in an in vivo system. This protocol outlines how to induce xenoGVHD and how to blind and standardize clinical scoring to ensure consistent results. Additionally, this protocol describes how to use digital PCR to detect human T cells in mouse tissues, which can subsequently be used to quantify efficacy of tested therapies. The xenoGVHD model not only provides a model to test GVHD therapies but any therapy that can suppress human T cells, which could then be applied to many inflammatory diseases.

Introduction

Allogeneic hematopoietic stem cell transplant (HSCT) has become routine treatment for patients suffering from hematological malignancies such as leukemia with poor prognosis. A significant complication of HSCT is acute graft-versus-host disease (GVHD). A 2012 study reported that acute GVHD developed in 39% of HSCT patients receiving transplants from sibling donors and 59% of patients receiving transplants from unrelated donors¹. Acute GVHD occurs when donor-derived T cells attack recipient’s organs. The only successful therapy for GVHD is treatment with highly immunosuppressive drugs², which are highly toxic and increase the risk of infection and tumor recurrence. Thus, despite improvements that have been made in acute GVHD survival in recent years^3,4,5, there is still a critical need for improved GVHD therapies with minimal toxicity that promote long-term remission.

The overall goal of the following methods is to induce and score xenogeneic GVHD (xenoGVHD). The xenoGVHD model was developed as a tool to induce acute GVHD with human cells rather than murine cells allowing for more direct translation of pre-clinical GVHD research to clinical trials⁶. This model involves intravenously injecting human peripheral blood mononuclear cells (PBMC) into NOD-SCID IL-2Rγnull (NSG) mice that are sublethally irradiated. Injected human T cells are activated by human antigen presenting cells (APCs) presenting murine antigen and the activated T cells migrate to distant tissues resulting in systemic inflammation and ultimately death^6,7,8,9,10. Disease pathology and progression in the xenoGVHD model closely mimic human acute GVHD. Specifically, the pathogenic human T cells are reactive to murine major histocompatibility complex (MHC) proteins, which is similar to the T cell alloreactivity in human GVHD^6,9. The primary advantage of the xenoGVHD model over the mouse MHC-mismatch model, the other widely used GVHD model, is it allows for testing of therapies on human cells rather than murine cells. This allows for testing of products that can directly be translated to the clinic without any modifications because they are made to target human cells. Recently, this model has been used to test a human anti-IL-2 antibody¹¹, human thymic regulatory T cells (Tregs)¹² and human mesenchymal stem cells¹³ as potential treatments for acute GVHD. In a wider context, this model can be used as an in vivo suppression assay for any drug or cell type that can suppress human T cell activity. For example, Stockis et al.¹⁴ used the xenoGVHD model to study the effect of blocking integrin αVβ8 on Treg suppressive activity in vivo. Thus, the xenoGVHD model can provide insight into the mechanism of any therapy targeting T cells in an in vivo setting.

An additional method described in this protocol is how to detect human T cells in mouse tissues using digital polymerase chain reaction (dPCR). The goal of this method is to offer a tool to quantify migration and proliferation of T cells in target tissues, which measure efficacy of immunosuppressive therapies being tested in this model. dPCR is a relatively novel method for quantification of nucleic acids¹⁵. Briefly, the PCR reaction mixture is divided into partitions that contain small numbers of the target sequence or no target at all. The target sequence is then amplified and detected using DNA intercalating dyes or fluorescent target-specific probes. dPCR quantifies the number of copies of target sequence based on the fraction of positive partitions and Poisson’s statistics^15,16. Detecting T cells with dPCR requires much less tissue compared with other alternative methods, including flow cytometry and histology, and can be performed on frozen or fixed tissue. dPCR does not require a standard curve to determine copy numbers, nor are technical replicates required. This reduces the amount of reagent and template DNA needed for dPCR compared to traditional quantitative PCR (qPCR)¹⁶. Partitioning the PCR reaction into sub-reactions in dPCR effectively concentrates targets¹⁷. Thus, dPCR is primarily a tool for detection of rare targets in a large amount of non-target DNA. For example, dPCR is being used to detect bacterial contamination in milk¹⁸, identify rare mutations in the estrogen receptor gene¹⁹, and detect circulating tumor DNA in the blood of patients²⁰. In this protocol, dPCR serves as an efficient tool for detecting and quantifying human T cells in tissues of mice with xenoGVHD.

