Preclinical Canine Model of Graft-versus-Host Disease after InUtero Hematopoietic Cell Transplantation

Jesse D. Vrecenak 1,2, Erik G. Pearson 1, Carlyn A. Todorow 1, Haiying Li 1, Mark P. Johnson 1,Alan W. Flake 1,*1 Children’s Center for Fetal Research, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania2 Division of Pediatric Surgery, Washington University in St. Louis, St. Louis, Missouri

Article history:Received 23 March 2018Accepted 15 May 2018

Key Words:In utero hematopoietic celltransplantationGraft-versus-host disease

A B S T R A C T

In utero hematopoietic cell transplantation (IUHCT) offers the potential to achieve allogeneic engraftment andassociated donor-specific tolerance without the need for toxic conditioning, as we have previously demon-strated in the murine and canine models. This strategy holds great promise in the treatment of manyhematopoietic disorders, including the hemoglobinopathies. Graft-versus-host disease (GVHD) represents thegreatest theoretical risk of IUHCT and has never been characterized in the context of IUHCT. We recently de-scribed a preclinical canine model of IUHCT, allowing further study of the technique and its complications.We aimed to establish a threshold T cell dose for IUHCT-induced GVHD in the haploidentical canine modeland to define the GVHD phenotype. Using a range of T cell concentrations within the donor inoculum, wewere able to characterize the phenotype of IUHCT-induced GVHD and establish a clear threshold for its in-duction between 3% and 5% graft CD3+ cell content. Given the complete absence of GVHD at CD3 doses of 1%to 3% and the excellent engraftment with the lowest dose, there is a safe therapeutic index for a clinical trialof IUHCT.

© 2018 American Society for Blood and Marrow Transplantation.

INTRODUCTIONIn utero hematopoietic cell transplantation (IUHCT) is a

nonmyeloablative approach capable of achieving allogeneicmixed hematopoietic chimerism and associated donor-specific tolerance (DST) without conditioning under specificexperimental circumstances [1]. We previously demon-strated that an optimized approach in the preclinical caninemodel yields stable and consistent engraftment associatedwith levels of hematopoietic chimerism sufficient for theuniform induction of DST and that are potentially therapeu-tic for many hematopoietic disorders [2]. That study confirmedin a preclinical model the potential for IUHCT in the treat-ment of many hematopoietic disorders, including thehemoglobinopathies [3-6]. However, its risks need to be wellunderstood before clinical translation. Although not ob-served in our canine model, graft-versus-host disease (GVHD)represents the greatest theoretical risk of this procedure, andthis risk has not previously been well characterized in thecontext of IUHCT.

Some murine studies have suggested that GVHD may occurafter IUHCT based on observations of decreased survival and/or upregulation of T cell regulatory populations [7], but effortsto define the relationship between IUHCT and GVHD inmurine models have been hampered by a bone marrow (BM)T cell profile requiring artificial supplementation withsplenocytes for sufficient T cell–effector function [8]. The dogmodel of GVHD has been well established in postnatal BMtransplantation (BMT) as an excellent predictor of clinicalresults [9,10], and in this study we sought to (1) determinewhether GVHD could be induced in recipients of IUHCT, (2)characterize the GVHD response in this context, and (3) definea safe T cell threshold for clinical application of IUHCT. Wenow report a canine model of IUHCT-induced GVHD depen-dent on T cell concentration within the graft, with clinicalfeatures similar to postnatal GVHD that supports a safe ther-apeutic index for clinical trials of IUHCT.

METHODSAnimals

Time-dated pregnant female beagles were purchased from Covance Re-search Products, Inc. (Cumberland, VA), which maintains a closed colony withan inbreeding coefficient of <.3%. Pregnancy was confirmed by transab-dominal ultrasound (Acuson Sequoia; Siemens, Malvern, PA) upon arrival.Animals were housed in the Laboratory Animal Facility of the Abramson Re-search Center at The Children’s Hospital of Philadelphia, which is approved

Financial disclosure: See Acknowledgments on page 1800.* Correspondence and reprint requests: Alan W. Flake, MD, Children’s

Center for Fetal Research, Children’s Hospital of Philadelphia, ARC 1116, 3615Civic Center Blvd., Philadelphia, PA 19104.

E-mail address: flake@email.chop.edu (A.W. Flake).

Biol Blood Marrow Transplant 24 (2018) 1795–1801

https://doi.org/10.1016/j.bbmt.2018.05.0201083-8791/© 2018 American Society for Blood and Marrow Transplantation.

