Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Feb 10.
Published in final edited form as: Cell Chem Biol. 2019 Feb 28;26(5):652–661.e4. doi: 10.1016/j.chembiol.2019.01.011

Activation of BMP Signaling by FKBP12 Ligands Synergizes with Inhibition of CXCR4 to Accelerate Wound Healing

Brandon J Peiffer 1,2,4, Le Qi 3,4, Ali R Ahmadi 3, Yuefan Wang 1,2, Zufeng Guo 1,2, Hanjing Peng 1,2, Zhaoli Sun 3,*, Jun O Liu 1,2,5,*
PMCID: PMC7008977  NIHMSID: NIHMS1068956  PMID: 30827938

SUMMARY

The combination of AMD3100 and low-dose FK506 has been shown to accelerate wound healing in vivo. Although AMD3100 is known to work by releasing hematopoietic stem cells into circulation, the mechanism of FK506 in this setting has remained unknown. In this study, we investigated the activities of FK506 in human cells and a diabetic-rat wound model using a non-immunosuppressive FK506 analog named FKVP. While FKVP was incapable of inhibiting calcineurin, wound-healing enhancement with AMD3100 was unaffected. Further study showed that both FK506 and FKVP activate BMP signaling in multiple cell types through FKBP12 antagonism. Furthermore, selective inhibition of BMP signaling abolished stem cell recruitment and wound-healing enhancement by combination treatment. These results shed new light on the mechanism of action of FK506 in acceleration of wound healing, and raise the possibility that less toxic FKBP ligands such as FKVP can replace FK506 for the treatment of chronic wounds.

Graphical Abstract

graphic file with name nihms-1068956-f0001.jpg

In Brief

Peiffer and colleagues report that a non-immunosuppressive FK506 analog (FKVP) significantly accelerates wound healing in diabetic rats when combined with AMD3100. FKVP was found to exert this FKBP12-specific effect through activation of the BMP pathway. Conversely, systemic BMP inhibition blocked treatment-enhanced healing and stem cell recruitment at wound areas.

INTRODUCTION

Wounding due to accidents, diseases, and armed conflict is one of the most common medical problems. The cost of care for chronic, non-healing wounds associated with severe burns and diseases such as diabetes has been estimated to exceed 50 billion US dollars per year (Fife and Carter, 2012). Chronic wounds make the human body more susceptible to infection, increasing the risk of acquiring opportunistic pathogens that can lead to sepsis. Thus, accelerating wound healing (WH) can reduce the risk of infection, improving the mortality and morbidity rates of wounded patients. However, there are limited options to shorten WH, warranting the development of new therapies.

We have previously reported the discovery of a synergistic drug combination for the acceleration of cutaneous WH (Lin et al., 2014) and the induction of long-term allograft survival through host repopulation (Okabayashi et al., 2011; Hu et al., 2016; Cameron et al., 2016). The combination of two Food and Drug Administration-approved drugs, tacrolimus (FK506) and plerixafor (AMD3100), reduced the complete healing time by 25% in mice with four circular full-thickness excisional wounds, which is unprecedented by existing therapeutic modalities. Accelerated WH is accompanied by the mobilization of bone marrow-derived stem cells (CD133, CD34, and cKit) and the recruitment of CD133 stem cells into wound sites, as well as augmented stromal derived factor 1 (SDF-1), fibroblast growth factor, and vascular endothelial growth factor release in granulation tissues (Lin et al., 2014). The underlying molecular mechanism by which the combination of FK506 and AMD3100 (AF) accelerates WH has not been extensively studied. AMD3100 is a selective antagonist of the chemokine receptor CXCR4 (Hatse et al., 2002) and has been used clinically to drive hematopoietic stem cells (HSCs) out of the bone marrow into the peripheral blood (Liles et al., 2003), where they can be recovered and preserved until the completion of ablative irradiation and/or chemotherapy. In addition to HSCs, the injection of AMD3100 augmented the mobilization of bone marrow-derived endothelial progenitor cells, which was associated with more rapid neovascularization and functional recovery after myocardial infarction in mice (Jujo et al., 2010; Balaji et al., 2013). However, an increased number of circulating stem cells by AMD3100 treatment alone exhibited only slightly faster healing due to reduced recruitment in wound sites (Lin et al., 2014).

