Journal Logo

Original Basic Science—General

Local Injections of Tacrolimus-loaded Hydrogel Reduce Systemic Immunosuppression-related Toxicity in Vascularized Composite Allotransplantation

Dzhonova, Dzhuliya V. PhD1,2; Olariu, Radu MD3; Leckenby, Jonathan MD3; Banz, Yara MD, PhD4; Prost, Jean-Christophe PhD5; Dhayani, Ashish MSc6,7; Vemula, Praveen K. PhD6; Voegelin, Esther MD3; Taddeo, Adriano PhD1,3; Rieben, Robert PhD1

Author Information
doi: 10.1097/TP.0000000000002283

Systemically administered tacrolimus (TAC) is the most commonly used immunosuppressant in vascularized composite allotransplantation (VCA).1 However, of the 66 registered in the International Registry on Hand and Composite Tissue Transplantation hand transplant recipients, 26% suffered from elevated creatinine values, 32.3%—from opportunistic bacterial infections and 3 of them developed malignancies.2 These TAC-mediated morbidities are a barrier to the broader adoption of VCA. Transitioning patients from TAC to other immunosuppressants has been attempted, but with limited success.2 Consequently, the field has turned to “increasingly bold approaches in modifying immunosuppression,”3 that need solid and conclusive preclinical data, demonstrating their feasibility, efficacy, and safety.

Our group developed a graft-targeted inflammation-responsive hydrogel4-6 delivering TAC “on demand”—only when needed, with the aim to provide an effective future alternative or addition to systemic immunosuppression for patients. The hydrogelator—triglycerol monostearate (TGMS)—is biocompatible, biodegradable, generally recognized as safe by the US Food and Drug Administration, and can be loaded with therapeutically relevant amounts of TAC. Tacrolimus-loaded TGMS hydrogel (TGMS-TAC) releases TAC in response to inflammatory stimuli, and prolongs VCA survival with a single injection.7

Here, we hypothesized that repeated subcutaneous intragraft injections of TGMS-TAC maintain long-term graft survival in the Brown Norway-to-Lewis rat hindlimb allotransplantation model. We expected that TGMS-TAC–treated animals would have higher TAC concentrations in the graft and lower in the blood compared to daily systemic TAC treatment (standard of care), which should result in reduced off-target effects and nephrotoxicity. Furthermore, we were interested in whether TGMS-TAC influences the dynamics of effector T (Teff) cells, regulatory T (Treg) cells, and chimerism.


Male Brown Norway and Lewis rats (6-8 weeks old weighing 200 to 250 g) were purchased from Charles Rivers Breeding Laboratories, Germany. Animals were kept in specific pathogen-free conditions. Experiments were planned and carried out in agreement with current 3R and ARRIVE guidelines and approved according to Swiss animal protection laws by the Veterinary Authorities of the Canton Bern, Switzerland, approval no. BE94/15.

Brown Norway-to-Lewis rat hindlimb transplantations were performed and animals were treated either with 1 mg/kg per day TAC systemically in the neck fold or every 70 days with 1 ml TGMS-TAC containing 7 mg TAC (n = 6 for each group). In the TGMS-TAC group, 4 subcutaneous TGMS-TAC depots of 250 μL each were injected in the zones of biceps femoris, gastrocnemius, tibialis anterior, and vastus muscles, taking great care to distribute the amount of drug as evenly as possible intra and interindividually. The reinjection time point was chosen based on a pilot study showing that transplanted animals (n = 5) treated with a single intragraft injection of 1 ml TGMS-TAC loaded with 7 mg TAC on postoperative day (POD) 1 rejected their graft on POD 83.4 ± 6.7. Reinjection time point was defined as 14 days before rejection and set to POD 70. Graft rejection was evaluated macroscopically and graded as 0, no rejection; 1, erythema and edema;2, epidermolysis and exudation; and 3, desquamation, necrosis, and mummification.

Rats were euthanized either once grade 3 rejection was reached or on day 280 (endpoint). Necropsy for immunosuppression-related side effects was performed. Kidney, as well as graft skin and muscle histology, was evaluated by a blinded pathologist (hematoxylin and eosin and/or Periodic acid-Schiff staining, as necessary). Kidney samples were graded according to the semiquantitative calcineurin inhibitor toxicity score by Kambham et al.8 Histological grading of skin rejection was according to Banff classification.9 Additionally, skin and muscle lymphocyte infiltration, vasculopathy and necrosis were graded as 0, none; 1, minimal; 2, moderate; 3, extensive. Immunofluorescence analyses of IgG, IgM, C3b/c, C4b/c, C5b-9, CD45RA in graft skin and muscle were performed. Blood urea nitrogen (BUN) and creatinine; cholesterol and triglycerides; aspartate aminotransferase (AST) and alanine aminotransferase (ALT) as kidney, metabolism, and liver markers, respectively, complete blood count, and donor-specific antibody formation were assessed in blood at endpoint.

