Minor Histocompatibility Antigen UTY as Target for Graft-versus-Leukemia and Graft-versus-Haematopoiesis in the Canine Model


  • D. Bund,

    Corresponding author
    1. Helmholtz Center Munich, German Research Center for Environmental Health, Clinical Cooperative Group Hematopoietic Cell Transplantation (CCG-HCT), Munich, Germany
    • Medical Department III, University Hospital Großhadern, Ludwig-Maximilians-University, Munich, Germany
    Search for more papers by this author
  • R. Buhmann,

    1. Medical Department III, University Hospital Großhadern, Ludwig-Maximilians-University, Munich, Germany
    2. Helmholtz Center Munich, German Research Center for Environmental Health, Clinical Cooperative Group Hematopoietic Cell Transplantation (CCG-HCT), Munich, Germany
    Search for more papers by this author
  • F. Gökmen,

    1. Medical Department III, University Hospital Großhadern, Ludwig-Maximilians-University, Munich, Germany
    2. Helmholtz Center Munich, German Research Center for Environmental Health, Clinical Cooperative Group Hematopoietic Cell Transplantation (CCG-HCT), Munich, Germany
    Search for more papers by this author
  • J. Zorn,

    1. Medical Department III, University Hospital Großhadern, Ludwig-Maximilians-University, Munich, Germany
    2. Helmholtz Center Munich, German Research Center for Environmental Health, Clinical Cooperative Group Hematopoietic Cell Transplantation (CCG-HCT), Munich, Germany
    Search for more papers by this author
  • H.-J. Kolb,

    1. Medical Department III, University Hospital Großhadern, Ludwig-Maximilians-University, Munich, Germany
    2. Helmholtz Center Munich, German Research Center for Environmental Health, Clinical Cooperative Group Hematopoietic Cell Transplantation (CCG-HCT), Munich, Germany
    Search for more papers by this author
    • These authors contributed equally to this work.
  • H. M. Schmetzer

    1. Medical Department III, University Hospital Großhadern, Ludwig-Maximilians-University, Munich, Germany
    Search for more papers by this author
    • These authors contributed equally to this work.

Correspondence to: D. Bund, Helmholtz Center Munich - German Research Center for Environmental Health, Clinical Cooperative Group Hematopoietic Cell Transplantation (CCG-HCT), Marchioninistr. 25, 81377 Munich, Germany. E-mail: dagmar_bund@gmx.de


Male patients with female-stem-cell donors have better prognosis compared to female-to-male combinations due to Y-encoded minor histocompatibility antigens recognized by female-alloimmune-effector lymphocytes in the context of a graft-versus-leukemia (GvL) effect. We provide data in a dog-model that the minor histocompatibility antigen UTY might be a promising target to further improve GvL-immune reactions after allogeneic-stem-cell transplantations. Female-canine-UTY-specific T cells (CTLs) were stimulated in vitro using autologous-DCs loaded with three HLA-A2-restricted-UTY-derived peptides (3-fold-expansion), and specific T cell responses were determined in 3/6 female dogs. CTLs specifically recognized/lysed autologous-female-peptide-loaded DCs, but not naïve-autologous-female DCs and monocytes. They mainly recognized bone-marrow (BM) and to a lower extent DCs, monocytes, PBMCs and B-cells from DLA-identical-male littermates and peptide-loaded T2-cells in an MHC-I-restricted manner. A UTY-/male-specific reactivity was also obtained in vivo after stimulation of a female dog with DLA-identical-male PBMCs. In summary, we demonstrated natural UTY processing and presentation in dogs. We showed that female-dog CTLs were specifically stimulated by HLA-A2-restricted-UTY peptides, thereby enabling recognition of DLA-identical-male cells, mainly BM cells. These observations suggest UTY as a promising candidate-antigen to improve GvL-reactions in the course of immunotherapy.


Allogeneic-stem-cell transplantation (alloSCT) represents the only curative therapy for many patients with haematological-malignancies including leukemia. The therapeutic-effect is mediated by donor-derived immune-effector cells infused with donor-lymphocyte transfusion (DLT) after transplantation. This approach is successful in treating relapsed myeloid-malignancies [1]. The favourable graft-versus-leukemia (GvL) effect of donor-lymphocytes is mainly mediated by allo-reactive T cells recognizing antigens (Ags) on hematopoietic-cells including the malignant leukemic-cells of the patients [2, 3]. These T cells can also be reactive towards healthy-tissues and cause graft-versus-host-disease (GvHD) [4, 5]. Own clinical observations demonstrated that in haploidentical-transplantations female-donors (especially mothers) show a higher GvL-effect against male-recipients (particularly sons) compared to all other haploidentical donor-recipient combinations [6, 7] (H. J. Kolb, unpublished data). These reactions might be due to the existence of male-associated antigens [8]. The Y-chromosome coded minor histocompatibility antigen (mHA) UTY (ubiquitously-transcribed-tetratrico-peptide-repeat-gene, Y-linked) could be a new immunotherapeutically useful potential candidate-target structure [8, 9]. Vaccination of patients after transplantation or vaccination of stem-cell donors before transplantation using mHA-specific-peptides, production of mHA-specific T cells and redirection of T cell specificity by gene-transfer of T cell receptors may be strategies to specifically eradicate the malignant cells after alloSCT.

Expression pattern and tissue restriction of antigens are essential for the clinical outcome of adoptive immunotherapy. Broadly expressed antigens cause not only T cell responses mediating the GvL-effect, but also GvHD. mHAs being expressed on hematopoietic-cells are representing the best antigens for GVL-reactions as T cells recognizing mHAs may mostly eliminate recipients'haematopoietic-cells including the malignant cells, without affecting donor-haematopoiesis or normal non-haematopoietic tissues [10]. Most Y-chromosome-coded proteins/mHAs show only few expression/presentation differences between donor and recipient and have a broad tissue-expression including UTY which is weakly expressed on non-hematopoietic cells and highly expressed on hematopoietic cells [11, 12]. The preferential immune recognition of male-cells may be caused by UTY-overexpression or -altered processing recognized by female-donor cells [9]. Therefore anti-UTY-specific T cell reactions after SCT or in the context of DLT might be a promising approach to improve GvL-reactions [6].

The UTY-gene and its X-chromosome-coded homologue UTX belong to the UTX/UTY-family [13]. UTY encodes a tetratricopeptide-repeat (TPR) protein with eight TPR-motifs and one JmjC-domain. TPR-motifs are believed to mediate protein-protein interactions. Some representatives of the JmjC-protein family have histone-demethylase properties and are involved in chromatin reorganization. For UTX, a regulating role in HOX-genes was reported implicating a function in development with nuclear subcellular localization [14]. UTX, in comparison to UTY, is involved in animal morphogenesis, as no enzymatic-demethylase activity was detectable for UTY [15]. For UTY, a nucleic-localization was determined but data according to its function are still lacking [16]. Moreover, a differential-expression profile of UTY and UTX was suggested [17].