Protocol

All mouse experiments were performed in compliance, and with approval from, the University of Kansas Medical Center Institutional Animal Care and Use Committee. All healthy human blood samples were obtained under informed consent and with approval from the Institutional Review Board at the University of Kansas Medical Center.

1. Irradiation of NSG Mice

1. One day prior to PBMC injection, irradiate 8–12-week old NSG mice (either sex can be used). In a sterile biosafety cabinet, place mice in a sterilized pie cage or microisolator. Irradiate mice in a Cs¹³⁷ source or small animal irradiator (e.g., RS 2000) with a total dose of 150 cGy with slow rotation to ensure even irradiation.
2. Place mice into clean cages in a sterile biosafety cabinet.

2. Preparation of human PBMC for injection

1. Collect enough healthy human blood to isolate 1.1 x 10⁷ PBMC per mouse. Dilute the heparinized blood in an equal volume of 2% fetal bovine serum (FBS) in phosphate buffered saline (PBS).
   NOTE: With this protocol, the PMBC yield is generally 0.5–1 x 10⁷ per 10 mL of whole blood. Each mouse will receive 10⁷ PBMC in 100 µL of PBS. The extra 1 x 10⁶ in 10 µL per mouse ensures that each mouse receives the full dosage of PBMC, should there be any issues with filling syringes.
2. Add 15 mL lymphocyte separation density gradient medium (e.g., Ficoll) to a 50 mL conical tube and then carefully overlay up to 25 mL of diluted blood on top of the density gradient. Centrifuge the tube containing density gradient and diluted blood at 400 x g for 40 min at room temperature without brake.
3. Harvest the PBMC interface into 10 mL of PBS in a 50 mL conical tube. Centrifuge the cells at 400 x g for 10 min. Remove the supernatant.
4. Loosen pellet by flicking tube and resuspend in 10 mL PBS to wash the PBMC. Centrifuge the cells at 400 x g for 5 min.
5. (Optional) Discard the supernatant and lyse red blood cells (RBCs) in the cell pellet by adding a volume of Ammonium-Chloride-Potassium (ACK) lysis buffer that is equal to the pellet volume. Gently resuspend the pellet and swirl the tube for 30–60 s. Fill tube with serum-free RPMI media and centrifuge the tube at 400 x g for 5 min.
6. Remove the supernatant and re-suspend the cell pellet in 5 mL of PBS. Count the cells using trypan blue exclusion with a hemocytometer and a microscope.
7. Centrifuge the cells at 400 x g for 5 min. Remove the supernatant and resuspend cells at 1 x 10⁸ cells/mL in PBS.

3. Retro-orbital injection of human PBMC into mice²¹

1. Place an anesthesia chamber in a laminar flow hood to maintain sterility. Pre-charge the anesthesia chamber with 5% isoflurane and 1 L/min oxygen flow rate for 5 min.
2. Reduce isoflurane to 2% and put mice from the same cage into the anesthesia chamber. Once mice lose righting, anesthetize mice for 5 min in chamber. During this time, pre-fill syringe with 100 µL (1 x 10⁷ cells) of cell suspension.
3. Place mouse on a heating pad with its nose in a nose cone to maintain anesthesia. Check for loss of consciousness by pinching a foot pad and checking for lack of reflex.
4. Restrain mouse with the thumb and middle finger. Insert the 28 G 1/2 in needle with the bevel down lateral to the medial canthus, through the conjunctival membrane until the back of the eye is reached. Slightly retract the needle and slowly inject 100 μL of cells (1 x 10⁷ cells).
5. Fully retract the needle and properly dispose of the syringe and needle. Close the eyelid and apply mild pressure to the injection site with a gauze sponge.
6. Examine the injection site for swelling or other visible trauma. Allow the mouse to regain consciousness in a sterile cage lined with paper towels to prevent aspiration of bedding before moving to home cage. After any bleeding has stopped, apply a single drop of proparacaine to the eye for analgesia.
   NOTE: Tail vein injection is an alternative method of intravenous injection of PBMC for this protocol as described by Macholz et al.²².