Throughout the study, TAC levels were measured in blood and skin biopsies retrieved from grafts and contralateral limbs at selected time points by LC-MS/MS.

Flow cytometry for obtaining Treg cell and chimerism levels was performed in blood at selected time points, and in graft and contralateral limb skin at endpoint.

Statistical analyses were executed with Prism software (GraphPad Software Inc., La Jolla, CA). Statistically significant data were presented as follows: *P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001. Statistical tests are specifically indicated under each figure.

Detailed materials and methods are available in Materials and Methods (SDC,


Periodic TGMS-TAC Injections Promote Long-term VCA Survival

To test if 1 mL TGMS-TAC loaded with 7 mg TAC reinjected every 70 days promotes long-term graft survival in a rat hindlimb transplantation model, we compared TGMS-TAC treatment to daily systemic immunosuppression using TAC at 1 mg/kg per day (Figure 1A). Five of 6 animals survived until endpoint in each group (Figure 1B).

Long-term graft survival and graft histology are comparable between TGMS-TAC and systemic tacrolimus treatment. A, Experimental scheme of the 2 treatment groups. Group 1, 1 mL hydrogel with 7 mg tacrolimus intragraft-treated group (TGMS-TAC). Group 2, daily systemic treatment with 1 mg/kg per day tacrolimus group (systemic tacrolimus), including treatment application frequencies and planned duration of the experiments. B, Kaplan-Meier survival curve of TGMS-TAC–treated and systemic TAC-treated animals (n = 6/group). C, Macroscopic grading of graft rejection over time in each of the rats (animals 1 to 6, TGMS-TAC–treated animals 7 to 12, systemically treated). Appearance of edema and erythema are defined as grade 1, epidermolysis and exudation as grade 2, desquamation, necrosis and mummification as grade 3. TGMS-TAC reinjection time points are indicated. Animal 1 was euthanized at POD 149 after reaching grade 3 rejection. D, Representative histological hematoxylin and eosin–stained sections of graft skin collected at endpoint—POD 280 from both treatment groups and their corresponding histopathological evaluation. Intergroup differences of the results presented are not significant as evaluated by Mann-Whitney test. Individual values are presented as dots with indication of the mean values by lines.

In the TGMS-TAC group, 1 animal was sacrificed due to grade 3 rejection at POD 149 (animal 1, Figure 1C). One animal experienced 3 rejection episodes (animal 2, Figure 1C). One animal started rejecting at POD 262, which reached grade 2 at endpoint (animal 3, Figure 1C). One animal experienced mild rejection, which resolved temporarily and later reoccurred (animal 4, Figure 1C). Two animals had no rejection episodes (animals 5 and 6, Figure 1C). No local complications due to TGMS-TAC injections were observed in any of the rats. In systemic TAC group, 1 animal was euthanized due to lymphoma on POD 275. Systemic immunosuppression was sufficient to maintain the grafts rejection-free throughout the duration of the treatment (animals 7-12, Figure 1C).

Histopathological evaluation of graft skin at endpoint (POD 280) revealed no necrosis in any of the animals (animals 1-12, Figure 1D). Two TGMS-TAC–treated animals were classified as Banff grade 3 (corresponding to macroscopic grade 2), and 1 as Banff grade 2 (corresponding to macroscopic grade 1) skin rejection (animals 2-4, Figure 1D). One systemically treated animal was classified as Banff grade 2 skin rejection, although no macroscopic lesions were detectable (animal 12, Figure 1D). No significant differences were found between the 2 groups in terms of lymphocyte infiltration (P = 0.2415), vasculopathy (P = 0.1411), or Banff grade (P = 0.1660, Mann-Whitney test, Figure 1D). All observed rejection episodes were restricted to graft skin, with no signs of rejection in graft muscle (Figure S1, SDC, However, TGMS-TAC–treated animals had significantly more pronounced atrophy of the muscle fibers as compared to systemic treatment (P = 0.0079, Mann-Whitney test, Figure S1, SDC,

As evident from Figures 1C and 2, rejection episodes in animal 2 markedly improved after TGMS-TAC reinjection. The first one—macroscopic grade 1 at the time of reinjection—was completely reverted within a week. The second one—macroscopic grade 2—required 2 to 3 weeks to reduce to grade 1 (Figure 2), and within a month, a complete recovery was observed, which, however, lasted only 3 weeks.

TGMS-TAC treatment does not completely prevent, but can revert rejection episodes. A, Representative photographs of rejecting graft before (POD 204) and after (POD 234) TGMS-TAC reinjection and (B) corresponding histological hematoxylin and eosin–stained sections of rejecting graft skin before and after TGMS-TAC reinjection.