In contrast to AMD3100, the precise role played by low-dose FK506 in the combination treatment (AF) has remained a mystery. FK506, a macrolide produced by the bacteria Streptomyces tsukubaensis, is an immunosuppressant widely used for prevention of transplant rejection as well as treatment of certain autoimmune disorders (Tanaka et al., 1987; Fung, 2004). The underlying mechanism for the immunosuppressive activity of FK506 has been well established. At the cellular level, FK506 inhibits the activation of T helper cells. At the pathway level, it blocks the intracellular signal transduction emanating from the T cell receptor, leading to transcriptional activation of interleukin-2 and other cytokine genes. At the molecular level, it binds to FK506 binding protein 12 (FKBP12) and other members of the FKBP family before the binary FKBP-FK506 complex associates with and inhibits the activity of the protein phosphatase activity of calcineurin, preventing calcium-dependent dephosphorylation of the nuclear factor of activated T cells (NFAT) (Liu et al., 1991; Griffith et al., 1995; Kissinger et al., 1995). A possible underlying mechanism for FK506 in WH is through inhibition of calcineurin. However, it has been shown that topical FK506 has a detrimental effect on WH (Schäffer et al., 1998). Furthermore, we have shown that animals treated with low-dose FK506 (0.1 mg/kg) alone exhibited slightly faster healing compared with the saline control group, but the standard dose of FK506 (1 mg/kg) for immunosuppression delayed healing time, leaving unanswered the question of whether calcineurin inhibition is responsible for the effect of FK506 on WH.

Although FKBP12 plays an accessory role in the immunosuppressive activity of FK506, it has also been shown to inhibit bone morphogenetic protein (BMP) type 1 receptor activation (Wang et al., 1996). Importantly, this interaction could be relieved by FK506 (Spiekerkoetter et al., 2013). BMP signaling has not yet been directly linked to any stage of WH, although it has been reported that epithelial cells downregulate many BMP receptors in response to injury (Lewis et al., 2014). Conversely, it has been recently reported that enhanced BMP signaling within myofibroblasts may promote scarless WH (Plikus et al., 2017). BMPs have been demonstrated to produce a pro-inflammatory phenotype in endothelial cells, thereby increasing leukocyte adhesion and SDF-1 secretion (Csiszar et al., 2006; Young et al., 2012). Upon activation, BMP receptors phosphorylate and activate the SMAD transcription factors 1, 5, and 8. One major target gene of these SMADs is inhibitor of differentiation 1 (ID-1), which inhibits transcription of several genes related to embryogenesis and stem cell self-renewal. Previous studies have examined some of the downstream effects of BMP receptor activation following FK506 treatment, which is accompanied by increases in SMAD1 and SMAD5 (denoted SMAD1/5) and/or SMAD8 (denoted SMAD1/5/8) phosphorylation in skeletal muscle cells (Spiekerkoetter et al., 2013) and human synovial stromal cells (Tateishi et al., 2007). Additionally, increases were observed in MAPKK phosphorylation and ID-1 expression, and the activity of FK506 was sufficient to rescue endothelial dysfunction in mice induced by a conditional BMP receptor type 2 (BMPR2) knockout (Spiekerkoetter et al., 2013). It has been reported that FK506 upregulated phosphorylation of SMADs downstream of the transforming growth factor β (TGF-β) signaling pathway (SMAD2 and SMAD3) in smooth muscle cells (Giordano et al., 2008; Bennett et al., 2016). However, downstream transcriptional activity was only seen in the presence of supplemented exogenous TGF-β (Spiekerkoetter et al., 2013; Wang et al., 1996). In another study, it was shown that FK506 increased expression of the TGF-β type 3 co-receptor endoglin, and stimulated both migratory and angiogenic activity of endothelial cells (Albiñana et al., 2011). Together, these observations raised the possibility that FK506 may exert its WH effect through FKBP12, independent of calcineurin inhibition.

In the present study, we attempted to deconvolute the underlying mechanism of low-dose FK506 in accelerating WH when combined with AMD3100. Using a non-immunosuppressive analog of FK506, we ruled out calcineurin as a mediator of the WH acceleration activity of FK506, suggesting that FKBP12 is the likely mediator. Among the different pathways affected by FKBP12, we demonstrate that FK506 and its non-immunosuppressive analog activated the BMP signaling pathway through BMP receptor type 1 (BMPR1) kinase activation. Moreover, we found that systemic administration of the BMP inhibitor LDN-193189 (LDN) abolished enhanced healing by AF therapy, indicating that activation of BMP signaling is necessary for the WH acceleration activity of FK506. Together, these results strongly suggest that enhanced BMP signaling is intimately involved with the enhanced healing by FK506, and that this activity may synergize with stem cell mobilization prompted by concomitant AMD3100 treatment. A non-immunosuppressive analog of FK506 that is equally potent in enhancing WH also raised the exciting possibility of developing a combination therapy devoid of the side effects associated with calcineurin inhibition.