Biology of Blood andMarrow Transplantationjournal homepage: www.bbmt.org

by the Association for Assessment and Accreditation of Laboratory AnimalCare. All experimental protocols were approved by the Institutional AnimalCare and Use Committee at The Children’s Hospital of Philadelphia and fol-lowed guidelines set forth in the National Institutes of Health Guide for theCare and Use of Laboratory Animals.

BM Harvest and ProcessingBM was harvested under sterile conditions from the pregnant mother

under general anesthesia (see Supplemental Methods). Mononuclear cellswere isolated using a Nycoprep 1.077A (Accurate Chemical, Westbury, NY)density gradient. CD3 depletion used monoclonal anti-canine CD3 CA17.2A12(AbD Serotec, Raleigh, NC). Cells were then incubated with rat anti-mouseIgG1 microbeads and sorted via AutoMACS (Miltenyi, Auburn, CA). CD26 in-hibition was performed as previously described [11] (see SupplementalMethods).

In Utero HCTCD3 selected and depleted fractions after AutoMACS processing were

analyzed for CD3 and CD34 content by flow cytometry, using anti-canineCD34PE (1H6) (BD Pharmingen, San Diego, CA) or anti-canine CD34FITC (ABDSerotec) and anti-canine CD3FITC (ABD Serotec). A portion of the CD3+ pop-ulation was added back to achieve final CD3+ concentrations of 3%, 5%, or7.5% of the total administered donor cells for each group. Additionally, 1 groupunderwent injection of whole maternal BM with a CD3+ concentration of16%. For injection, cells were resuspended in normal saline with 1% heat-inactivated donor serum, .03% DNAse, and .1% heparin. Injections wereperformed via laparotomy under ultrasound guidance, and fetal viability wasconfirmed.

Chimerism AnalysisChimerism was assessed using the variable number of tandem repeats

assay, which identifies microsatellite repeats with a sensitivity of approx-imately 2%, as previously described [2,11]. Because of overlap between donorpeak and stutter patterns, no more than 2 primers were informative for anypairing; in most cases only 1 could be used. This assay was validated againstboth quantitative Taqman PCR for the SRY gene and CD18 FACS, usinglong-term chimeras from our previous study [11]. As reported previously,variable number of tandem repeats results correlated well with bothtechniques [2].

Clinical ObservationAnimals were examined at least twice daily after birth to evaluate for

signs of GVHD, including skin lesions (flaking, erythema), weight loss, poorfeeding, and oral/mucosal lesions. For all groups tube feedings were initi-ated upon evidence of weight loss persisting on 3 subsequent measurements.All animals showing consistent signs of distress (lethargy, pain, inability totolerate oral intake) were killed in accordance with our protocol. Oral ste-roids and rapamycin were administered as deemed clinically appropriateby the supervising veterinarian. Surviving pups were observed for a periodof 15 to 22 months.

ImmunohistochemistryTissue specimens were fixed in 10% buffered formalin and embedded

in paraffin using a Sakura Tissue-Tek embedder (Sakura Finetek USA, Tor-rance, CA). Using a paraffin microtome (Leica RM2035; Instrument GmbH,Nussloch, Germany), 4-μm sections were obtained. Paraffin sections wereincubated overnight at 55°C and deparaffinized in serial xylene washes, fol-lowed by rehydration through a graded alcohol series to deionized water.Slides were blocked for specific protein for 30 minutes at room tempera-ture. For CD3 immunoperoxidase staining, mouse anti-canine CD3 (ABDSerotec) was applied as primary antibody, and slides were incubated over-night at 4°C. The slides were washed with PBS at room temperature for 10minutes, and species-specific secondary antibody was applied (anti-mouse IgG; Vector Lab, Burlingame, CA). After room temperature incubation

for 30 minutes, slides were rinsed in PBS for 10 minutes, and avidin-biotincomplex (1:200 dilution; Vector Lab) was added for 30 minutes. After a PBSrinse the slides were developed with peroxidase substrate kit SK-4100 (VectorLab), lightly stained with Harris hematoxylin, dehydrated in alcohol, clearedin xylene, and mounted using Acrymount (Statlab Medical Products, Lewis-ville, TX). Slides were imaged using a DMR microscope (Leica Microsystems,Wentzler, Germany).

StatisticsA chi-square test was applied to compare survival with birth among

groups.