For the human-(h)-UTY, different CTL-epitopes were identified being leukemia-associated and HLA-B8-, HLA-B60- and HLA-B52-restricted [12, 18, 19]. A promising way to treat (relapsed)-leukemia was shown to be provided by adoptive-immunotherapy via CTLs in allogeneic-chimeras [20]. Great progress in transplantation-biology has been derived from canine-(c)-preclinical-studies. Adoptive immunotherapy with DLT was developed by our group in a dog-model: Tolerance was induced by transplanting dogs with T cell-depleted stem cells from dog-leukocyte-antigen-(DLA)-identical littermates followed by DLT 61/62 days later. This enabled a conversion of a mixed-chimerism to full-donor type without inducing GvHD [21]. Before a clinical-application of UTY-specific T cells to patients, this strategy should be tested in dogs as a preclinical-large-animal model.

As the canine-UTY sequences were not available at the time-point of our study (and in conse-quence no canine-peptides), we decided to use peptides derived from the human-UTY-sequence. Experimental data of other groups have not only demonstrated homology between human-HLA- and canine-DLA-sequences [22-24], but also that human-peptides can bind to canine-DLA [23, 25-30]. Although MHC-class-I-clusters have been demonstrated as partially divergent between human and canines (conservation in DLA-B and -C, but divergence in DLA-A [24]), the DLA has a multiple number of class-I-genes characterized by moderate levels of polymorphism, thereby encoding functional class-I-antigens [24, 31, 32]. Furthermore, the potential peptide-restriction of UTY to one or more DLA-class-I-molecules can be predicted.

The clinical observation of a better outcome and prognosis for male patients transplanted with female transplants prompted us to hypothize an improved GvL-effect against male-recipient cells caused by anti-male-specific antigen reactions. Here, we wanted to address following questions in the dog-model: Is it possible to (1) induce an improved GvLT cell response in a female-cellular system by pulsing female-DCs with UTY-derived male-antigens? (2) generate canine-UTY-specific T cells to characterize the functional-repertoire and Y-restriction of these T cells to increase GvL-specificity by adding DLA-identical-male-cells? (3) What is the potential of UTY-derived peptides to induce a specific GvL-effect (graft-versus-male-haematopoiesis effect)?

Materials and methods

Animals, cell-lines, peripheral-blood and bone-marrow samples

Fifteen 3–6 year-old purebred beagles were used (Table 1). Animals were housed and cared for in the facilities of the Helmholtz Center Munich (Neuherberg, Germany). Dogs were healthy, regularly de-wormed and vaccinated against distemper, leptospirosis, parvovirus and canine-hepatitis. DLA-typing was performed by using MLRs and MHC-I- and MHC-II-loci microsatellite-PCR: two dogs were defined as DLA-identical if both showed the same fragments in the microsatellite-PCR/MLRs [33]. All animal-experiments were in compliance with protocols approved by the local Animal-Care and Use-Committee.

Table 1. Animal characteristics
Dog ID #DLA-typeGender



  1. f, female; m, male.

  2. Shown are the dogs included in this study, their DLA-type, gender and age. DLA-identical dogs are clustered together which are siblings of the same littermate. DLA-types indicated as the parental haplo-types (DLA-A (MHC-I) - DLA-B (MHC-II)/DLA-A (MHC-I) - DLA-B (MHC-II)).


Peripheral-blood was sampled by venipuncture. PBMCs were separated over Ficoll-Hypaque (Biochrom, Berlin, Germany), washed twice and kept in serum-free X-Vivo15-Medium (BioWhittaker, Walkersville, MD, USA). A normal, healthy composition of dog-blood cells contained on average 13 % B-cells (range: 5–34%), 36% CD3+ T cells (range: 22.4–48.4%) and 10% monocytes (range: 4.3–23.9%) in the mononuclear fraction.

Monocytes were gathered from the isolated PBMC-fraction by adherence to plastic-flasks bottoms in RPMI1640 with 10% dog-serum (PAN-Biotech, Aidenbach, Germany) for 2 h (38 °C, 5%CO2). Supernatant was removed and collected. Adherent cells were scraped-off, washed twice (PBS) and resuspended in X-Vivo15. Supernatant fraction was used to gain B-cells using MACS-separation technology (Miltenyi-Biotech, Bergisch Gladbach, Germany) according to the manufacturers' instructions, using canine-CD21-PE-antibody (Serotec, D FCsseldorf, Germany) and Anti-PE-beads (Miltenyi, Bergisch Gladbach, Germany).

Bone-marrow samples were aspirated from the dogs' iliac-crests under general anaesthesia and bone-marrow-mononuclear cells were isolated corresponding to canine-PBMCs.

Human T2-cells (HLA-A2+, no endogenous MHC-I-peptide loading/presentation due to TAP-deficiency [34]), provided by Dr. Elfriede Nössner, Helmholtz Center Munich) were maintained in culture as recommended by ATCC (Rockville-USA).


HLA-A2-binding peptides of hUTY-sequence were identified using the publicly available peptide-motif-scoring systems http://www.bimas.cit.nih.gov/molbio/hla_bind/ and http://www.syfpeithi.de. Their potential natural-processing by proteasomal-cleavage was checked using http://www.paproc.de. Following nonameric-peptides were defined: W248: WMHHNMDLV; T368: TLAARIKFL; K1234: KLFEMIKYC. As controls we used I540S (HFLLWKLIA; non-HLAA0201-binding [35]), a MAGE-3-derived-(MAGE-3: FLWGPRALV [36]) and an influenza-matrix-protein-derived, HLA-A2-binding peptide (IMP: GILGFVFTL [37]). Peptides were synthesized and purified by Peptide-Specialty-Laboratories-GmbH (Heidelberg, Germany; Dr. H.R. Rackwitz) and dissolved in DMSO (10 mg/ml). In an HLA-A2-T2-binding assay [38], MAGE-3, IMP and all UTY-derived-peptides efficiently bound to the hHLA-A2-molecule (data not shown).

Binding of the HLA-A2-restricted hUTY-derived peptides to canine-DLA molecules was verified by testing the reactivity of female-canine-UTY-primed effector T cells (CTLs) against hUTY-peptides loaded on cDLA (DCs; n = 3). To exclude unspecific-reactions, autologous-female cells (DCs, monocytes) were used as controls (see 'Generation of UTY-specific-CTL responses in vitro using peptide-pulsed-autologous-female DCs (APCs)'). Only DCs presenting the loaded hUTY-peptides by cDLA were targeted specifically indicating the presence/recognition of the hUTY-peptide sequences in the DLA-system. As controls for male-specific reactivity and the presence of hUTY-derived peptides in the canine-DLA-context, different male-cell types were investigated (see 'Generation of UTY-specific-CTL responses in vitro using peptide-pulsed-autologous-female DCs (APCs)') showing natural presentation of the chosen hUTY-peptides in the dog via cDLA.