RESULTS

Modification of FK506 at C40 Led to a Non-immunosuppressive Analog, FKVP

The in vivo immunosuppressive activity of FK506 has been established to be mediated through the inhibition of calcineurin (Bueno et al., 2002). Previous studies have shown that modification at the terminal alkene of FK506 could block calcineurin inhibition while retaining FKBP binding (Clemons et al., 2002). To determine whether calcineurin inhibition was required for enhanced healing by the AF combination, we designed and synthesized a non-immunosuppressive analog of FK506 by using cross-metathesis to fuse a vinyl pyridine moiety to the terminal alkene as a “bump” in the effector domain of FK506. The resultant analog was named FKVP (Figure 1A). The newly added pyridine moiety was intended to increase water solubility while providing steric bulk to disrupt its interaction with calcineurin (Figure S1A). We assessed the cytotoxicity of FKVP and compared it with that of FK506 in both Jurkat T (Figures 1B and S1B) and primary human umbilical vein endothelial cells (HUVECs) (Figure S2C). Like FK506, FKVP did not affect cell viability at concentrations up to 10 μM (Figure 1B). We then determined the effect of FKVP on a phorbol myristate acetate (PMA)/ionomycin-activated NFAT-luciferase reporter gene in Jurkat T cells (Clemons et al., 2002). While both FK506 and cyclosporine A (CsA) exhibited potent inhibition of the reporter, FKVP did not cause significant inhibition at concentrations of up to 10 μM (Figure 1C), suggesting that FKVP is no longer capable of inhibiting calcineurin. To confirm that FKVP retained the ability to bind FKBP, we applied it to a competition assay in combination with FK506 and rapamycin, as sequestration of free FKBP will prevent the formation of active FKBP12-FK506 or FKBP-rapamycin complexes and thus antagonizing the activity of both drugs (Rao et al., 1997; Abraham and Wiederrecht, 1996). The effect of FK506 on calcineurin was determined using as a readout the phosphorylation state of NFATc2. Thus, FK506 blocked the dephosphorylation of NFATc2 in response to stimulation with PMA and ionomycin (Figure S1D). The presence of 10 μM FKVP reversed the inhibitory effect of FK506 on NFATc2 dephosphorylation, suggesting mutual antagonism between FKVP and FK506. Similarly, we examined the effect of rapamycin on mammalian target of rapamycin (mTOR) activity as judged by the phosphorylation state of its substrate p70s6k. Once again, a high concentration of FKVP reversed the inhibition of rapamycin on p70s6k phosphorylation (Figure S1E). Together, these results clearly showed that FKVP is capable of antagonizing the activities of both FK506 and rapamycin through competitive binding to endogenous FKBP proteins.

Figure 1. Generation of a Non-immunosuppressive Analog (FKVP) by Modifying FK506 at C40 Position.

Figure 1.

Open in a new tab

(A) Chemical structures of FK506, FKVP, synthetic ligand of FKBP (SLF), and CsA.

(B) Resazurin-based cell viability assay of Jurkat cells after 3 days of FKVP or FK506 treatment (n = 3). Absorbance values were normalized to DMSO control. Error bars represent standard deviation (SD).

(C) NFAT-Luciferase reporter activity of PMA/ionomycin-activated Jurkat cells is inhibited by FK506 and CsA, but not by FKVP and SLF. Dose-response curves were obtained by treating Jurkat cells expressing the NFAT-luciferase reporter gene with serial dilutions of indicated compounds, and the relative luciferase activities were determined upon normalization to DMSO control values (n = 3).

See Quantification and Statistical Analysis for more information on analysis.