Reduced Systemic But Relevant Tissue TAC Levels With TGMS-TAC

To understand TAC distribution over time, TAC concentrations were measured in blood (biweekly), and in graft and contralateral limb skin (monthly). In blood, a burst release of TAC was detected in the first 72 hours after the first TGMS-TAC injection. This peak was beyond the upper quantification limit of the LC-MS/MS analysis (ie, 65 ng/mL) during the first 24 to 48 hours, afterward, it normalized to therapeutic levels (5-20 ng/mL). At POD 46 most animals had subtherapeutic levels (<5 ng/mL). Compared with the first TGMS-TAC injection, the following injections induced markedly weaker burst releases, followed by similar TAC release kinetics (P < 0.0001 for first peak versus second, P < 0.0001 for first peak versus third, and P = 0.0009 for second peak vs third, standard 1-way analysis of variance [ANOVA], Figure 3A). Trough levels of systemically treated animals were within the therapeutic range throughout the duration of the experiments (Figure 3A). Biweekly average TAC levels in blood of TGMS-TAC and animals showed that TGMS-TAC–treated animals had significantly lower TAC blood levels compared with systemically treated animals (9.29 ± 5.89 ng/mL vs 13.44 ± 4.44 ng/mL, respectively, P = 0.0060, Mann-Whitney test, Figure 3B).

Blood tacrolimus levels are lower with TGMS-TAC therapy. A, Tacrolimus levels were measured by LC-MS/MS in blood of TGMS-TAC and systemic TAC-treated rats (trough levels) over time. TGMS-TAC reinjection time points are indicated by syringe symbols on the x-axis. Statistical analyses of the differences between the peaks of tacrolimus release under TGMS-TAC treatment are shown. Data are shown as individual values and means, ***P < 0.001, ****P < 0.0001 by ordinary 1-way ANOVA. Individual values have been manually shifted to left or right to make each value visible. B, Mean biweekly tacrolimus blood levels between TGMS-TAC and systemic TAC-treated rats are compared. Each data point represents the mean value of pooled tacrolimus measurements acquired over the course of each 2 consecutive weeks (starting from postoperative week 1 + 2, then 3 + 4, and so on, until the end of the study). Statistical analyses of the differences between the 2 groups are shown. Data are presented as individual values, mean ± SD. are indicated, **P < 0.01 by Mann-Whitney test.

In TGMS-TAC–treated animals, there was a nonsignificant trend toward higher TAC concentrations in the graft as compared with the contralateral limb skin during the first week, corresponding to the burst release observed in blood (P = 0.1532, paired t test, Figure 4A). For the rest of the measured time points, there were no statistically significant differences between TAC levels in graft and contralateral limb skin. In systemically treated animals, there were no statistically significant differences between the TAC levels in graft and contralateral limb skin at all time points (Figure 4B). Monthly average skin TAC levels in TGMS-TAC graft skin were higher than in the respective contralateral limb skin (1.1 ± 1.49 ng/mg vs 0.25 ± 0.19, respectively, P = 0.0195, Figure 4C). Under systemic treatment, there was no significant difference between graft and contralateral limb skin (0.46 ± 0.39 ng/mg vs 0.29 ± 0.16 ng/mg, respectively, P = 0.1094, Wilcoxon test, Figure 4C). When compared between groups, TAC levels in both graft and contralateral limb skin were comparable (P = 0.7785 and P = 0.4755, respectively, Mann-Whitney test, Figure 4C).

Tacrolimus levels in graft skin are higher with TGMS-TAC therapy. A, LC-MS/MS tacrolimus measurements in graft and contralateral limb skin biopsies of TGMS-TAC–treated rats over time. TGMS-TAC reinjection time points are indicated. B, LC-MS/MS tacrolimus measurements in graft and contralateral limb skin biopsies of systemic TAC-treated rats over time. Data are shown as individual values and means. Individual values have been manually shifted to left or right to make each value visible. C, Mean TAC levels in graft and contralateral limb skin of TGMS-TAC–treated or systemic TAC-treated animals. Each data point represents the mean value across the experimental group at each time point of TAC measurements in either graft or contralateral limb skin. Statistical analyses of the differences between the 2 groups are shown. Data are shown as individual values, mean ± SD are indicated, *P < 0.05, Mann-Whitney test is used for comparisons between the 2 treatment groups, Wilcoxon test is used for comparisons within a group between the 2 sites of biopsy collections.

Low Antidonor Antibody and Complement Activation

The formation of donor-specific antibody (DSA) was assessed by incubating donor thymocytes with plasma of transplanted animals and subsequent analysis by flow cytometry. There was no significant intergroup difference for binding of IgG to donor thymocytes, whereas IgM was significantly lower in the TGMS-TAC group at POD 280 as compared with systemic treatment (mean fluorescence, 75.56 ± 18.62 and 143.8 ± 41.09, respectively, P = 0.013, Student t test, Figure 5A). At endpoint, no significant antibody (IgG and IgM, Figure 5B) and complement (C4b/c, C5b-9, Figure S2, SDC, deposition, or B-cell infiltration was observed (CD45R, Figure S3, SDC,, except for increased C3b/c deposition under systemic treatment (integrated density 1018 ± 155.6 and 1500 ± 300.1 for TGMS-TAC and systemic treatment respectively, P = 0.029, Student t test, Figure S2, SDC,