Generation of canine-DCs

PBMCs were isolated from heparinized whole-blood-samples by density-gradient-centrifugation using Ficoll-Hypaque (density 1.078 g/ml). Cells were washed and resuspended in PBS [39]. Cell-counts were quantified and PBMCs were pipetted in 12-well-tissue-plates (X-Vivo15-Medium, Bio-Whittaker, Walkersville, MD, USA) for serum-free culture experiments. Pretesting of the three serum-free-methods (Calcium-Ionophore (A23187), Picibanil (OK-432) and Cytokines) successfully generating canine-DCs resulted in at least one optimal DC-generating method for each dog being selected for DC-generation (data not shown) [39]. DCs were generated, according to the different protocols, harvested and counted. During the maturation-period, peptides (20 mg/ml final concentration) were added to the medium to permit peptide-uptake. A refined gating strategy was applied for DC-analysis and DC-quantification (FACS) [39].


Anti-cmAbs (anti-canine-monoclonal antibodies) and anti-hmAbs (anti-human-monoclonal antibodies) were used for analysis of canine-cell surface antigen-expressions to evaluate and quantify amounts and phenotypes of DCs, monocytes, B and T cells in the PBMC-fractions on day 0 and day of harvest by FACS. Used anti-hmAbs were described being cross-reactive with the homologous canine-antigens [39].

mAbs were directly FITC- or PE-labelled. Canine (c) and human (h) Abs were purchased from Serotec (S), BD/Pharmingen (B; Heidelberg, Germany), Immunotech/Beckmann Coulter (I; Krefeld, Germany) and Caltag (C; Frankfurt, Germany): hCD1a-PES, cCD3-FITCS, cCD3-PES, cCD4-FITCS, cCD4-PES, cCD8-PES, cB-cells-PES, hCD14-FITCB, hCD40-PEI, hCD54-PEI, hCD56-PEI, hCD58-FITCB, hCD80-PEB, hCD83-FITCI, hCD86-FITCC, hCD116-PEI, hCD206-PEI, hCD209-FITCB, hMHC-class-I-FITCB and cMHC-II-FITCS. PBMCs/cultured-cells were incubated with mAbs (PBS) according to manufacturer's instructions, including appropriate isotype controls. Expression data were evaluated on a FACS-Calibur-Flow-Cytometer using Cell-Quest-data acquisition and analysis software (BD).

UTY-mRNA expression (RT-PCR)

Total dog-RNA was extracted from female and male cells (PBMCs, DCs, B cells, monocytes, BM) using RNeasy Mini Kit (Qiagen, Hilden, Germany) and cDNA synthesis was performed for each sample with 1 μg total-RNA using the SuperScript II Reverse Transkriptase (Invitrogen, Darmstadt, Germany) according to the manufacturer's protocols. 100 ng cDNA was applied in the PCR-reaction using the Red-Taq-Readymix PCR-Reaction-Mix (Sigma-Aldrich, Hannover, Germany). For the detection of UTY-specific cDNA, 4 μl of the following primers were used (100 pmol/μl, Metabion, Martinsried, Germany): 5′ ttc agg aaa tcg atc ctt gg 3′ and 5′ ttg tca cag gct tcc cta cc 3′. Samples were normalized for beta-Actin RNA-expression with the following primer (1 μl): 5′ gtg ggg cgc ccc agg cac ca 3′ and 5′ ctc ctt aat gtc acg cac gat ttc 3′. Cycling conditions were 95 °C for 2 min, and 35 cycles of 95 °C for 1 min, 55 °C for 1 min and 72 °C for 1 min and a final extension step of 72 °C for 7 min. PCR fragments (UTY: 237 bp; beta-Actin: 540 bp) were separated on 1% Agarose gels (120 V, 1 h) and visualized by Ethidium bromide under UV-light.

Generation of DC-primed canine-UTY-specific effector T cells (CTLs)

CD3+ T cells were positively selected from female-cPBMCs using cCD3-PE (Serotec) and Anti-PE-beads as recommended by the manufacturer. 1–2 × 106 T cells/well were co-cultured with autologous-mature DCs (5 × 104) pulsed with male-hUTY-derived peptides (20 mg/ml) in 2 ml X-Vivo15 containing hIL-2 (80 U/ml) and hIL-7 (8 ng/ml; PAN). Restimulation with peptide-loaded autologous-DCs was performed on days 0, 7, 14 and 21 (X-Vivo15 and cytokines) [39, 40].


Lytic-activity of cCTLs was assessed after 3–4 stimulations in a [51Cr]-release-assay [39, 40]: Target cells were labelled with 100 μCi [51Cr] for 1.5 h (37 °C) in dog-serum, washed and resuspended in X-Vivo15. [51Cr]-labelled target cells (2000 cells/well) were incubated with effector cells for 4 h (37 °C; E:T = 80:1) in 96-well-microtiter plates. Radioactivity of culture-supernatant was measured by a γ-counter and percentage of specific-lysis (cytotoxicity) was calculated:% cytotoxicity = (experimental release-spontaneous release)/(maximum release-spontaneous release) × 100. For the blocking-experiments, we used the monoclonal human canine-cross-reactive MHC-I antibody (clone G46-2.6, end-concentration of 40 μg/ml, BD, Heidelberg, Germany).

Canine-IFN-γ-ELISPOT assay

Canine-IFN-γ-ELISPOT assay (R&D-Systems, Minneapolis, MN, USA) used to quantify peptide-epitope-specific, IFN-γ-releasing effector cells, performed according to the manufacturers' instructions and examined on day 21 or 28 of T cell stimulation. Precursor frequency of cUTY-specific T cells in dogs' peripheral blood was evaluated on day 0. Spots were counted by visualization using a dissecting-microscope. For the blocking-experiments, we used the monoclonal human canine-cross-reactive MHC-I antibody (clone G46-2.6, end-concentration of 40 μg/ml, BD).

In vivo immunization

In vivo generation of hUTY-specific-CTLs was tested by immunizing a female dog with PBMCs from a DLA-identical-male dog. On day 0, 50 ml heparinized peripheral-blood was taken from the male-donor and PBMCs were isolated as described above. 2.5 × 108 cells were resuspended (5 ml warm-RPMI1640) and applied in equal-amounts subcutaneously to the four limbs, followed by a second-immunization on day 14. There, PBMCs (3.2 × 108 in 20 ml RPMI1640) were injected intravenously with 100 ml NaCl. 35 days after the second-injection, blood-derived T lymphocytes were harvested and studied for their UTY-specific reactivity. Distribution of the different cell-populations was monitored at day 0, 14 and 35 via flow-cytometry (donor and recipient).

Statistical analysis

Mean- and standard-deviation were performed using microsoft® excel xp, and Statistical-calculations were achieved using spss-Version 11.5 (SPSS, Chicago, IL, USA). A statistical significance was accepted for P ≤ 0.05.


Generation of UTY-specific-CTL responses in vitro using peptide-pulsed-autologous-female DCs (APCs)

Canine-female-UTY-specific CTLs were induced in vitro using autologous-DCs derived from monocytes of healthy female dogs (#1, #4, #6). DCs were pulsed with the identified HLA-A2-binding hUTY-derived peptides W248, T368 and K1234. T cells decreased during the first 2 weeks of stimulation, but then the surviving T cells proliferated, resulting in a 1.5-2.9-fold percentage-increase of successfully expanded cCTLs (Fig. 1), whereas the amount of CD4+ T cells decreased (1.6–2.9-fold; data not shown). That means that the absolute T cell number increased after 3–4 weeks of in vitro culture.