FKVP in Combination with AMD3100 Accelerated Wound Healing

We have previously reported synergistic activity of AMD3100 and low-dose FK506 (AF) in accelerating WH after full-thickness skin excision (Lin et al., 2014). To determine whether FKVP has the equivalent effect, we performed a WH experiment in a rat model of type 2 diabetes. Four full-thickness wounds were generated by 8-mm diameter circular excisions on the shaved back of a diabetic Goto-Kakizaki (GK) rat and each wound site was photographed digitally at the indicated time intervals (Figure 2A). Re-epithelialization of entire wound areas was used as a defining parameter of complete healing, and the complete healing time of four wounds in each animal was calculated in days (Figure 2B). Wounded rats were divided randomly into three experimental groups and received subcutaneous injections of saline, AF (AMD3100 [1.0 mg/kg] plus FK506 [0.1 mg/kg]), or AV (AMD3100 plus FKVP [0.1 mg/kg]) immediately after wounding and every other day until complete healing. While the saline control group showed an average complete healing time of 26 days, the animals treated with AF exhibited significantly faster healing as wounds reached complete re-epithelialization at day 21, which is consistent with our report in non-diabetic rodent models of surgical excisional wounds (Lin et al., 2014). Importantly, ten rats receiving AV therapy displayed an AF-equivalent effect of significantly reduced time for complete healing from 26 to 20 days (Figure 2C). These results strongly suggest that inhibition of calcineurin is not involved in the synergistic activity of AMD3100 and FK506 in accelerating WH.

Figure 2. Accelerated Wound Healing in Diabetic GK Rats Treated with a Combination of AMD3100 and FK506 or FKVP.

Figure 2.

Open in a new tab

(A) The wound model: four circular excisional wounds (8 mm in diameter) were created on the dorsal of GK rats.

(B) Representative photographs of wounds in GK rats for each treatment group (AF = AMD3100 + FK506, AV = AMD3100 + FKVP), at days 0, 6, 12, 18, and 24.

(C) Quantitative analysis of complete healing time in GK rats. All data represent mean ± SEM.

See Quantification and Statistical Analysis for more information on analysis.

FKVP Activates ID-1 Reporter and SMAD1/5 Phosphorylation through BMP Type 1 Receptor

Having ruled out calcineurin as a relevant mediator of the WH acceleration activity of FK506, we turned to FKBP12 and the BMP signaling pathway it is reported to regulate. We began by determining whether FK506 and FKVP are both capable of activating the BMP signaling pathway by employing a BMP-response-element (BRE) pathway reporter (luciferase under the control of the ID-1 gene promoter) (Spiekerkoetter et al., 2013) in Jurkat (E6.1) T cells. The Jurkat line was found to express working components of BMP signaling (BMPRs, SMAD1/5) in addition to high levels of FKBP12 and CXCR4, making it an excellent model system. To confirm reporter selectivity, we used rBMP-4 and rTGF-β1 as positive and negative controls, respectively. Treatment with both FKVP and FK506 caused dose-dependent activation of the reporter, and the increases in luciferase activity were completely blocked by the selective BMP1R inhibitor LDN (Figure 3A). In contrast to FK506 and FKVP, CsA did not activate the reporter (Figure 3A), consistent with the notion that calcineurin is not involved in BMPR1 kinase activation by FK506.

Figure 3. Non-immunosuppressive Analogs Preferentially Activate BMP Pathway Signaling through Increased SMAD1/5 Phosphorylation and ID1 Promoter Activity.

Figure 3.

Open in a new tab

(A) BMP-response-element (BRE) reporter activity in Jurkat cells after treatment with increasing amounts of FK506 and FKVP. This activity is strongly inhibited by the addition of 100 nM LDN-193182. TGF-β1 serves as negative control, while BMP4 shows strong induction of luciferase expression after 18 h. Relative luciferase activities were determined upon normalization to DMSO control values (n = 3). Error bars represent SD from mean. See Quantification and Statistical Analysis for more information on analysis.

(B) Jurkat cells show no appreciable increase in SMAD2/3 phosphorylation after 2 h of FKVP treatment compared with positive control (TGF-β1).

(C) Dose-dependent induction of SMAD1/5 phosphorylation by FKVP in Jurkat cells.

(D) BMPR1-selective inhibitor LDN-193189 inhibits SMAD1/5 phosphorylation induced by either BMP-4 or FKVP in Jurkat cells.