RESULTSGVHD after IUHCT in the Optimized Canine Model

As per our previous study [2], we elected to target ges-tational age (GA) 40 to 42 days using intracardiacadministration of maternal BM. Our published cohort con-sisting of 29 fetuses of 6 pregnant dams served as a controlgroup at a CD3+ concentration of 1%, with absolute CD3+

counts ranging from 3.7 × 108 to 2.7 × 109/kg. In the 3% group15 fetuses of 2 pregnant dams were injected at GA 37 to 43days, with absolute CD3+ counts ranging from 4.5 × 108 to1.2 × 109/kg. Seven fetuses of a single pregnant dam were in-jected at GA 43 days using a graft containing 5% CD3+ cellswith an absolute CD3+ count of 8.75 × 108. Five fetuses of asingle pregnant dam were injected at GA 41 days using a CD3+

concentration of 7.5%, with an absolute CD3+ count of2.25 × 109/kg, and 9 fetuses of a single pregnant dam wereinjected at GA 43 days with whole maternal BM at a CD3+

concentration of 16%, resulting in an absolute CD3+ count of2.0 × 109/kg (Table 1). Absolute CD3+ counts were determinedbased on estimation of fetal weight using ultrasound mea-surement of crown–rump length at the time of injection [12],resulting in some overlap in absolute counts between groups.

Differences in survival to birth between groups did notreach statistical significance (chi-square statistic, 4.141;P = .39), and no mucosal or skin lesions or other evidence ofGVHD was clinically evident at birth in any animal. Our 1%cohort had 86% survival to birth (25/29), as previously re-ported [2], which compares favorably with our historicalintraperitoneally injected animals [11] and survival rates re-ported previously by Blakemore et al. [13]. Fetuses neitherlive nor stillborn were likely resorbed after the injection pro-cedure, which has been reported to occur at a rate of 11% to25% during normal canine gestation [14,15], and thus nopathologic evaluation could be conducted. In the 3% group9 of 14 pups were live born (65%), and an additional pup wasstillborn because of birth complications but without evi-dence of GVHD on pathologic evaluation. The 5% group yielded5 of 8 live-born pups (63%), whereas 3 of 5 pups in the 7.5%group were live born after a difficult labor lasting > 21 hoursand assisted by oxytocin for failure to progress. One addi-tional pup in this litter was stillborn because of the prolongedlabor but without clinical or pathologic abnormality. In the

Table 1Injection Summary and Graft Composition in All Groups

Dose No. of Live-Born Injected Pups GA at IUHCT(days)

CD3 cells/kg CD34 cells/kg GVHD

1% 20 39-41 3.71-27.2 × 108 5.8-58.2 × 108 N3% 8 38-43.5 4.5-12.6 × 108 5.0-9.8 × 108 N5% 5 43 8.75 × 108 5.39 × 108 Y7.5% 3 41.5 2.26 × 109 7.0 × 108 Y16% (Whole BM) 7 43 2.0 × 109 5.1 × 108 Y

Significant overlap in CD3 cells/kg suggests that the concentration, rather than the absolute CD3 content of the donor graft, is a significant contributing factorto the development of GVHD.

1796 J.D. Vrecenak et al. / Biol Blood Marrow Transplant 24 (2018) 1795–1801

Whole BM group 7 of 9 fetuses were live born (78%), al-though 1 was runted. No premature rupture of membranesor preterm delivery was observed after IUHCT, consistent withthe very low rates observed with clinical intrauterine trans-fusion [16].

Acute GVHD was observed in all recipients of 5% and 7.5%CD3+ cells as well as all pups receiving Whole BM (16%).Symptoms became evident between 1 and 14 days post-birth, and all pups ultimately were killed by 23 days of age(Figure 1). Signs of GVHD included weight loss and failure tothrive despite enteral supplementation, which were univer-sally evident among affected pups (Figure 2A,B). Both skin andmucosal lesions were observed clinically in selected pups(Figure 2C). Although jaundice was observed, gut manifes-tations of GVHD were difficult to confirm clinically becauseof maternal cleaning and a wide range of “normal” outputin breastfed animals. Laboratory evaluation revealed signif-icant anemia and eosinophilia in most affected animals as wellas frequent hyperbilirubinemia and transaminitis. Patholog-ic examination revealed punctate gastrointestinal hemorrhage(Figure 3A) and gross hepatomegaly as well as hemorrhagiclesions throughout the lungs (Figure 3B). Histologic evalua-tion showed lymphocytic infiltration and CD3 activity withinthe skin (Figure 4A,B), bowel wall (Figure 4C,D), canalicularbile stasis in liver specimens (Figure 4E), and inflammatoryhemorrhage within the lungs (Figure 4F). No evidence ofGVHD was observed in recipients of 1% or 3% CD3+ contain-ing grafts.