Figure 1.

Expansion of female CD8+ T cells upon stimulation with autologous-UTY-pulsed DCs in the presence of IL-2 and IL-7. (A) Percentages of CD3+/CD8+-female T cells and isotype controls on day 0 and after 3 weeks of stimulation with autologous UTY-pulsed DCs as APCs are given. Presented are data of (a) W248-specific CTLs (dog #4), (b) K1234 CTLs (dog #6) and (c) T368-specific CTLs (dog #6). (B) Shown are the increasing percentages of CD3+/CD8+-female T cells within 3 weeks of stimulation (d0, d14, d21) of the three UTY-specific cCTLs (#4: filled symbols, #6: open symbols) from panel (A). In addition, for both dogs a positive control (Mage-3) is shown. (*) In this panel CD8-FITC/CD3-PE were used instead of CD3-FITC/CD8-PE (all the other panels).

Cytotoxicity of the provoked cCTLs was assessed in a [51Cr]-release-assay: T cells demonstrated peptide-specific killing against autologous-DCs pulsed with synthetic-peptides derived from hUTY-protein in vitro (E:T = 80:1; Fig. 2A). CTLs only recognized DCs loaded with cognate-peptides (lysis: W248 (n = 3): 15.4 ± 2.9%; T368 (n = 2, #4 + 6): 47.9 ± 10.0%; K1234 (n = 2, #4 + 6): 28.5 ± 14.7%; P < 0.024 to P < 0.026, Wilcoxon-test), whereas they did not lyse naïve DCs (W248: 2.3 ± 1.2%; T368: 9.1 ± 12.8%; K1234: 1.7 ± 2.4%) and autologous-monocytes (W248: 1.0 ± 2.1%; T368: 0%; K1234: 7.3 ± 3.6%).

Figure 2.

Generation of UTY-specific CTL-responses in vitro using autologous-female-peptide-pulsed DCs as APCs. DCs generated from autologous-female monocytes were pulsed with the synthetic-peptides derived from the UTY-protein (W248, T368, K1234) and used to induce female-CTL-responses in vitro. (A) CTL-responses of stimulated female-UTY-specific CTLs can be achieved against autologous-peptide-loaded DCs in vitro. Cytotoxic activity of the W248-CTLs (black; n = 3, dog #1, #4, #6), T368-CTLs (grey; n = 2, dog #4, #6) and K1234-CTLs (stripes; n = 2, dog #4, #6) was analysed in a standard-[51Cr]-release-assay. The female T cells only recognized autologous-DCs loaded with the cognate-peptides (auto DCs + peptide), whereas they did not lyse unpulsed autologous-DCs (auto DCs) as well as autologous-monocytes (auto Monos). E:T = 80:1. (B) Specific IFN-γ-secretion of stimulated female anti-UTY-CTLs can be achieved against autologous peptide-loaded DCs in vitro in IFN-γ-ELISPOT assays (E:T = 40:1) using the same CTLs and target cells. Target cells and the dogs' T cells served as negative controls and the number of spots detected was subtracted from the experimental-values. All microcultures/controls were performed in triplicates. Given are the specific spots/100,000 CD3+ T cells investigated. *Significant difference in T cell reactivity against autologous-female-peptide-loaded DCs, unpulsed autologous DCs and autologous-monocytes, respectively.

Parallel, canine-IFN-γ-ELISPOT assays (E:T = 40:1; Fig. 2B) were performed using the same target cells. There, UTY-specific CTLs generated from healthy female dogs recognized hUTY-peptide-loaded-DCs with 281–3106 specific-spots/100,000 T cells (median: 900/100,000; P < 0.042, Wilcoxon-test). Control cells, i.e. unpulsed-autologous DCs and monocytes, were not recognized (0–55/100,000 T cells, median: 19/100,000; P < 0.024 to P < 0.026, Wilcoxon-test). W248-specific-CTLs reacted with UTY-loaded-autologous DCs within a range of 280–540/100,000 T cells (median: 392), T368-specific-CTLs with 2807–3106/100,000 T cells (median: 2957) and K1234-specific T cells with 900–965/100,000 IFN-γ-secreting T cells (median: 932). Unloaded autologous-DCs and monocytes were not recognized or only at background-levels (W248: 2–55/100,000, median: 19; T368: monocytes: 12–55/100,000, median: 34; K1234: 0–12/100,000, median: 6).

UTY peptides are endogenously and differentially presented by male-cell types

We wanted to generate cUTY-specific T cells, characterize their functional-repertoire and their Y-restriction to possibly increase GvL-specificity by investigating DLA-identical male-cells: T cells from six female dogs (#1, #4, #6, #9, #11, #14) were expanded using autologous-female DCs pulsed with the hUTY-derived peptides W248, T368 and K1234. We evaluated the ability of the in vitro induced female CTLs to recognize male-DLA-identical cells via hUTY-peptides (UTY-specific-reactivity) in IFN-γ-ELISPOT assays: female T cells were investigated in the presence of T2-cells (Table 2) and different target cells from the autologous-female-dogs, DLA-identical females and DLA-identical male-dogs (BM, DCs, monocytes, B cells, PBMCs and peptide-loaded-DCs, Fig. 3). UTY-specific-CTL reactivity was only detected in 50% of dogs tested (3/6: #1, #4, #6). Accordingly, T cell/target cell combinations of autologous-female-dogs, DLA-identical-females and DLA-identical-male-dogs were tested (#1/#2/#3; #4/#6/#5; #6/#4/#7; Table 1).

Table 2. UTY-specific CTLs derived from female donors specifically recognize peptide-loaded T2-cells
Spots/100,000 T-cells (E:T = 40:1)W248 CTLsT368 CTLsK1234 CTLs
dog #1dog #4dog #6dog #6dog #4dog #6
  1. <MHC-I>, <MHC-I > -antibody.

  2. UTY-specific T-cells from 3 female-donors were additionally tested in IFN-γ-ELISPOT assays using T2-cells as targets (E:T = 40:1). CTLs only recognized T2-cells loaded with the cognate-peptides whereas T2-cells loaded with the other two peptides or the non-T2-binding I540S-peptide were only targeted unspecifically. This was shown by blocking-experiments with an <MHC-I >-antibody. Furthermore, unpulsed T2 cells were only recognized to a low extent. W248-CTLs showed reactivity in all three dogs tested, K1234-CTLs in two and T368 in one dog. Listed are the counted spots/100,000 T-cells indicating IFN-γ secreting T-cells. The number of spots detected by the coincubation of T2-cells with DMSO and the dog T-cells served as negative controls and were subtracted from the experimental values.

  3. Bold values indicate the specific reactivity of the CTLs against their peptides and therefore seperates them from the controls.

T2 +  W248 23 34 65 271810
T2 +  W248 + <MHC-I> 6 22 44 291511
T2 +  T3687933 42 372
T2 +  T368 + <MHC-I>8644 17 783
T2 +  K1234862713 34 106
T2 +  K1234 + <MHC-I>6112516 22 68
T2 + I540S01243201586
Figure 3.