Local and systemic TAC treatment is associated with low DSA formation. A, Levels of IgG and IgM DSA in the plasma of TGMS-TAC or systemic TAC-treated rats at POD 280. Donor-thymocytes have been incubated with heat-inactivated recipients’ plasma isolated at POD 280, followed by staining with anti-IgG and anti-IgM antibody. DSA deposition on thymocytes have been analyzed by flow cytometry and expressed as MFI. Lower threshold for positivity is defined as mean + 2× SD of the MFI measured in donor-thymocytes incubated with naive Lewis rats’ plasma. Data are shown as individual values, mean ± SD are depicted, *P < 0.05 by Student t test. B, Representative results of immunostained graft skin and muscle cryosections with anti-IgG or anti-IgM antibody and (C) their corresponding quantification. Data are shown as individual values, mean ± SD are depicted. Intergroup differences are not significant as evaluated by Student t test. MFI, mean fluorescence intensity.

TGMS-TAC Mitigates Immunosuppression-related Side Effects

Blood urea nitrogen in TGMS-TAC animals was higher than that in naive age-matched animals and lower than that in systemically treated animals—12.62 ± 1.6, 5.9 ± 0.2, and 20.86 ± 4.86 mmol/L, respectively (P = 0.0023 TGMS-TAC vs systemic treatment, P = 0.0095 TGMS-TAC vs naive, and P < 0.0001 systemic treatment vs naive, Figure 6A). Creatinine was lower in TGMS-TAC and naive age-matched animals compared with systemic treatment (27.2 ± 4.21, 23.2 ± 2.68, and 49.8 ± 17.08 μmol/L, respectively, P = 0.0117 TGMS-TAC vs systemic treatment, P = 0.0039 systemic treatment vs naive, Figure 6B). Histological analysis of kidneys revealed only minimal damage under both treatments (Figure S4, SDC,

TGMS-TAC treatment mitigates elevation of kidney function markers and occurrence of malignancy and opportunistic infections. A-F, Biochemical analyses of the plasma levels of (A) BUN, (B) creatinine, (C) cholesterol, (D) triglycerides, (E) AST, (F) ALT in plasma collected at POD 280 from TGMS-TAC and systemic TAC-treated rats. Data are shown as individual values, mean ± SD are depicted, *P < 0.05, **P < 0.01, ****P < 0.0001 by 1-way ANOVA. G, Representative histological hematoxylin and eosin–stained section of pseudocyst infected with Staphylococcus aureus and Proteus mirabilis in a systemic TAC-treated rat and of (H) ipsilateral inguinal lymph node with lymphoma in another systemic TAC-treated rat.

Cholesterol was comparable between groups (2.48 ± 0.39 mmol/L, 2.35 ± 0.48 mmol/L, and 2.88 ± 0.47 mmol/L for TGMS-TAC, systemic treatment, and naive Lewis rats, respectively, Figure 6C). Triglycerides were similar in TGMS-TAC and naive rats and decreased in systemic treatment group as compared to naive rats (1.22 ± 0.54 mmol/L, 0.44 ± 0.29 mmol/L, and 1.75 ± 0.99 mmol/L for TGMS-TAC, systemic treatment and naive Lewis rats, respectively, P = 0.0236 between naive rats and systemic treatment, Figure 6D).

Hepatic enzymes were not significantly different between naive rats and the 2 treatment groups. Aspartate aminotransferase was 140.6 ± 97.18 U/L, 109 ± 52.72 U/L, and 80 ± 15.64 U/L for TGMS-TAC, systemic treatment, and naive rats, respectively, Figure 6E. Alanine aminotransferase was 73.4 ± 29.57 U/L, 48.4 ± 29 U/L, and 50.4 ± 7.8 U/L for TGMS-TAC, systemic treatment, and naive rats, respectively (Figure 6F).

The complete blood count of TGMS-TAC and systemically treated animals at endpoint was comparable to naive age-matched rats, except for total hemoglobin, mean corpuscular hemoglobin, platelet distribution width, and median platelet volume, which were lower under systemic treatment, compared either to naive animals or to both naive and TGMS-TAC–treated animals (1-way ANOVA, Table 1).

Complete blood count of TGMS-TAC and systemic TAC-treated groups at POD 280 versus naive age-matched animals

As mentioned above, under systemic treatment 1 of the 6 animals was euthanized at POD 275, due to markedly enlarged ipsilateral inguinal lymph node, accompanied by elevated white blood cell count (75.1 × 103 cells/μL) and apathetic behavior indicating pain and/or suffering. Histopathological analyses of the lymph node revealed aggressive lymphoma, most consistent with diffuse large B cell lymphoma (Figure 6G). Another animal from the same group had an increasingly firm and growing solid mass circumventing the graft, accompanied by a slow but steady increase of the white blood cell count until endpoint. Necropsy revealed a large encapsulated granuloma-like formation filled with granulated yellow-green substance. Histopathological analysis confirmed that the formation was an infected pseudocyst (Figure 6H). Polymerase chain reaction analyses of its content revealed Staphylococcus aureus and Proteus mirabilis, commensal skin bacteria. In the TGMS-TAC group neither malignant nor infectious complications were observed.