Chimerism after IUHCTPeripheral blood samples for chimerism were obtained at

1 month of age or when affected animals were killed, becauseall were less than 1 month of age at the time of death. Nosignificant differences were observed in initial chimerismbetween groups (Figure 5A), although the varying ages atwhich samples could be obtained confounds any directcomparison. One pup in the 3% group never showed anymeasureable chimerism, likely because of technical failureat injection. Otherwise, the 3% group maintained stable chi-merism over the period of analysis (15 to 22 months,Figure 5B), consistent with our prior results.

DISCUSSIONIUHCT represents a promising strategy to achieve stable

multilineage engraftment in an immunocompetent hostwithout the need for toxic conditioning and has potential forclinical application in any hematopoietic disorder that canbe diagnosed early in gestation and is treatable by postna-tal BMT. We have previously investigated the requirementsfor successful IUHCT extensively in the murine model [17-22]and gained better understanding of the barriers to engraft-ment [23,24], whereas large animal studies in sheep [25], pigs[26], and nonhuman primates [27] have yielded inconsistentresults. We have reported the development of an optimizedcanine model yielding stable and consistent engraftment atdonor cell levels associated with DST in nearly all recipi-ents of haploidentical IUHCT using maternal donor cells [2]

Figure 1. Survival curve for all groups. All pups affected by GVHD died by 23 days of age, whereas all pups in the 1% and 3% groups appeared normal through-out the period of analysis.

1797J.D. Vrecenak et al. / Biol Blood Marrow Transplant 24 (2018) 1795–1801

to avoid the confounding effects of maternal immunization[24]. Because IgG is known to cross the placenta in both dogsand humans, the use of maternal donor cells in this modelis necessary due to the risk of maternal immunization againstthird-party donor cells. Although this has obvious implica-tions in the treatment of inherited hematologic disease, theexpectation that maternal antibodies could limit donor chi-merism has significant experimental support [24,28]. Theseresults represent a significant step toward adequate exper-imental justification for clinical trials of IUHCT, but a betterunderstanding of the associated risks is necessary before clin-ical translation.

Beyond failure of engraftment and subsequent lack of ef-ficacy, GVHD represents the greatest theoretical risk of IUHCT.Based on Billingham’s tenets, the fetus should be at high riskfor GVHD, because IUHCT grafts contain immunologicallycompetent cells, the recipients express tissue antigens not

present in the donor, and the recipients are incapable ofmounting an effective response to eliminate the trans-planted cells [29]. Despite these risk factors, GVHD has provendifficult to study in the context of IUHCT in no small part dueto the lack of a consistent and valid preclinical large animalmodel. Despite MHC mismatch, mouse BM contains a rela-tively low T cell concentration and generally requiressupplementation with a T cell source such as splenocytes tohave any graft-versus-host activity [8,30] due to the pres-ence of natural killer T cell subsets with protective activity[31] and T cell regulatory suppression [32-34]. Althoughmurine models have proven valuable in the mechanistic studyof GVHD associated with postnatal BMT, most reports ofIUHCT-induced GVHD have been characterized primarily bydecreased survival [7,35-37]. In large animal models links havebeen proposed between in utero or perinatal lethality andGVHD in sheep [38,39] and nonhuman primates [27,40,41],

Figure 2. Clinical manifestations of GVHD. (A) Weight gain across 1%, 3%, 7.5%, and 15% groups. Failure to thrive occurred in all GVHD pups despite enteralnutritional supplements and steroid and rapamycin treatment. The 5% group was excluded from this analysis because these pups were killed upon emer-gence of GVHD symptoms, whereas other groups were treated clinically and observed throughout the disease course. (B) Failure to thrive manifest as growthretardation in a pup affected by GVHD (white arrow) compared with littermates who had not yet exhibited symptoms. (C) Skin lesions characteristic of IUHCT-induced GVHD.

Figure 3. Gross pathologic evidence of GVHD. (A) Punctate hemorrhages throughout the gastrointestinal tract of a GVHD-affected pup (arrows). (B) Hemor-rhagic lesions throughout the lungs of a GVHD-affected pup.