4. Clinical Scoring of acute GVHD in Mice (Figure 1)²³

1. Measure GVHD score every other day until mice reach a score of a 2 and then every day until day of sacrifice.
   1. Place the cage in a laminar flow hood, remove the food and water and put the lid back on. Score activity by observing mice for 5 min and assign scores as follows: 0 = mouse starts walking within a couple minutes and keeps walking around cage, 1 = mouse takes a more than a couple minutes to get up and walks slowly around cage, 2 = mouse doesn’t get up in 5 min and only walks when touched.
2. Weigh each mouse in a glass beaker and give a weight loss score: 0 = <10% change, 1 = 10–25% change and 2 = >25% change.
3. While the mouse is still in the beaker, inspect for posture: 0 = normal, 1 = hunching at rest, 2 = hunching impairs movement; fur texture: 0 = normal, 1 = mild to moderate ruffling, 2 = severe ruffling, and skin integrity : 0 = normal, 1 = scaling of paws/tail, 2 = obvious areas of denuded skin (look at ears, tail and paws for scaling).
4. With five categories and a score of 0–2 for each category, a mouse can reach a maximum score of 10. When a mouse reaches a score of 7 or greater or reaches 42 days post-injection, euthanize mouse via CO₂ euthanasia or other method approved by the local Institutional Animal Care and Use Committee.
   NOTE: To ensure accuracy of results, the scoring is performed by a researcher who is blinded to the treatment groups²⁴. Additionally, although >20% of body weight loss is the recommended humane endpoint for many IACUC protocols, with additional justification IACUC approval of this protocol can be obtained.

5. Harvesting tissues for genomic DNA from euthanized mice and isolating genomic DNA

1. Dissect mice using sterile surgical tools. Cut a small piece of tissue about 3 mm x 0.5 mm in size from an organ of interest, e.g., lung, liver, or spleen. Weigh the tissue sample and place the tissue piece in a sterile 1.5 mL tube. If collecting multiple tissues, wash tools with ethanol between cutting each organ.
2. Freeze tissues by submerging the tubes containing the tissue in liquid nitrogen until it stops bubbling and store in -80 °C overnight. Thaw the tissue samples and lyse them according to a genomic DNA (gDNA) isolation kit.
3. Isolate genomic DNA according to manufacturer’s instructions as described in the kit. Be sure to note the amount of tissue that was processed per microliter of eluted gDNA (mg/μL).
   NOTE: Frozen tissues can be stored at -80 °C for processing at a later time.

6. Quantification of human T cells using digital PCR (Figure 2)

1. Prepare digital PCR reactions according to the protocol for DNA binding dyes for the digital PCR machine being used. Use the following primers specific for human CD3 epsilon genomic DNA (NCBI Reference Sequence: NG_007383.1); Forward Primer: AGGCTGCCTTAACTCCCAAG, Reverse Primer: GCCCTACCAGCTGTGGAAAC. These primers will yield a single band of 105 bp.
   NOTE: Add 0.5 μL of a restriction enzyme, such as HindIII, to each reaction to digest the genomic DNA. Be sure to test out different gDNA dilutions to optimize separation of positive and negative droplets. Additionally, a subset of infiltrating T cells may be non-pathogenic regulatory T cells. These cells can be quantified using additional primers for regulatory T cell markers.
2. Carry out the digital PCR reaction under the following conditions: Lid at 105 °C, 95 °C for 10 min (1 cycle); 95 °C for 30 s, ramp 2 °C/s and 55 °C for 1 min, ramp 2 °C/s (40 cycles); 72 °C for 10 min, 12 °C hold.
   NOTE: These parameters may need to be adjusted depending on digital PCR equipment.
3. Acquire digital PCR data with analysis software. Report data as numbers of copies per mg of tissue using the following equation: (number of copies/μL) x (μL of gDNA in reaction) x (dilution factor)/(mg of tissue/μL of total gDNA )

Results

Sublethally irradiated 8-12-week old NSG mice of both sexes that received human PBMC started displaying clinical signs of GVHD around day 10 post injection compared to negative control mice that received PBS only (Figure 1A). XenoGVHD mice had a median survival of 23.5 days (Figure 1B). With digital PCR, CD3 epsilon positive human T cells could be detected in the lung and liver samples of mice that received human PBMC. Tissue sam...