To study downstream BMPR1 signaling events, we determined the effects of FK506 and FKVP on the phosphorylation of SMAD1/5. Both compounds induced SMAD1/5 phosphorylation in a dose-dependent manner (Figure 3C), an effect abolished by LDN treatment (Figure 3D). In contrast, there were no increases in phosphorylated SMAD2/3 under the same conditions (Figure S3B), suggesting that while FKBP12 may bind TGF-β receptors (Chen et al., 1997), its dissociation is insufficient to activate receptor kinase activity in the absence of TGF-β. This result is in agreement with previous studies showing that FK506 and non-functional analogs are incapable of activating TGF-β pathway reporters without the addition of exogenous ligand (Spiekerkoetter et al., 2013; Wang et al., 1996). In the BRE-luciferase reporter, AMD3100 was found to have no effect on ID-1 promoter stimulation alone or in combination with FKVP (Figure S2A), suggesting that the synergistic activity of these drugs does not occur at the level of BMP signaling. Furthermore, we found that the extracellular BMP-inhibitor protein, Noggin, was not effective at reducing FK506- or FKVP-mediated induction of the BMP reporter (Figure S2B). These observations suggest that endogenous BMP is not required for signaling activation by FK506 or FKVP, and that these compounds could activate BMP in tissues with underex-pression of BMP protein or overexpression of extracellular inhibitors such as Noggin. Both SMAD1/5 phosphorylation and ID-1 expression were significantly increased in HUVECs after FKVP treatment (Figures S3A–S3C), providing evidence that more than one cell type is sensitive to BMP activation by FKVP. Together, these results indicated that both FK506 and FKVP are capable of activating BMPR1 kinase signaling, raising the possibility that this activation plays a key role in the healing acceleration activity of FK506 in the AF combination therapy.

FKBP12 Alone Is Required for FKVP-Induced SMAD1/5 Phosphorylation

FKBP12 is a member of the FKBP superfamily of proteins. In previous work, FKBP12 has been shown to be associated with the BMPR1 activin-like kinase 2 (ALK2). However, attempts to knock down several FKBPs failed to reveal a specific effect on BMPR1 signaling (Spiekerkoetter et al., 2013), likely due to the relative stability and high abundance of FKBPs. To address this problem, we generated CRISPR/Cas9 knockouts of three cytosolic FKBPs, namely FKBP12, FKBP51, and FKBP52. All three proteins have been reported to bind FK506 (Kozany et al., 2009), and in the BRE-luciferase reporter assay it was found that only FKBP12 knockout (KO) cells lost sensitivity to FK506 and FKVP (Figures 4A and 4B). It can also be seen that FKBP12 KO cells showed constitutively elevated levels of SMAD1/5 phosphorylation (Figure 4A), explaining the low value of BMP-4 treatment in FKBP12 KO cells relative to an already elevated DMSO-treated sample. Moreover, reconstitution of FKBP12 by an SNAP-tagged fusion construct restored FK506/FKVP sensitivity in KO cells to that of the original parental line (Figure 4C). We were able to use this fusion protein to pull down calcineurin and mTOR in the presence of FK506 and rapamycin, respectively, suggesting that the fused SNAP tag did not interfere with the interactions of the FKBP12-FK506 and the FKBP12-rapamycin complexes with calcineurin or mTOR (Figure S4). We then applied the same construct to pull down V5- or HA-tagged ALK receptors from transfected HEK293T cells. We observed that the SNAP-FKBP12 and ALK receptors did indeed interact with each other, and the association was sensitive to competition by FKVP (Figure 4D). These observations strongly suggest that FKBP12-BMPR1 interaction is solely responsible for mediating the effect of FKVP-induced BMP activation.

Figure 4. FKBP12 Alone Is Required for FK506 and FKVP-Induced SMAD1/5 Phosphorylation.

Figure 4.

Open in a new tab

(A) Induction of SMAD1/5 phosphorylation by FKVP and BMP-4 in wild-type (WT) and different FKBP isoform knockout cells.

(B) Activation of BMP pathway reporter by FKVP and FK506 in different FKBP isoform knockout cells and inhibition of the reporter gene activation by LDN (n = 3). Relative luciferase activities were determined upon normalization to DMSO control values. Error bars represent SD from mean. See Quantification and Statistical Analysis for more information on analysis.

(C) Expression of FKBP12-SNAP fusion protein restores BMP pathway activation by FKVP and FK506 in FKBP12KO Jurkat cells (n = 3).

(D) Pull-down of ALK receptors using FKBP12-SNAP in conjunction with SNAP-functionalized beads in the absence and presence of FKVP.

BMP Signaling Is Required for the Effect of AF Combination in Accelerating Wound Healing