UTY-specific CTLs derived from female-donors specifically recognize male-DLA-identical cells, but not autologous or DLA-identical female cells. T cells from 6 female-donors were expanded using autologous-female DCs as APCs pulsed with the UTY-derived peptides W248, T368 and K1234. Specific T cell recognition was examined in IFN-γ-ELISPOT assays using these female-UTY-specific CTLs and were performed on day 21–28 (E:T = 80:1) in the presence of different target cells: DLA-identical-male-samples (DCs ± cognate and unspecific-peptide (grey), Monos (dark grey), BM (black), PBMCs (white) and B-cells (striped)) and the control-cells from autologous-female/DLA-identical-female-cells (DCs ± peptide (grey), Monos (dark grey), BM (black), PBMCs (white) and B-cells (striped)). The number of spots detected was pooled to show the immunogenicity of the UTY-derived peptides and the MHC-I-restriction of the female-T cells. Target cells and T cells alone served as negative controls. The number of spots detected with the negative-control cells was subtracted from the experimental values. This was tested in two independent experiments and all microcultures/controls were performed in triplicates. Shown are specific spots/100,000 CD3+ T cells investigated indicating IFN-γ-secreting T cells. Specific CTL-responses of female-T cells against male-DLA-identical-cells (A) for W248 (dogs #1, #4, #6; n = 3), (B) for T368 (dog #6; n = 1) and (C) for K1234 (dogs #4, #6; n = 2), but not against female-autologous or DLA-identical-female-cells. Following combinations of autologous-, DLA-identical-female- and DLA-identical-male-canine-cells were used according to the female-dog from which UTY-specific T cells were generated: #1-#2-#3; #4-#6-#5; #6-#4-#7; see Table 1). DCs = Dendritic-cells, Monos = monocytes, PBMCs = peripheral-blood-mononuclear-cells, BM = bone-marrow, <MHC-I> = <MHC-I > -antibody. *Significant difference in T cell-reactivity against DLA-identical-male-samples and autologous-/DLA-identical-female-samples.

To demonstrate, whether the hUTY-peptides are presented via MHC-I and whether these antigens could be specifically recognized by CTLs, peptides were loaded on hT2-cells, and CTL-reactivity was monitored with and without a canine-cross-reactive MHC-I-blocking antibody. CTLs could specifically, i.e. in an MHC-I-restricted-fashion, recognize peptide-loaded hT2-cells as shown in Table 2 (E:T = 40:1; W248-CTLs: 65–23/100,000 T cells, <MHC-I>: 44–6/100,000; T368-CTLs: 42, <MHC-I>: 17; K1234-CTLs: 106–34/100,000, <MHC-I>: 68–22/100,000; P < 0.026 to P < 0.024, Wilcoxon-test). Unpulsed T2-cells, pulsed with the two other UTY-peptides or the non-T2-binging-I540S-peptide served as controls (W248-CTLs: 0–43/100,000 T cells, median: 10; T368-CTLs: 13–27/100,000 T cells, median: 18; K1234-CTLs: 3–86/100,000 T cells, median: 17; P < 0.046 to P < 0.023, Wilcoxon-test, exceptions: T2-cells versus T2-cells + W248 and K1234 + <MHC-I>: P < 0.113 and P < 0.335, respectively).

Generated female-canine-W248-specific CTLs (Fig. 3A) recognized DLA-identical-male cell types in all three cases tested with up to 98/100,000 specific-spots (median: 28/100,000; E:T = 80:1; n = 3) in an MHC-I-restricted manner (<MHC-I>: 2-30/100,000, median: 19/100,000), T368-specific cCTLs (Fig. 3B) specifically reacted against DLA-identical male-cells only in one dog (#6) (<38/100,000 T cells; <MHC-I>: 0–6/100,000; n = 1) and K1234-specific cCTLs (Fig. 3C) induced MHC-I-restricted IFN-γ-secretion in 2/3 samples (#4 + #6) towards male-cells (up to 338/100,000 K1234-specific T cells, median: 39/100,000; <MHC-I>: 0–113/100,000, median: 15/100,000; P < 0.041 to P < 0.001, Wilcoxon-test; n = 2). In all cases, controls, i.e. the corresponding female-DLA-identical and autologous-female cell-types (without presentation of male-restricted Y-chromosomal-peptides like UTY) were not recognized or only to low extent (W248: <29/100,000 T cells; T368: <20/100,000 T cells; K1234: <59/100,000 T cells; P < 0.046 to P < 0.002, Mann–Whitney-U-test). Supplementary exogenous peptide-addition to male-DCs revealed an increased cCTL-reactivity for all three peptides compared to the naïve male-DCs (W248: 54 ± 26 versus 35 ± 25 spots/100,000 T cells; T368: 20 ± 4 versus 11 ± 3/100,000; K1234: 117 ± 102 versus 107 ± 104/100,000; P < 0.025 to P < 0.024, Wilcoxon-test). In contrast, male-DCs loaded with an unspecific peptide revealed low CTL-reactivity, showing the CTLs′ peptide restriction and specificity (W248 (K1234): 17 ± 11/100,000 T cells; T368 (W248): 5 ± 3; K1234 (W248): 39 ± 12; P < 0.043 to P < 0.010, Wilcoxon-test). Female-autologous and DLA-identical-female DCs were not targeted (W248: 1 ± 2/100,000 T cells; T368: 6 ± 2/100,000; K1234: 20 ± 25/100,000; all P < 0.025, Mann–Whitney-U-test), but when pulsed with hUTY-peptides, cCTL-reactivity increased (W248: 29 ± 20 spots/100,000 T cells; T368: 20 ± 4/100,000; K1234: 59 ± 40/100,000; P < 0.026 to P < 0.024, Wilcoxon-test).

Besides, male-BM was the cell-type being mostly recognized by the in vitro-generated female-canine CTLs (38–338 spots/100,000 T cells), followed by male-DCs (11–181/100,000), male-PBMCs (5–109/100,000), male-monocytes (<79/100,000) and male-B cells (<33/100,000). This pattern was detected for each of the three UTY-peptides. Additionally, UTY-mRNA-expression levels (total-dog-RNA; RT-PCR) of the different hematopoietic cell-types from all animals investigated were determined semi-quantitatively (Fig. 4): Only male-cells showed differentially extended UTY-mRNA-expressions (male-BM≫DCs, PBMCs, monocytes, B cells), correlating directly with the ELISPOT-data (Table 2, Fig. 3), whereas female-tissues lack UTY-mRNA.

Figure 4.

UTY-expression in different male and female hematopoietic-derived cell-types. UTY-mRNA expression-levels were determined by RT-PCR from total dog-RNA. Different cell-types (Monocytes, PBMCs, bone-marrow (BM), B-cells and DCs (Dendritic cells)) from animals investigated were observed. Here, the results of four female (#1, #2, #4 and #6) and two male dogs (#3 and #7) are given. All cells showed UTY-mRNA expression in male dogs (237 bp) to different extents, whereas female tissues did not express UTY. As internal control the house-keeping gene beta-Actin was used (arrow, 540 bp). M = Marker, 100 bp-ladder; -RT = RT-control, w/o enzyme; -PCR = PCR control, w/o template (cDNA).