1798 J.D. Vrecenak et al. / Biol Blood Marrow Transplant 24 (2018) 1795–1801

but no clinical or histologic confirmation has been re-ported. No GVHD has previously been observed after IUHCTin dog [2,11,13,42] or pig [26] models, underscoring the needfor better characterization of this risk.

Despite the theoretical risk factors, our experience overmany years of experimental work in IUHCT has suggested thatthe fetus may actually be surprisingly resistant to GVHD rel-ative to recipients of postnatal myeloablative BMT. We havepreviously speculated that such resistance may be related toa robust host T cell–regulatory response from the fetal thymus[7,24,43]. In our characterization of GVHD, the most strik-ing finding was the clear threshold between 3% and 5%, abovewhich GVHD was observed in all recipients. In contrast, nearlyall animals injected with grafts containing either 1% or 3%CD3+ cells engrafted and maintained chimerism throughout

the duration of the study; none developed any clinical signof GVHD. Our results mirror those obtained in sheep byCrombleholme et al. [38], who observed no GVHD until greaterthan 3% of donor cells were T cells. Furthermore, we foundthat the development of GVHD was more clearly related toT cell concentration than to the absolute CD3+ dose per weight,although a limitation of the study is the dependence on ac-curate ultrasound-based fetal weight estimation for calculationof absolute T cell dose.

Mechanistically, this concentration threshold implies thatthe development of GVHD depends on some interplaybetween donor T cells and some other donor cell subset withinthe donor cell inoculum. Given the inability to differentiatebetween donor and recipient by flow cytometry, the caninemodel is ill-suited to further explore the donor cell type(s)

Figure 4. Histologic evidence of GVHD. (A) Hematoxylin and eosin (H&E) and (B) anti-CD3 immunoperoxidase staining of GVHD-affected skin (10×) demon-strating breakdown of normal barrier function and inflammatory ulceration with CD3 infiltration (black arrow). (C) H&E and (D) anti-CD3 immunoperoxidasestaining of GVHD-affected intestine (10×) demonstrating blunted villous architecture, sloughing (white arrow) and CD3 infiltration (black arrows). (E) H&E stain-ing of liver demonstrating canalicular bile stasis (white arrows) (10×). (F) H&E staining demonstrating pulmonary inflammation and hemorrhage (black arrow)(10×).

Figure 5. Chimerism after IUHCT. Initial chimerism after IUHCT by group (A) and long-term stable engraftment (B) as measured by variable number of tandemrepeats. No significant differences in initial chimerism could be discerned because of the limited sample size.

1799J.D. Vrecenak et al. / Biol Blood Marrow Transplant 24 (2018) 1795–1801

involved in the regulation of GVHD, but extrapolation ofobservations made in postnatal BMT would suggest cross-presentation by donor antigen-presenting cells as a likely co-stimulatory candidate, allowing higher T cell concentrationsto overwhelm the suppressive effect of host regulatory T cells[44-47]. In the treatment of nonmalignant disorders, the oc-currence of GVHD by any mechanism would represent anunacceptable clinical outcome. Our results in this preclini-cal model confirm that GVHD is indeed a risk of IUHCT, butonly at relatively high T cell concentrations, providing supportfor a clear margin of safety for clinical application.

Interestingly, we saw no evidence of prenatal lethality atany T cell concentration, based on the lack of significant dif-ference in survival to birth. Although the onset of GVHD wasrapid in all cases, no affected animal had obvious evidenceof any GVHD-related pathology at birth apart from a singlerunted pup in a very large litter of 9. These results contrastthose previously reported in large animal models, includ-ing sheep. The lack of in utero manifestations might be relatedto the relatively short gestation of the canine model, to pro-tective factors associated with the intrauterine environment[48], or to a lack of sufficient inflammatory stimulus in theabsence of conditioning or any bacterial exposure.

Notably, our model of IUHCT-induced GVHD shares manyclinical features with classical descriptions of GVHD, includ-ing examples of skin, gut, and liver involvement [49,50].Weight loss and wasting were primary features of IUHCT-induced GVHD and were perhaps more evident in ournonconditioned model than in the setting of postnatal trans-plants wherein multiple confounding factors might contributeto such findings. Our results echo those observed in themurine model of GVHD proposed by Misra et al. [8]. Al-though GVHD is primarily a clinical diagnosis, laboratoryresults in the affected animals provided additional supportfor our clinical findings; anemia was nearly universal, andmost also had evidence of hepatic dysfunction. The etiolo-gy of the eosinophilia observed in affected animals is not clear,although eosinophilia has been reported to occur frequentlyin chronic GVHD after nonmyeloablative transplantation andhas not been shown to correlate with outcome [51]. Becauseour affected pups presented with a clinical course more similarto acute GVHD, the question of whether the presence of eo-sinophilia in GVHD-affected animals after IUHCT has anymechanistic implications requires further study.