TGMS-TAC Therapy Favors Hematopoietic Chimerism

To understand the dynamics of Teff and Treg cells and chimerism under both treatments, we analyzed blood at selected time points throughout the study. The gating strategy for Teff and Treg cells enumeration is shown in Figure S5 (SDC,, and for chimerism in Figure S6 (SDC,

Both treatment groups had significantly decreased amounts of circulating T cells compared with naive rats. Initially, there were significantly more T cells in the TGMS-TAC group than in systemically treated group (for example, in postoperative week 2: 2087 ± 427 T cells/μL vs 1163 ± 359 T cells/μL, respectively, P = 0.0074, Student t test). After 17 weeks of gradual decrease the difference of T-cell counts between the TGMS-TAC and systemically treated group were no longer statistically significant (for example, in postoperative week 19, 1825 ± 767 T cells/μL vs 1290 ± 336 T cells/μL, P = 0.1486, Student t test, Figure 7A). Three T-cell populations—cytotoxic T lymphocytes (CTL), T helper cells and Treg cell—were separately analyzed, with additional focus on Helios+ and Helios Treg cell populations. The T helper cells were the most abundant T-cell population and followed the total T-cell dynamics (Figure 7B). There were no major differences in the CTL or the Treg cell populations between the 2 treatment groups over time (Figure 7C-F, respectively).

TGMS-TAC treatment depletes T cells to a lesser extent than systemic TAC treatment. A-F, Absolute number of T cell populations in blood of TGMS-TAC and systemic tacrolimus treatment at selected time points. The values measured in naive animals are reported (black dots). Cells are enumerated by flow cytometry as (A) T cells (CD45+, CD3+), (B) cytotoxic T cells (CD45+, CD3+, CD8+), (C) T helper cells (CD45+, CD3+, CD4+), (D) Treg cells (CD45+, CD3+, CD4+, FoxP3+, CD25high), (E) Helios+ Treg cells (CD45+, CD3+, CD4+, FoxP3+, CD25high, Helios+), (f) Helios Treg cell cells (CD45+, CD3+, CD4+, FoxP3+, CD25high, Helios). Data are shown as mean ± SD. Statistical analyses between naive and the 2 treatment groups—1-way ANOVA. Highlighted in orange are the time points, which are significantly different between TGMS-TAC and systemic tacrolimus treatment groups. Significant differences between naive animals and treated animals are not shown.

In terms of chimerism, in the first 11 weeks, there was a significantly higher amount of circulating donor-derived cells in the TGMS-TAC group (Figure 8A). Donor-derived B cells, T helper cells, CTL, and monocytes were all significantly increased in the TGMS-TAC–treated group compared with systemic treatment for up to 23 weeks (Figures 8B-E). Circulating donor-derived granulocytes were initially high in both treatment groups (for example, in postoperative week 2, 367 ± 92 and 314 ± 113 donor-derived granulocytes/μL for TGMS-TAC and systemic treatment, respectively). This number dropped to 58 ± 52 cells/μL and 90 ± 36 cells/μL, respectively, at postoperative week 10 and remained low until termination of the experiment (Figure 8F). At endpoint, peripheral blood monocytes isolated from graft and contralateral limb skin of both groups were analyzed using the same flow cytometry protocol. The cell count was low and revealed no significant differences between the 2 treatment groups (Figure S7, SDC,

TGMS-TAC boosts multilineage hematopoietic chimerism in blood. A-F, Donor-derived white blood cell populations in blood from TGMS-TAC and systemic tacrolimus at selected time points. The values measured in naive animals (unspecific-staining) are reported (black dots). Donor cells are enumerated by flow cytometry as (a) all donor-derived white blood cells (CD45+, RT1a+), (b) donor-derived B cells (CD45+, CD3, CD4, SSClow, RT1a+), (c) donor-derived T helper cells (CD45+, CD3+, CD4+, RT1a+), (d) donor-derived cytotoxic T cells (CD45+, CD3+, CD8+, RT1a+), (e) donor-derived monocytes (CD45+, CD3, CD4+, RT1a+), (f) donor-derived granulocytes (CD45+, CD3, CD4, SSChigh, RT1a+). Naive Lewis rats’ white blood cells are RT1a. Data are shown as mean ± SD. Statistical analyses between naive and the 2 treatment groups—1-way ANOVA. Highlighted in orange are the time points, which are significantly different between TGMS-TAC and systemic tacrolimus treatment groups. Significant differences between naive animals and treated animals are not shown.