Human W248 is the most immunogenic UTY-derived-peptide in dogs

Although non-homologous amino-acids may play a role in T cell-recognition by the TCR (T cell-receptor)-peptide (possibly resulting in more potent or weaker reactions than the natural dog peptide) we could work out an immunogenicity-hierarchy of the human-peptides in the dog model. The most immunogenic human-UTY-derived peptide in the canine-system was W248 with 85 ± 21 specific-spots/100,000 T cells (BM; E:T = 80:1) in 3 dogs (Fig. 3). K1234 could provoke a higher specific T cell amount in one dog compared to W248 (338/100,000 T cells; 80:1; BM), but in total it was less immunogenic regarding reactive-dogs (n = 2) and counted spots (202 ± 192/100,000 T cells; E:T = 80:1; BM). T368 was the less immunogenic hUTY-peptide with 38/100,000 T cells (E:T = 80:1; BM; n = 1). Altogether, the most immunogenic human-UTY-derived peptide was W248 (3/3 = 100%), followed by K1234 (2/3 = 67%) and T368 (1 dog = 33%).

UTY-specific generation of IFN-γ-producing cells in vivo

As a proof-of-principle we wanted to confirm our in vitro data in an in vivo experiment. UTY-specific CTLs were obtained by immunizing a female dog (dog #6) twice (day 0 and 14) with DLA-identical-male PBMCs (dog #7). Thirty-five days after the second injection peripheral-blood T cells were harvested and studied for their UTY-specific reactivity in IFN-γ-ELISPOT assays (E:T = 20:1, Fig. 5). Monocytes, PBMCs and BM (Fig. 5A–C) from the DLA-identical male-dog served as target cells verifying the endogenous cUTY-presentation on male cell-types, cells from a DLA-identical female-dog (dog #4) and autologous female-cells (#6) served as controls. Additionally, cAPCs and hT2-cells (Fig. 5D) were pulsed with hUTY-derived peptides. Female T cells’ MHC-I-restriction was confirmed with Anti-MHC-I-mAb.

Figure 5.

Specific recognition of UTY-derived peptides by expanded DLA-identical-recipient T cells presented on male-donor-cells and T2-cells. T cells from the female-dog #6 were expanded in vivo using male-DLA-identical-PBMCs from dog #7 as APCs. The recipient was immunized twice (day 0 and 14). Thirty-five days after the second injection T cells’ IFN-γ-ELISPOT assays were performed with (A) monocytes, (B) PBMCs and (C) BM from the DLA-identical-donor to verify the male-specific UTY-presentation. Cells from a DLA-identical-female-dog (dog #4) as well as autologous-cells served as controls. Target cells were additionally pulsed with the three UTY-derived-peptides W248, T368 and K1234 or the I540S-control-peptide. (D) Furthermore, T2-cells were pulsed with these peptides. MHC-I-restriction of the female-T cells was shown by adding an Anti-MHC-class-I-mAb. Coincubation of T2-cells with DMSO, APCs alone, and the T cells served as negative controls and the spots detected were subtracted from the experimental values. Microcultures were analysed as triplicates. Given are the numbers of UTY-specific spots per 100,000 CD3+  T cells investigated. E:T = 20:1; <MHC-I> = <MHC-I > -antibody. *Significant difference in T cell reactivity against DLA-identical-male-samples and autologous-/DLA-identical-female-samples as well as significant reduction in T cell reactivity against T2-cells using the MHC-I-antibody.

Compositions of the different cell-populations (T cell-subtypes CD4 and CD8, monocytes, B cells and NK cells) of the male-donor and the female-recipient were separately controlled before (day 0), after 14 and 35 days of immunization via flow-cytometry (data not shown). Donor-cell-compositions did not show significant variations during in vivo culture, but a 2-fold-increase in percentage of all cell-populations of recipient cells was observed.

In vivo-generated canine-female T cells showed low reactivity (IFN-γ-ELISPOT assay) against female-control-cells and autologous-cells (Monocytes, PBMCs and BM: range: 3–5/100,000 T cells, median: 4), whereas T cells secreted IFN-γ in the presence of the male-cell-types (15–45/100,000 T cells, median: 29; P < 0.044 to P < 0.001, Mann–Whitney-U-test) being UTY-specific (<MHC-I>: 2–25/100,000 T cells, median: 7/100,000; P < 0.048 to P < 0.003, Wilcoxon-test; Fig. 5). When pulsing male-target cells (Monocytes, PBMCs and BM) with hUTY-peptides, female-T cells specifically reacted against them, shown by MHC-I-blocking-experiments (12–35/100,000 T cells, median: 20; <MHC-I>: 3–15/100,000, median: 7; P < 0.048 to P < 0.003, Wilcoxon-test). Male-target cells pulsed with the control-peptide I540S did not influence T cell reactivity compared with naïve cells (I540S: 12–29/100,000; median: 23; P < 0.106 to P < 0.066). In vivo-primed female T cells recognized peptide-loaded T2-cells (W248: 85 ± 28/100,000 T cells; T368: 35 ± 12/100,000; K1234: 50 ± 17/100,000) being UTY-specific as indicated by Anti-MHC-I-antibody-blockage (W248: 30 ± 10/100,000 T cells; T368: 26 ± 9/100,000; K1234: 10 ± 3/100,000; P < 0.026 to P < 0.018, Wilcoxon-test). In contrast, T2-cells alone or loaded with the I540S-control-peptide demonstrated only low unspecific-reactions (20–1/100,000 T cells, median: 9; P < 0.113 to P < 0.018, Wilcoxon-test).

According to the in vitro experiments (Table 2, Fig. 3) in vivo primed female T cells mostly reacted with male-BM (<45 specific-spots/100,000 CD3+T cells) followed by monocytes (<29 spots) and PBMCs (<15 spots) and in vivo immunogenicity of the hUTY-peptides was comparable with those in vitro: W248 exhibited the most immunogenic potential on T2-cells (85 spots/100,000 T cells > K1234 (50 spots) >T368 (35 spots)).


We provide evidence that hUTY-derived male-peptides specifically expand T lymphocytes derived from female-DLA-identical-dogs either using autologous-peptide-pulsed-female DCs as APC in vitro or male-DLA-identical PBMCs in vivo. The expanded female T cells recognized HLA-A2-binding hUTY-derived endogenously presented peptides W248, K1234 and T368 only on DLA-identical male-cells (mostly BM) representing a male-specific restriction. Thereby, W248 appeared to be the most immunogenic-peptide. Importantly, no response against autologous- and female-DLA-identical cells, not expressing the male-specific-UTY antigen, was detected. Therefore, we conclude that the mHA UTY is very homologous in male-humans and dogs, and the canine-system could serve as a large-animal model to study T cell applications in terms of immunotherapeutic approaches after alloSCT in male patients with female donors. Consequently hUTY-(especially W248)-pulsed female DCs might be used in male hematopoietic-SCT recipients with female stem-cell donors [3, 6, 7].