To determine whether BMP activation by FK506 is responsible for accelerated WH, we administered a selective BMPR1 kinase inhibitor, LDN (2 mg/kg/day, intraperitoneally) to wounded GK rats treated with saline or AF combination. LDN has been shown to be effective in vivo (Cuny et al., 2008; Sun et al., 2013), and alone showed no effect on WH. Interestingly, LDN abolished the beneficial effect of AF combination therapy and increased the time for complete healing from 21 to 25 days (Figures 5A and 5B). We have reported that FK506 plays a key role in recruitment of AMD3100 mobilized CD133 stem cells into wound sites (Lin et al., 2014) or injured organs (Okabayashi et al., 2011; Hu et al., 2016; Cameron et al., 2016; Zhai et al., 2018). To further confirm whether blocking BMP signaling inhibits the recruitment of stem cells, we performed immunohistochemistry staining for CD133 in wound tissue sections recovered from animals at day 7 after surgery. A few CD133+ cells were identified in wound tissue sections from animals treated with saline (Figure 5C). The number of CD133+ cells was significantly increased in newly formed granulation tissues of the wounds in animals receiving AF combination therapy. Strikingly, administration of BMP inhibitor LDN dramatically reduced the number of CD133+ cells in the wounds in animals with AF combination treatment. Taken together, these results suggest that the recruitment of more CD133+ stem cells into the wound sites by AF combination treatment depends on BMP activation by FK506, and that blockade of BMP signaling with LDN eliminates the beneficial effect of AF combination therapy (Figure 5D).

Figure 5. BMP Signaling Is Required for the Beneficial Effects of AF Therapy in Both Wound Healing and Stem Cell Recruitment.

Figure 5.

Open in a new tab

(A) Representative photographs of wounds in GK rats treated with saline, LDN, AF, or AF + LDN showing difference between AF and AF + LDN beginning at day 12.

(B) Quantitative analysis of complete healing time in GK rats. All data represent mean ± SEM. See Quantification and Statistical Analysis for more information on analysis.

(C) Representative immunohistochemical stainings for the stem cell marker CD133 in granulation tissues of GK rats at day 7 (Magnification: 40×). The rats receiving AF treatment had a significantly higher number of CD133 cells (brown) in granulation tissues compared with the saline control group, while LDN treatment dramatically reduced the number of CD133 stem cells in the wound sites.

(D) Proposed mechanism for enhanced healing through BMP modulation. AMD3100 releases CD133+ stem cells into circulation, where FKVP-mediated BMP activation influences recruitment to wounded tissues. Systemic FKBP12 antagonism by FKVP increases BMP-related gene expression in both mobilized stem/progenitor cells and granulation tissues without affecting the necessary immune responses to wounding.

DISCUSSION

In this study, we investigated the mechanism by which FK506 accelerated WH when used in combination with AMD3100. Using FKVP, a non-immunosuppressive analog of FK506, we ruled out calcineurin as a mediator of both WH acceleration, raising the possibility that FKBPs are the primary target for both effects. Moreover, we demonstrated that macrocyclic FKBP ligands activate BMP signaling by relieving the inhibition of BMPR1 by endogenous FKBP12. We showed that FKBP12 plays an essential role in the BMP signaling pathway, an effect that can be mediated without calcineurin inhibition through the use of non-immunosuppressive FK506 analogs. We found that BMP receptor signaling is required for WH enhancement by FK506, and that blocking this activation results in fewer numbers of stem cells recruited to the wound area. BMP signaling may manipulate several cell types in the WH mechanism, such as chemotaxis of stem cells or endothelial adhesion of mobilized cells in the wounded tissue.

FKBP12, a founding member of the FKBP superfamily, has been shown to possess multiple cellular and physiological functions in addition to its role in mediating inhibition of calcineurin and T cell activation by FK506. The association with, and inhibition by FKBP12 adds another layer of BMPR1 kinase regulation. That relieving FKBP12 inhibition by FK506 or FKVP is sufficient to activate the ID-1 luciferase reporter gene suggests that there is a basal level of activity of BMPR that is normally suppressed by FKBP12, and relief of this inhibition leads to significant, albeit moderate, activation of the signaling pathway in comparison with BMP-4 binding. Thus, BMPR may exist in three distinct activation states: upon release of FKBP12 inhibition, upon BMP binding, and both. Our results revealed that the partial activation of BMP pharmacologically with FK506 or FKVP is sufficient to accelerate WH in combination with AMD3100. It remains to be seen whether that activation state is also achievable under physiological stimulation, and is involved in regulation of healing in vivo.

Type 2 BMP receptors are reported to constitutively phosphorylate the GS domain of type 1 receptors (ALKs). FKBP12 is believed to inhibit random activation of ALKs by binding to residues in the GS domain (Chaikuad et al., 2012). Our results indicated that this association is competed by FKVP for all BMP-specific ALKs. We showed that loss of FKBP12 results in elevated basal phosphorylation of SMAD1/5, suggesting a partially activated state of ALKs in the absence of its endogenous intracellular inhibitor FKBP12. Furthermore, addition of the BMP inhibitor Noggin did not prevent ID-1 reporter stimulation by FKVP, suggesting that the regulation of BMPR by endogenous FKBP12 is independent of BMP protein-receptor binding.