Our data show that repeated intragraft injections of TGMS-TAC sustain long-term graft survival with better toxicological and immunological outcomes as compared to systemic TAC delivery. Markers of kidney function (ie, BUN and creatinine) and complete blood analysis at endpoint, showed preserved kidney and hematological parameters of TGMS-TAC–treated rats as compared to systemic treatment. Unlike humans,10 rat models require sodium depletion to develop significant TAC-induced kidney damage.11-13 Therefore, we speculate that the toxic effects reported in this study may be underrepresented, and that the TGMS-TAC treatment may potentially have more visible benefits in humans, especially in the kidney on a histological level.

Our study was also of sufficient duration to reveal possible complications of long-term immunosuppression. One systemically treated animal developed an infected pseudo-cyst containing commensal skin bacteria. Another developed an aggressive lymphoma. Lymphomas can arise spontaneously in aging Lewis rats; however, their incidence during the first year of life of a male Lewis rat is extremely low,14 suggesting that systemic immunosuppression contributed to its development. Necropsy of TGMS-TAC–treated animals did not reveal any malignancy or opportunistic infection, suggesting that localized immunosuppression could mitigate immunosuppression-related complications.

Local complications related to TGMS-TAC treatment, such as rash, alopecia, discoloration, atrophy or thinning of skin, or extracutaneous hydrogel extrusions were not observed. Animals did not extensively groom or scratch the limb after injection indicating absence of local irritation. Stool was firm and urine was clear, suggesting no acute gastro-intestinal or renal complications resulting from TGMS-TAC either.

Although providing better recipient outcomes, TGMS-TAC treatment resulted in inferior graft outcomes as compared to systemic treatment. Four of the 6 TGMS-TAC–treated animals experienced at least 1 rejection episode. Rejecting TGMS-TAC–treated animals had comparable systemic TAC levels to the nonrejecting TGMS-TAC–treated animals. Moreover, we7 and others15 have demonstrated that localized immunosuppression promotes extended rejection-free graft survival in the setting of subtherapeutic systemic TAC levels. Therefore, we hypothesize that these rejections are not due to subtherapeutic systemic TAC levels, but rather to low intragraft TAC levels. According to Capron et al,16 tissue levels of immunosuppression provide a more accurate insight into actual efficiency of immunosuppression. Reinjecting TGMS-TAC guided by local TAC levels, instead of fixed time points, could mitigate the observed rejections. To test this hypothesis, we plan to conduct TGMS-TAC studies in a porcine VCA model. In addition to being more clinically relevant, pigs provide the opportunity to collect frequent biopsies, sufficient to identify minimal threshold for intragraft TAC levels.

The TGMS-TAC–treated animals demonstrated increased muscle atrophy as compared to systemically treated animals. Calcineurin is involved in skeletal muscle hypertrophy and tacrolimus counteracts this effect.17 However, to our knowledge, there have been no studies demonstrating that tacrolimus monotherapy causes direct myotoxicity, as conversely reported for tacrolimus in conjunction with statins.18 Moreover, clinical cases of tacrolimus overdose have not reported effects on skeletal musculature,19 suggesting that muscle atrophy is not TAC-related. Mechanical pressure of the hydrogel deposits on graft vessels or nerves resulting in muscle atrophy is not a likely explanation either, because grafts were all well perfused and all animals used their limbs for walking until endpoint in both groups. The hydrogel itself has been previously described to be safe, biocompatible and biodegradable.7 However, our data cannot rule out the possibility that muscle atrophy may be a hydrogel-related side effect, which could not develop in studies of shorter duration. Importantly, muscle atrophy is a known manifestation of chronic rejection, and we believe that this is the most likely explanation for our observations. Indeed, multiple acute rejection episodes have been correlated to chronic rejection, particularly in rat.20 However, nonrejecting TGMS-TAC–treated animals also had high muscle atrophy scores, keeping the question of muscle atrophy a matter requiring further investigation.

A third and potentially problematic aspect of TGMS-TAC hydrogel could be the TAC burst release following TGMS-TAC injection. The TGMS-TAC injections led to peaks in TAC blood levels that, with each subsequent injection, became significantly lower. Because of the enzyme responsiveness of the hydrogel, the most likely reason for the very high first peak is the elevated levels of inflammation-related enzymes resulting from the surgical trauma and ischemia-reperfusion injury. However, in our view, the burst release has arguably a negative impact, as high intragraft perioperative TAC levels were shown to prolong graft survival in the same experimental model.21

Each TGMS-TAC injection contained 7 mg of TAC and the total amount of TAC given over 280 days to TGMS-TAC–treated animals was 28 mg. In contrast, animals treated systemically with 1 mg TAC/kg per day received a total of 84 to 112 mg of TAC, depending on the weight of the rats, which ranged from 300 to 400 g. Consequently the systemic TAC levels in TGMS-TAC–treated animals were significantly lower than the trough TAC levels in systemically treated animals. Nevertheless, TAC levels in skin were comparable between the 2 groups. Interestingly, although in nonrejecting TGMS-TAC–treated animals, TAC levels were similar in transplanted and contralateral limb skin, upon rejection levels increased in the transplanted limb as compared with nonrejecting grafts and contralateral limb skin (Figure S8, SDC, This supports the idea that rejection triggers the local release of the drug. We also found that at endpoint systemically treated animals had significantly decreased tissue TAC levels (P < 0.05 in graft and P < 0.001 in contralateral limb skin as compared to previous time point by paired t-test). We do not have an explanation for this observation, as reduced systemic TAC levels, or any observable physiological changes, did not accompany it.