Discussion

Disease progression is generally consistent in the xenoGVHD model, even with injection of PBMC from different donors, so multiple experiments can be combined. The key steps required to maintain this consistency are proper i.v. injection technique, blinding and consistent scoring. A study by Nervi et al.²⁵ demonstrated that compared to intravenous tail vein injection, retro-orbital injections of PBMC resulted in more consistent engraftment and more severe GVHD. Leon-Rico et al.^26<...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1.5 mL eppendorf tubes	Fisher	05-408-129
10 mL serological pipet	VWR International	89130-898
10mL BD Vacutainers - Green capped with Sodium Heparin	Becton Dickinson	366480
250 µL Ranin pipette tips	Rainin	17001118	Do not use other pipettes or pipet tips for droplet generation
50 mL conical tube	VWR International	89039-656
96-Well ddPCR plate	Bio-Rad	12001925
ACK (Ammonium-Chloride-Potassium) Lysing Buffer	Lonza	10-548E	Optional
Alcohol Wipes	Fisher Scientific	6818
Anesthesia Chamber	World Precision Instruments	EZ-178	Provided by animal facility
Anesthesia Machine	Parkland Scientific	PM1002	Provided by animal facility
BD Vacutainer Safety-Lok Blood Collection Set	Becton Dickinson	367281
DG8 Cartridges and Gaskets for QX100/QX200 Droplet Generator	Bio-Rad	1864007
DNAse and RNAse free Molecular Grade H2O	Life Technologies	1811318
Ethyl alcohol, Pure,200 proof, for molecular biology	Sigma-Aldrich	E7023-500ML
Fetal Bovine Serum	Atlanta Biologicals	S11150
Ficoll	Fisher Scientific	45001750
Insulin Syringe	Fisher Scientific	329424
Isoflurane	Sigma-Aldrich	CDS019936	Provided by animal facility
Liquid nitrogen	N/A	N/A
Mouse Irradiator Pie Cage	Braintree Scientific, Inc.	MPC 1	Holds up to 11 mice
Nexcare Gentle Paper Tape (a.k.a. 3M Micropore Surgical Tape / 3/4")	Fisher Scientific	19-027-761
P1000 pipetman	MidSci	A-1000
P200 pipetman	MidSci	A-200
Pierceable Foil Heat Seal	Bio-Rad	1814040
Pipetaid Gilson Macroman	Fisher Scientific	F110756
Pipet-Lite Multi Pipette L8-200XLS+	Rainin	17013805	Do not use other pipettes or pipet tips for droplet generation
Qiagen DNeasy Blood and Tissue Kit	Qiagen	69506
qPCR plates	VWR International	89218-292
QX200 Droplet Digital PCR System	Bio-Rad	12001925	Includes droplet generator, droplet reader, laptop computer, software, associated component consumables, for EvaGreen or probe-based digital PCR applications
QX200 Droplet Generation Oil for EvaGreen	Bio-Rad	1864006
QX200 ddPCR EvaGreen Supermix	Bio-Rad	1864033
RNase and DNase-free plate seal	Thermo Scientific	12565491
RPMI Advanced 1640	Life Technologies	12633012
Sterile Gauze Pads (2" x 2", 12-Ply)	Fisher Scientific	67522
Sterile Phosphate Buffered Saline	Fisher Scientific	21040CV
Sterile reservoir	VWR International	89094-662
Surgial Scissors	Kent Scientific	INS600393-4
Surgical Forceps	Kent Scientific	INS650914-4