Although GVHD is possible after IUHCT, our results confirmthat its development can be prevented through the use oflower T cell concentrations within the donor graft while pre-serving engraftment sufficient for induction of DST andpotentially therapeutic for many target disorders. In creat-ing a model of IUHCT-induced GVHD for which the clinicalfeatures show marked similarity to the constellation of symp-toms and signs observed in GVHD associated with postnatalBMT, we have both defined the GVHD response and con-firmed its absence at the T cell concentrations proposed forclinical IUHCT. Having established the relevance of this pre-clinical model for IUHCT-induced GVHD, we plan to evaluatefuture modifications of the donor inoculum intended toenhance engraftment for GVHD risk before clinical applica-tion. Importantly, we have demonstrated a safe therapeuticindex such that even a 3-fold increase over planned T cellconcentrations did not show evidence of GVHD in ourpreclinical model. In combination with our prior demonstra-tion of long-term stable hematopoietic chimerism in thepreclinical canine model, these results provide necessary char-acterization of the greatest theoretical risk of IUHCT and

support its translation into clinical trials for inherited he-matologic disorders.

ACKNOWLEDGMENTSThe authors thank Aaron Weilerstein and Laboratory

Animal Services staff for their assistance in animal care andJane Wright for her sonographic expertise.

Financial disclosure: This work was supported by charita-ble contributions from the Albert M. Greenfield Jr. Foundationand the Ruth and Tristram C. Colket Jr. Endowed Chair (toA.W.F.)

Conflict of interest statement: There are no conflicts of in-terest to report.

REFERENCES1. Vrecenak JD, Flake AW. In utero hematopoietic cell

transplantation—recent progress and the potential for clinical application.Cytotherapy. 2013;15:525-535.

2. Vrecenak JD, Pearson EG, Santore MT, et al. Stable long-term mixedchimerism achieved in a canine model of allogeneic in uterohematopoietic cell transplantation. Blood. 2014;124:1987-1995.

3. Flake AW, Zanjani ED. Treatment of severe combined immunodeficiency.N Engl J Med. 1999;341:291-292.

4. Flake AW, Zanjani ED. In utero hematopoietic stem cell transplantation.A status report. JAMA. 1997;278:932-937.

5. Fleischman RA, Mintz B. Prevention of genetic anemias in mice bymicroinjection of normal hematopoietic stem cells into the fetal placenta.Proc Natl Acad Sci USA. 1979;76:5736-5740.

6. Surbek D, Schoeberlein A, Wagner A. Perinatal stem-cell and genetherapy for hemoglobinopathies. Semin Fetal Neonatal Med. 2008;13:282-290.

7. Hayashi S, Hsieh M, Peranteau WH, Ashizuka S, Flake AW. Completeallogeneic hematopoietic chimerism achieved by in utero hematopoieticcell transplantation and cotransplantation of LLME-treated, MHC-sensitized donor lymphocytes. Exp Hematol. 2004;32:290-299.

8. Misra MV, Gutweiler JR, Suh MY, et al. A murine model of graft-vs-hostdisease after in utero hematopoietic cell transplantation. J Pediatr Surg.2009;44:1102-1107, discussion 1107.

9. Storb R, Thomas ED. Graft-versus-host disease in dog and man: theSeattle experience. Immunol Rev. 1985;88:215-238.

10. van Bekkum DW. Biology of acute and chronic graft-versus-hostreactions: predictive value of studies in experimental animals. BoneMarrow Transplant. 1994;14(suppl 4):S51-S55.

11. Peranteau WH, Heaton TE, Gu YC, et al. Haploidentical in uterohematopoietic cell transplantation improves phenotype and can inducetolerance for postnatal same-donor transplants in the canine leukocyteadhesion deficiency model. Biol Blood Marrow Transplant. 2009;15:293-305.

12. Yeager AE, Mohammed HO, Meyers-Wallen V, Vannerson L, ConcannonPW. Ultrasonographic appearance of the uterus, placenta, fetus, and fetalmembranes throughout accurately timed pregnancy in beagles. Am J VetRes. 1992;53:342-351.

13. Blakemore K, Hattenburg C, Stetten G, et al. In utero hematopoietic stemcell transplantation with haploidentical donor adult bone marrow in acanine model. Am J Obstet Gynecol. 2004;190:960-973.