In terms of immunological outcomes, the amount of circulating and intragraft Treg cell was comparable between TGMS-TAC and systemic treatment. However, TGMS-TAC treatment was associated with higher and more persistent hematopoietic chimerism compared with systemic treatment. Chimerism is a protolerogenic factor22 and boosting it without aggressive preconditioning or bone marrow transplantation may be an attractive option to control antigraft immunity in VCA. Despite elevated chimerism, most TGMS-TAC–treated animals experienced rejection episodes. Chimerism alone is not sufficient to prevent rejection and requires the support of higher Treg cell counts,23 which was not the case under both treatments. Moreover, it has been shown that robust chimerism cannot prevent rejection once immunosuppression is tapered in a porcine VCA model.24 The levels of chimerism in TGMS-TAC–treated animals, despite being elevated compared to systemic treatment, were still below the threshold required for tolerance.25 Therefore, rejection due to reaching low intragraft TAC levels was not preventable by the achieved increase in chimerism with TGMS-TAC. Nevertheless, the possibility to use localized immunosuppression to increase chimerism levels to the “tolerogenic threshold”25 represents an interesting opportunity that deserves further investigation.

In 2014, for the first time, the VCA society dealt with antibody-mediated rejection in a presensitized face recipient, raising patient sensitization as the next frontier in the field.26 Recent studies in rat VCA model have clearly demonstrated that sensitized recipients experience accelerated rejection of both cell- and antibody-mediated nature.27 In our study, we have not included a presensitized group, neither did our animals develop de novo DSA. Complement deposition28 and tertiary lymphoid structure formation,29 which were also described as participants in the VCA rejection process, were also not detected, consistent with previous studies.30 Future studies addressing the efficacy of TGMS-TAC in a sensitized animal model would provide a strong argument on the potential and limitations of this therapeutic modality.

In view of clinical application, a combination of “the best of both worlds”—combining use of reduced systemic immunosuppression and local “on demand” immunosuppression—might be envisaged to balance the outcomes of graft and recipient. Moreover, single-drug immunotherapies are not successful in clinical VCA. Multidrug immunosuppressive protocols are currently used in transplanted patients to guarantee an effective level of immunosuppression. Therefore, we believe that protocols involving localized immunosuppression in humans should further evolve by including multiple drugs to better control graft rejection.

In summary, this study demonstrates that the use of an enzyme-responsive drug delivery system for localized immunosuppression in VCA results in long-term graft survival with reduced drug-related side effects. These findings support the safety of this therapeutic possibility and suggest a potential to mitigate immunosuppression-related morbidities in patients.


The authors would like to thank Jane Shaw-Boden, Catherine Tsai, and Tsering Wuethrich for technical assistance and proofreading. LC-MS/MS analyses were performed at the Clinical Metabolomics Facility, Center of Laboratory Medicine from the Bern University Hospital “Inselspital”. The authors thank Michael Hayoz for blood LC-MS/MS routine analyses, Alexander Leichtle for BUN, creatinine, cholesterol, triglycerides, AST and ALT analyses, and Gabriela Mäder for help with tissue LC-MS/MS TAC analyses.