CD8+T cell-proliferation was induced up to 3-fold within 3–4 weeks (Fig. 1). After in vitro stimulation expanded CD8+T cells specifically reacted against the hUTY-derived peptides presented on autologous-female DCs in up to 3.1% of all T cells (IFN-γ-ELISPOT assay, Fig. 2), but not against autologous-naïve DCs and monocytes. This proves that HLA-A2-restricted peptides selected from human-UTY protein bind to canine-DLA-identical molecule(s), and these peptides are immunogenic in dogs and can induce UTY-specific T cell reactivity. Detected amounts of reactive-UTY-specific CD8+T cells after in vitro culture with IFN-γ-ELISPOT and [51Cr]-release-assays were comparable. This is in accordance with findings by others, although both the assays address different CTL-mechanisms [41].

Demonstrating canine-UTY-specific T cell functionality and Y-restriction of the canine-female T cells to increase GvL-specificity by adding DLA-identical male-cells, we generated female-CTLs from 6 female-dogs using autologous-peptide-loaded DCs. We were not able to generate UTY-specific CTLs in every case, depending on the tested dogs and the investigated peptide: UTY-specific CTLs were found in 50% (3/6) of dogs investigated for W248, in 33% (2/6) for K1234 and in 17% (1/6) for T368 (Fig. 3). This indicates a restriction of the selected-peptides to a homologue of hMHC-class-I-subtype HLA-A2 in dogs peptides’ immunogenicity and functionality of the generated female CTLs [24]: In this setting, we can only state that UTY-specific MHC-I-restricted CTLs can be generated, but not to which MHC-I-molecule the peptides are restricted. Five class-I-antigens are characterized in dogs [32]. Potentially, the most common and highly polymorphic canine-MHC-I-molecule DLA-88 (99% homology was predicted for the human-MHC-I-locus HLA-A2, and partially of DLA-12 and DLA-64 [22-24, 31]) could represent the involved MHC-I-antigen in UTY-presentation or others being not yet identified. Moreover, in the ELISPOT-analysis MHC-I-blocking-experiments showed MHC-I-restriction of the generated CTLs, which strengthens that peptides are endogenously presented via MHC-I. The individual case of dog #6 represented a peculiarity: Its CTLs revealed reactivity against all three hUTY-peptides. In analogy to human-experimental data those variations within single-dogs can be assumed [40]. In vitro-induced female T cells specifically recognized only male-DLA-identical cells (BM, DCs, monocytes, B cells) in IFN-γ-ELISPOT assays. Low unspecific T cell reactivity against control-cells (autologous/female-DLA-identical) might arise from unspecifically time-induced immune-reactive cells (e.g. NK cells) secreting IFN-γ or mediating target-lysis [42, 43]. Additionally, female-UTY-specific T cells only recognized hUTY-peptides presented on hT2-cells specifically. Furthermore, reactivity against the hUTY-derived peptides was detectable in three dogs (#1, #4, #6). The DLA-genotype of dogs #4 and #6 (2-5/1-13) seems to represent most likely a homologous cMHC-I-type to the human-HLA-A2-molecule, presenting all three peptides. Dog #1 (3–12/9–4-genotype) apparently has overlapping recognition-sites with 2–5/1–13-genotype, as T cell reactivity could be determined for W248. Our results clearly show evidence that UTY is not only expressed and immunogenic in canine-male-restricted- or male-cells, but additionally, that they naturally process and present hUTY-derived-peptides in sufficient amounts (UTY-restriction). Generally, reactivity of various female-effector cells against diverse cell-types in different female dogs tested, as measured by IFN-γ-secretion, was comparable.

We showed in vitro that the phenotypic analysis of the generated female T cells bearing male-specific reactivity was mainly mediated by CD8+T cells, indicating the presence/presentation of MHC-class-I-restricted UTY-peptides. To compare the efficacy of these female T cells, we also immunized one female dog in vivo with PBMCs from a DLA-identical male littermate. Male-specific recognition induced by UTY-specific CTLs after in vitro immunization was comparable to those with T cells after in vivo immunization of a female dog (Figs. 3 and 5) as male-BM was targeted with the highest efficiency, followed by DCs and PBMCs, monocytes and B cells indicating an elevated presentation of male-antigens in BM, as previously assumed by others [11, 12, 44, 45]. Nevertheless, male-BM represents the most-affected target of female T cells, indicating higher and presumably different UTY-expression/UTY-presentation [46] and confirms the high-potential of UTY in female-to-male-transplantation settings (Figs. 3 and 5). Immunization of the female dog with DLA-identical-male-PBMCs induced UTY-specific CD8+T cells, as indicated by increasing amounts of donor-PBMCs. Other mononuclear-cells, like CD4+, CD14+ and B cells, also increased within the experiment, indicating an intense immune-response (data not shown). IFN-γ-secretion was detectable against the UTY-peptides when loaded on different target cells and hT2-MHC-I-restricted cells. Thereby, immunogenicity of investigated peptides was W248 > K1234 > T368 (T2-cells). Detailed characterization of the UTY-response of female T cells exhibited a male-specific T cell response acting in an MHC-I-restricted-fashion (Figs. 3 and 5). As these blocking-experiments did not reveal a purely CTL-directed T cell reactivity (incomplete blocking), these data could implement an additional CD4+T cell-driven reactivity [43, 47]. With these data, we can evidently exclude xenogeneic-CTL-activity as in vitro data were reproducible in vivo showing male-cell-type-specific comparable results.

UTY-mRNA-expression was determined in cell-types of hematopoietic-origin (Fig. 4) and confirmed our in vitro and in vivo data: Only male cells expressed UTY-mRNA (male-BM≫DCs, PBMCs, monocytes, B-cells), whereas corresponding female cells lack UTY-mRNA proofing male-restricted UTY-expression in hematopoietic-cells [48]. UTY-expression in non-hematopoietic cells was not shown in our study. This is compelling to demonstrate as UTY is ubiquitously expressed and would lead not only to GvL-reaction after adoptive immunotherapy, but also to GvHD after transfusion of CTLs. In order to show that our UTY-derived peptides are good targets for canine-/human-GVL-reactions, canine non-hematopoietic-cells like male-fibro-blasts/keratinocytes as well as gut-, liver- and epithelium-cells should be investigated in further experiments to prove this hypothesis. Three isoform-variants of the hUTY-gene are known (additional isoforms seem to exist) and splice-variants show different expression-profiles and tissue-distributions of resulting peptides [49, 50]. In humans UTY-epitopes restricted to different HLA-types revealed different expression-patterns in skin-derived fibroblasts: HLA-A2- and -B7-restricted peptides were found to be highly expressed on hematopoietic-cells, but also on non-hematopoietic tissues, HLA-B8-restricted UTY-specific CTLs lysed male-hematopoietic-cells efficiently, whereas no or limited reactivity was detected against HLA-B8+ male-fibroblasts [12, 44, 51]. Apparently, T cell-reactivity depends on HLA-restriction of the UTY-peptides which might be due to differential tissue-distribution of tissue-specific splice-variants. In dogs, splice-variants might also exist and be differentially expressed in organs/cell-types. Another possibility to identify UTY-tissue-distribution is to test UTY-specific CTLs in a skin-explant-model [52]. In any case only transplantation and adoptive immunotherapy will give answers regarding GvHD and conversion of chimerism after transfusion of UTY-specific CTLs obtained from immunized female donors or generated in vitro using autologous-DCs + peptides [53].