14. Ortega-Pacheco A, Rodriguez-Buenfil JC, Segura-Correa JC,Montes de Oca-Gonzalez AR, Jimenez-Coello M. Prevalence of fetalresorption in stray dogs in Yucatan, Mexico. J Small Anim Pract.2006;47:266-269.

15. England GC. Ultrasonographic assessment of abnormal pregnancy. VetClin North Am Small Anim Pract. 1998;28:849-868.

16. Lindenburg IT, van Kamp IL, Oepkes D. Intrauterine blood transfusion:current indications and associated risks. Fetal Diagn Ther. 2014;36:263-271.

17. Hayashi S, Abdulmalik O, Peranteau WH, et al. Mixed chimerismfollowing in utero hematopoietic stem cell transplantation in murinemodels of hemoglobinopathy. Exp Hematol. 2003;31:176-184.

18. Hayashi S, Peranteau WH, Shaaban AF, Flake AW. Complete allogeneichematopoietic chimerism achieved by a combined strategy of in uterohematopoietic stem cell transplantation and postnatal donor lymphocyteinfusion. Blood. 2002;100:804-812.

19. Kim HB, Shaaban AF, Yang EY, Liechty KW, Flake AW. Microchimerismand tolerance after in utero bone marrow transplantation in mice. J SurgRes. 1998;77:1-5.

20. Kim HB, Shaaban AF, Milner R, Fichter C, Flake AW. In utero bone marrowtransplantation induces donor-specific tolerance by a combination ofclonal deletion and clonal anergy. J Pediatr Surg. 1999;34:726-729,discussion 729-730.

21. Peranteau WH, Hayashi S, Hsieh M, Shaaban AF, Flake AW. High-levelallogeneic chimerism achieved by prenatal tolerance induction and

1800 J.D. Vrecenak et al. / Biol Blood Marrow Transplant 24 (2018) 1795–1801

postnatal nonmyeloablative bone marrow transplantation. Blood.2002;100:2225-2234.

22. Peranteau WH, Endo M, Adibe OO, Merchant A, Zoltick PW, Flake AW.CD26 inhibition enhances allogeneic donor-cell homing and engraftmentafter in utero hematopoietic-cell transplantation. Blood. 2006;108:4268-4274.

23. Peranteau WH, Endo M, Adibe OO, Flake AW. Evidence for an immunebarrier after in utero hematopoietic-cell transplantation. Blood.2007;109:1331-1333.

24. Merianos DJ, Tiblad E, Santore MT, et al. Maternal alloantibodies inducea postnatal immune response that limits engraftment following in uterohematopoietic cell transplantation in mice. J Clin Invest. 2009;119:2590-2600.

25. Flake AW, Harrison MR, Adzick NS, Zanjani ED. Transplantation of fetalhematopoietic stem cells in utero: the creation of hematopoieticchimeras. Science. 1986;233:776-778.

26. Lee PW, Cina RA, Randolph MA, et al. Stable multilineage chimerismacross full MHC barriers without graft-versus-host disease following inutero bone marrow transplantation in pigs. Exp Hematol. 2005;33:371-379.

27. Shields LE, Gaur LK, Gough M, Potter J, Sieverkropp A, Andrews RG. Inutero hematopoietic stem cell transplantation in nonhuman primates:the role of T cells. Stem Cells. 2003;21:304-314.

28. Nijagal A, Wegorzewska M, Jarvis E, Le T, Tang Q, MacKenzie TC. MaternalT cells limit engraftment after in utero hematopoietic cell transplantationin mice. J Clin Invest. 2011;121:582-592.

29. Billingham RE. The biology of graft-versus-host reactions. Harvey Lect.1966;62:21-78.

30. Schroeder MA, DiPersio JF. Mouse models of graft-versus-host disease:advances and limitations. Dis Model Mech. 2011;4:318-333.

31. Zeng D, Lewis D, Dejbakhsh-Jones S, et al. Bone marrow NK1.1(-) andNK1.1(+) T cells reciprocally regulate acute graft versus host disease.J Exp Med. 1999;189:1073-1081.

32. Kohrt HE, Pillai AB, Lowsky R, Strober S. NKT cells, Treg, and theirinteractions in bone marrow transplantation. Eur J Immunol.2010;40:1862-1869.