1. Howsare M, Jones CM, Ramirez AM. Immunosuppression maintenance in vascularized composite allotransplantation. Curr Opin Organ Transplant. 2017;22:463–469.
2. Petruzzo P, Sardu C, Lanzetta M, et al. Report (2017) of the International Registry on Hand and Composite Tissue Allotransplantation (IRHCTT). Curr Transpl Rep. 2017;4:294–303.
3. Dean WK, Talbot SG. Vascularized composite allotransplantation at a crossroad. Transplantation. 2017;101:452–456.
4. Qiu Y, Park K. Environment-sensitive hydrogels for drug delivery. Adv Drug Deliv Rev. 2001;53:321–339.
5. Patterson J, Hubbell JA. Enhanced proteolytic degradation of molecularly engineered PEG hydrogels in response to MMP-1 and MMP-2. Biomaterials. 2010;31:7836–7845.
6. Purcell BP, Lobb D, Charati MB, et al. Injectable and bioresponsive hydrogels for on-demand matrix metalloproteinase inhibition. Nat Mater. 2014;13:653–661.
7. Gajanayake T, Olariu R, Leclere FM, et al. A single localized dose of enzyme-responsive hydrogel improves long-term survival of a vascularized composite allograft. Sci Transl Med. 2014;6:249ra110.
8. Kambham N, Nagarajan S, Shah S, et al. Semiquantitative, clinically correlated calcineurin inhibitor toxicity score for renal allograft biopsies. Clin J Am Soc Nephrol. 2006;2:135–142.
9. Cendales LC, Kanitakis J, Schneeberger S, et al. The Banff 2007 working classification of skin-containing composite tissue allograft pathology. Am J Transplant. 2008;8:1396–1400.
10. Kershner RP, Fitzsimmons WE. Relationship of FK506 whole blood concentrations and efficacy and toxicity after liver and kidney transplantation. Transplantation. 1996;62:920–926.
11. Kędzierska K, Sindrewicz K, Sporniak-Tutak K, et al. Does immunosuppressive therapy affect markers of kidney damage? Ann Transplant. 2016;21:137–144.
12. Andoh TF, Burdmann EA, Lindsley J, et al. Functional and structural characteristics of experimental FK 506 nephrotoxicity. Clin Exp Pharmacol Physiol. 1995;22:646–654.
13. Stillman IE, Andoh TF, Burdmann EA, et al. FK506 nephrotoxicity: morphologic and physiologic characterization of a rat model. Lab Invest. 1995;73:794–803.
14. Baum A, Pohlmeyer G, Rapp KG, et al. Lewis rats of the inbred strain LEW/Han: life expectancy, spectrum and incidence of spontaneous neoplasms. Exp Toxicol Pathol. 1995;47:11–18.
15. Unadkat JV, Schnider JT, Feturi FG, et al. Single implantable FK506 disk prevents rejection in vascularized composite allotransplantation. Plast Reconstr Surg. 2017;139:403e–414e.
16. Capron A, Haufroid V, Wallemacq P. Intra-cellular immunosuppressive drugs monitoring: a step forward towards better therapeutic efficacy after organ transplantation? Pharmacol Res. 2016;111:610–618.
17. Dunn SE, Burns JL, Michel RN. Calcineurin is required for skeletal muscle hypertrophy. J Biol Chem. 1999;274:21908–21912.
18. Dalakas MC. Toxic and drug-induced myopathies. J Neurol Neurosurg Psychiatry. 2009;80:832–838.
19. Mrvos R, Hodgman M, Krenzelok EP. Tacrolimus (FK 506) overdose: a report of five cases. J Toxicol Clin Toxicol. 1997;35:395–399.
20. Unadkat JV, Schneeberger S, Horibe EH, et al. Composite tissue vasculopathy and degeneration following multiple episodes of acute rejection in reconstructive transplantation. Am J Transplant. 2010;10:251–261.
21. Olariu R, Denoyelle J, Leclère FM, et al. Intra-graft injection of tacrolimus promotes survival of vascularized composite allotransplantation. J Surg Res. 2017;218:e107–e107.
22. Siemionow M, Cwykiel J, Madajka M. Bone marrow-derived ex vivo created hematopoietic chimeric cells to support engraftment and Maintain Long-Term Graft Survival in Reconstructive Transplantation. In: Brandacher G, editor. The Science of Reconstructive Transplantation. Stem Cell Biology and Regenerative Medicine. New York, NY: Springer New York; 2015:227–254.
23. Pilat N, Farkas AM, Mahr B, et al. T-regulatory cell treatment prevents chronic rejection of heart allografts in a murine mixed chimerism model. J Heart Lung Transplant. 2014;33:429–437.
24. Shanmugarajah K, Powell H, Leonard DA, et al. The effect of MHC antigen matching between donors and recipients on skin tolerance of vascularized composite allografts. Am J Transplant. 2017;17:1729–1741.
25. Lin J-Y, Tsai F-C, Wallace CG, et al. Optimizing chimerism level through bone marrow transplantation and irradiation to induce long-term tolerance to composite tissue allotransplantation. J Surg Res. 2012;178:487–493.
26. Chandraker A, Arscott R, Murphy GF, et al. The management of antibody-mediated rejection in the first presensitized recipient of a full-face allotransplant. Am J Transplant. 2014;14:1446–1452.
27. Wang HD, Fidder SAJ, Miller DT, et al. Desensitization and prevention of antibody-mediated rejection in vascularized composite allotransplantation by syngeneic hematopoietic stem cell transplant. Transplantation. 2018;102:593–600.
28. Sarhane KA, Tuffaha SH, Broyles JM, et al. A critical analysis of rejection in vascularized composite allotransplantation: clinical, cellular and molecular aspects, current challenges, and novel concepts. Front Immunol. 2013;4:406.
29. Hautz T, Zelger BG, Nasr IW, et al. Lymphoid neogenesis in skin of human hand, nonhuman primate, and rat vascularized composite allografts. Transpl Int. 2014;27:966–976.
30. Unadkat JV, Schneeberger S, Goldbach C, et al. Investigation of antibody-mediated rejection in composite tissue allotransplantation in a rat limb transplant model. TPS. 2009;41:542–545.

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.