Inhibition of calcineurin by FK506 has been shown to be responsible for both its potent immunosuppressive activity and a number of its side effects, including nephrotoxicity and neurotoxicity (Bechstein, 2000; Naesens et al., 2009). By placing a molecular “bump” on the calcineurin-interacting effector domain of FK506, the resultant FKVP lost its immunosuppressive activity as judged by the NFAT reporter gene assay. In comparison with calcineurin, the loss of function of FKBP12 and other members seems to have much fewer and less drastic impacts on both yeast and mammals. Aside from BMP receptors, FKBP12 has been reported to modulate calcium flux in inositol 1,4,5-trisphosphate and ryanodine receptors (Cameron et al., 1995; Jayaraman et al., 1992), suggesting that FK506 may affect vascular or cardiac smooth muscle contractility. However, calcineurin inhibition alone has been recognized as a key potentiator of hypertension (Hoorn et al., 2012). As such, non-immunosuppressive FKBP ligands should have fewer side effects, resulting in safer and more selective pharmacological BMP agonists. Moreover, FKBP52 inhibition by FK506 has been shown to augment nerve regeneration (Gold, 1999; Gold et al., 1999), suggesting that the effect may synergize with FKBP12–mediated tissue regeneration through BMP signaling. In addition to WH, FK506 has also been shown to be effective against pulmonary arterial hypertension and invasiveness in bladder cancer through activation of BMPR signaling (Spiekerkoetter et al., 2013; Shin et al., 2014). It is possible that FKVP and other analogs may find a use in treating these conditions.

That FKVP is as effective in the enhancement of WH as FK506 also has important clinical implications due to its lack of immunosuppressive effect. In patients with a greater risk of infection, such as those with diabetes, treatment with FKVP will provide effective treatment without the risks associated with immune-suppressants. This is highlighted in our use of GK rats for this study: a rat model that spontaneously develops type 2 diabetes after 3–4 months of age and suffers from many of the same physiological manifestations that affect humans with the disease, including significantly impaired WH. By using such a model for this study, we aim to convey the power of this treatment and its application to clinical use. Our demonstration that FKVP recapitulates the WH efficacy of FK506 in the challenging rat model makes FKVP an attractive lead compound, and it is expected that other non-immunosuppressive ligands devoid of inhibitory activity toward calcineurin similar to FKVP will possess similar beneficial effects. Furthermore, the synergistic activities of AMD3100 and FKVP demonstrate a regenerative therapy that could potentially be applied to several other types of tissue damage. Beyond WH, our lab has shown improved liver regeneration after partial hepatectomy and AF combination treatment (Zhai et al., 2018). Thus, FKVP and other non-immunosuppressive FKBP12 ligands may find use in both WH and regenerative therapies.

STAR★METHODS

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact Jun O. Liu (joliu@jhu.edu).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Cell Culture and Transfections

Jurkat E6.1(ATCC, male) cells were cultured in RPMI with 10% FBS and 1.5% PennStrep. Jurkat cells (1 × 106) were transfected with 10 μg of BRE-Luciferase (kindly provided by Martine Roussel & Peter ten Dijke) or NFAT-Luciferase cDNA by electroporation (BioRad, square-wave, 250V, 950μF) in 400 μL serum/antibiotic free RPMI with 0.5% DMSO. Thirty minutes after transfection, cells were transferred to complete RPMI and rested overnight. Before plating, cells were re-suspended in fresh media and diluted to0.5×106 cells/mL. HEK293T (ATCC, fetal) cells were cultured in DMEM with 10% FBS, 1% PennStrep, and 500 μg/mL G418 (Corning). Cells were transfected using SuperFect reagent and supplied transfection protocols. HUVEC (Lonza, donor pooled) cells were cultured in Lonza Endothelial Cell Growth Medium (EGM-2) and used between passages 3 and 7. All cells were cultured at 37°C with 5% CO2.

Jurkat lymphoblasts were originally cultured from the peripheral blood of a 14 year old male with T cell leukemia. These immortal cells have since been cloned to obtain the current line (E6.1). Hek293T cells were first obtained from female human embryonic kidney and immortalized with large T antigen. Clone 17 was chosen for its high transfection efficiencies (ATCC).HUVEC cells are a primary cell line obtained from human umbilical cord. All HUVEC cultures were “donor pooled” via ATCC, and therefore may contain both male and female cells. All cell lines were authenticated by manufacturer prior to distribution. Cell cultures were monitored daily for morphological and growth rate changes.