33. Pillai AB, George TI, Dutt S, Strober S. Host natural killer T cells inducean interleukin-4-dependent expansion of donor CD4+CD25+Foxp3+ Tregulatory cells that protects against graft-versus-host disease. Blood.2009;113:4458-4467.

34. Nguyen VH, Zeiser R, Negrin RS. Role of naturally arising regulatory Tcells in hematopoietic cell transplantation. Biol Blood Marrow Transplant.2006;12:995-1009.

35. Chou SH, Chawla A, Lee TH, et al. Increased engraftment and GVHD afterin utero transplantation of MHC-mismatched bone marrow cells andCD80low, CD86(-) dendritic cells in a fetal mouse model. Transplantation.2001;72:1768-1776.

36. Chen JC, Ou LS, Yu HY, Chang HL, Chang PY, Kuo ML. Allogeneiclymphocytes exerted graft-versus-host rather than tolerogenic effectson preimmune fetuses. J Surg Res. 2013;183:405-411.

37. Bhattacharyya S, Chawla A, Smith K, et al. Multilineage engraftment withminimal graft-versus-host disease following in utero transplantation ofS-59 psoralen/ultraviolet a light-treated, sensitized T cells and adult Tcell-depleted bone marrow in fetal mice. J Immunol. 2002;169:6133-6140.

38. Crombleholme TM, Harrison MR, Zanjani ED. In utero transplantationof hematopoietic stem cells in sheep: the role of T cells in engraftmentand graft-versus-host disease. J Pediatr Surg. 1990;25:885-892.

39. Zanjani ED, Lim G, McGlave PB, et al. Adult haematopoietic cellstransplanted to sheep fetuses continue to produce adult globins. Nature.1982;295:244-246.

40. Roodman GD, Vandeberg JL, Kuehl TJ. In utero bone marrowtransplantation of fetal baboons with mismatched adult marrow: initialobservations. Bone Marrow Transplant. 1988;3:141-147.

41. Roodman GD, Kuehl TJ, Vandeberg JL, Muirhead DY. In utero bonemarrow transplantation of fetal baboons with mismatched adult baboonmarrow. Blood Cells. 1991;17:367-375.

42. Petersen SM, Gendelman M, Murphy KM, et al. Use of T-cell antibodiesfor donor dosaging in a canine model of in utero hematopoietic stemcell transplantation. Fetal Diagn Ther. 2007;22:175-179.

43. Nijagal A, Derderian C, Le T, et al. Direct and indirect antigen presentationlead to deletion of donor-specific T cells after in utero hematopoieticcell transplantation in mice. Blood. 2013;121:4595-4602.

44. Wang X, Li H, Matte-Martone C, et al. Mechanisms of antigenpresentation to T cells in murine graft-versus-host disease: cross-presentation and the appearance of cross-presentation. Blood. 2011;118:6426-6437.

45. Shlomchik WD. Graft-versus-host disease. Nat Rev Immunol. 2007;7:340-352.

46. Hoffmann P, Ermann J, Edinger M, Fathman CG, Strober S. Donor-typeCD4(+)CD25(+) regulatory T cells suppress lethal acute graft-versus-hostdisease after allogeneic bone marrow transplantation. J Exp Med.2002;196:389-399.

47. Schneidawind D, Pierini A, Negrin RS. Regulatory T cells and natural killerT cells for modulation of GVHD following allogeneic hematopoietic celltransplantation. Blood. 2013;122:3116-3121.

48. Erkers T, Nava S, Yosef J, Ringden O, Kaipe H. Decidual stromal cellspromote regulatory T cells and suppress alloreactivity in a cell contact-dependent manner. Stem Cells Dev. 2013;22:2596-2605.

49. Holtan SG, Pasquini M, Weisdorf DJ. Acute graft-versus-host disease:a bench-to-bedside update. Blood. 2014;124:363-373.

50. Shulman HM, Kleiner D, Lee SJ, et al. Histopathologic diagnosis of chronicgraft-versus-host disease: National Institutes of Health consensusdevelopment project on criteria for clinical trials in chronic graft-versus-host disease. II. Pathology Working Group report. Biol Blood MarrowTransplant. 2006;12:31-47.

51. Ahmad I, Labbe AC, Chagnon M, et al. Incidence and prognosticvalue of eosinophilia in chronic graft-versus-host disease afternonmyeloablative hematopoietic cell transplantation. Biol Blood MarrowTransplant. 2011;17:1673-1678.

1801J.D. Vrecenak et al. / Biol Blood Marrow Transplant 24 (2018) 1795–1801