During our dog-UTY-studies, canine-Y-chromosome-/UTY sequence was not available in database (canine-genome data rose from female-dog material), but finally the dog-UTY sequence was published [54]. Blast-analysis of canine-UTY- and human-UTY-protein-/peptide sequences including their corresponding X-chromosomal counterparts (UTX) was used to confirm the postulated UTY-analogies. Amino acid (AA) differences were present for W248 (AA6 + 9) and T368 (AA4 + 8) in the canine-sequence but substituted AAs bear comparable chemical properties (exception: T368-AA9: human: F-polar; dog: Y-unpolar), therefore showing high similarity. K1234-peptide sequence and UTY-homologue UTX sequences for all three peptides were identical in dogs. These alterations can also explain the different recognition-patterns of the three peptides in the context of the different dogs' DLA-genotypes producing UTY-specific T cell reactivity or not (#1, #4, #6 versus #2, #3, #5, #7–#15). Therefore, the supposed similarities of canine- and human-UTY sequences were evidently proved by dog-UTY sequence explaining binding of human-peptides to canine-DLA [32]. Despite the use of the cUTY-sequence in our experiments, we could clearly demonstrate the generation of specific male- and MHC-I-restricted cCTL-reactivity evidently verifying UTY-expression, presentation and immunogenicity in dogs, although we cannot show data with the native canine-UTY peptides. As canine-sequences are expected to be highly homologous to their human orthologues, further scientific strategies have to focus on the amplification and sequence of the relevant canine cDNA-sequences using human-, mouse- and rat-UTY-sequences, resulting in the use of completely authentic canine-minor-epitopes. Indeed, BLAST-sequence alignments of dog-UTY with human-, mouse- and rat-UTY DNA, mRNA and protein revealed accordance in 89% for humans, 86% and 84% for mouse and rat, respectively. Designing and testing multiple PCR-primer-sets would have enabled us to be capable of amplification of genuine canine-UTY cDNA-fragments. Retrospectively alignments of the investigated hUTY-peptides with those of canine-, murine- and rat-sequences (Table 3) revealed conservation of K1234 in all four species. For W248, the most immunogenic peptide in our setting (positive in 3 dogs) changes only appeared in 0-2 AAs which had no influence on peptide-sequence/properties and their immunogenic potential. Furthermore, W248-data could already be shown in mice [55]. Highest divergence was determined for T368, with the human- and canine-peptides being similar at most. As determined in this work for UTY, substantial homologies and conservation of immune-reactivity, functionality, proteins, peptides (including MHC-presentation) and isoforms were already described for canines in comparison to humans, cats, mice, rats, apes and cows by others in vitro and in vivo [30, 56-68].

Table 3. Sequence alignments of canine, human, murine and rat UTY-peptides
  1. AA, amino acids.

  2. Retrospectively BLAST-sequence alignments of dog-UTY with human-, mouse- and rat-UTY DNA, mRNA and protein revealed accordance in 89% for humans, 86% and 84% for mouse and rat, respectively. Alignments of the investigated human-UTY-peptides with those of canine-, murine- and rat-sequences revealed conservation of K1234 in all four species. For W248, changes only appeared in 0-2 AAs which had no influence on peptide-sequence/properties. Highest divergence was determined for T368, whereas the human and canine peptides were the ones being similar at most. Amino-acid-changes in the sequences are marked in grey and amino-acids typed in bold share same chemical properties compared to human peptide-sequence.


Further in vitro-culture experiments of lymphocytes from in vivo immunized females with DLA-identical-male cells should be performed to strengthen our preliminary data of our first proof-of-principle experiments. Furthermore, higher response of in vivo T cell proliferations might be exhibited by peptide-loaded (single-peptides or peptide-mix/pool) male-DCs or male-PBMCs, as well as investigating human- and canine-UTY-peptides in parallel. Thereby, using human- and canine-UTX-homologue peptides, unspecific X-chromosomally derived reactivity can be excluded and the DLA-binding efficacy of the human/canine-UTY/UTX peptides will be verified as well. Non-hematopoietic-cells (fibroblasts, keratinocytes) should be examined with respect to their target cell function as well as their UTY-expression profiles. In further studies we want to transfer our setting in clinical settings, especially in a context of stem-cell transplantation or T cell transfer for treatment of human leukemia: Normally, UTY is not restricted to cells of hematopoietic origin, but the level of expression may differ in various tissues. Adoptive immunotherapy with Y-chromosome-encoded UTY would be feasible in certain circumstances. This first proof-of-principle experiment should demonstrate that hUTY-peptides are presented on male-canine cell-surfaces triggering a male-specific immune-response. Interpretation of our experiments could be enhanced by cloning some canine-T cells via limiting-dilution-culture recognizing one of the three hUTY-derived-peptides, permitting more detailed examinations of the antigenic specificity and functional properties of CD8+ as well as CD4+cells [43].

Adoptive immunotherapy with DLT after SCT provides a potent strategy to curatively treat haematological malignancies [69]. However, the use of DLT is limited by occurrence of GvHD and sometimes by the poor-response of the patients [70]. Optimized sensitization of donor T cells against antigens presented by leukemic cells could improve DLT. Therefore, ex vivo CTL-generation against UTY for the treatment of recurrent leukemia is reasonable. After adoptive immunotherapy host-haematopoiesis (including residual-leukemic-cells) is completely replaced by the graft [71]. Dogs are a valuable preclinical model for transplantation studies, including adoptive immunotherapy with donor lymphocytes. Conversion of mixed-haematological-chimerism into complete-donor-chimerism thereby simulate efficacy of transplantation [21, 72, 73].

In conclusion, after establishing the implements for the generation of cUTY-specific CTLs, we are able to use this mixed-chimerism model as an in vivo model for the treatment of leukemic relapse with UTY-specific CTLs. In up to 50% of the females we could induce a UTY-specific reaction (W248) in male-DLA-identical animals in vitro and in vivo. This is a very promising starting point for exploitation of our preclinical canine-model for leukemia treatment in humans: Ex vivo-generated UTY-specific-female-donor CTLs using UTY-derived-peptide-loaded DCs will be transfused to male-recipients in the course of DLT after transplantation in order to prevent or cure AML-relapse.


We thank the people from the animal facility (Helmholtz Center Munich), especially M. Hagemann, S. Schlink and V. Terkowski for taking care of the dogs. We also thank I. Laaser and J. Adamski (Helmholtz Center Munich, Neuherberg) for providing the canine-UTY-mRNA sequence. Supports: DLR-grant 01GU0516 (D. Bund); Deutsche-José-Carreras-Stiftung-e.V. (H.J. Kolb).

Conflict of interest statement

All authors concur with the manuscript submission and have no financial/commercial conflict of interest to disclose.