Declaration of Ethical Animal Care and Use

Goto-Kakizaki (GK) rats were obtained from Taconic (Hudson, NY) and housed in a pathogen-free facility and cared for according to NIH guidelines and a protocol approved by the Johns Hopkins University Animal Care and Use Committee (ACUC). Both male and female GK rats at age of 4–5 months were used in this study.

Diabetic Animal Model

GK rats represent a non-obese model of Non-Insulin Dependent Diabetes Mellitus (NIDDM), type II diabetes. It exhibits similar metabolic, hormonal, and vascular disorders to the human diabetes disease. Males develop type II diabetes at approximately 14 – 16 weeks of age. GK rats are continuously inbred to ensure development of spontaneous hyperglycemia, and were not subject to genotyping by the distributor (Taconic).

Rat Strain Profile: Aco1b, Alpa, Amy2a, Anpepa, Es1a, Es2a, Es3a, Es4a, Es4b, Es6a, Es7b, Es9a, Esda, Es14b, Es15a, Es18a, Fh1b, Hao1a, Hbba, Pepcb, Pgd1b, Svp1a.

Animal Housing and Diet

Animal room temperature and humidity was controlled to 22 ±2°C and 40% (range 30%70%). Light was controlled by timer, turned on at 7 am and off at 9 pm. Animal diet was Teklad 2018SX Global 18% protein extruded rodent diet (Envigo, Frederick, MD).

METHOD DETAILS

Cell Viability Assays

Jurkat or HUVEC cells were plated at 1000 cells/well in 180 μL growth media before addition of 20 μL of 10× drug/protein stock. After incubation for 72 h, 22 μL of a resazurin sodium salt solution (0.1 mg/mL stock in water) was added to each well and allowed to incubate at 37°C. The metabolic conversion of resazurin dye was monitored by absorbance at 570 nm after 6 h. After background subtraction (media only + dye), absorbance values were left as arbitrary absorbance units or normalized to those obtained from cells treated with DMSO.

Western Blot

Jurkat T cells were collected by centrifugation (300g, 5 min), washed with PBS, and lysed in RIPA buffer containing protease and phosphatase inhibitors (Cell Signaling) with sonication. Lysates were normalized using DC assay (BioRad) and run on SDS-PAGE gels. Proteins were transferred to nitrocellulose membranes overnight at 100mA. After blocking with 5% milk in TBS-T for 20 min, membranes were incubated overnight at 4°C with primary antibodies (Table S1). After washing three times with TBS-T, membranes were incubated with secondary antibody (Table S1) for 1 hour. After 3 additional washes, blots were visualized using SynGene, either using ECL substrate (Thermo) or laser excitation and filter (647nm).

FKBP12-SNAP Pull-Downs

FKBP12-SNAP was cloned using pSNAPf vector (New England Bio) and PCR-amplified FKBP12 (a gift from Tobias Meyer) with added EcoRI and BsrGI restriction sites. Gel-purified plasmids were ligated using T4 DNA ligase (Thermo), transformed into DH5α, and plated on LB-agar plus ampicillin for single colony selection, sequencing, and plasmid purification.

Alk1-V5 plasmid was generated by gateway cloning of pDONR223-ACVRL1(Alk1) (a gift from William Hahn & David Root) and pEF-DEST51 vectors using LR Clonase enzyme mix (Thermo). Plasmid was transformed into DH5α and plated on LB-agar plus ampicillin for single colony selection, sequencing, and plasmid purification. HA-tagged Alk2, Alk3 and Alk6 plasmids were a gift from Aristidis Moustakas.

FKBP12-SNAP (5μg) and tagged ALK receptor plasmids (5μg) were co-transfected into Hek293T cells using SuperFect and supplied protocol. After 48 hours, cells were treated with DMSO (0.1%) or 1 μM FKVP for 30 min. After 1 hour, cells were lysed in vessel with 1mL lysis buffer (150mM NaCl, 50mM Tris-HCl, 0.1% Trition-100, 5% glycerol, protease and phosphatase inhibitors) and plate scraper. Lysate was transferred to 2mL eppendorf and rotated at 4C for 20min. Lysates were centrifuged at 14000g for 10min, and ~1mL supernatant was transferred to a new tube with 200nM drug or 0.2% DMSO. 20uL input was taken and mixed with 20uL 2× loading buffer before boiling. Each lysate sample was then mixed with 250uL of SNAP buffer (lysis buffer + 5mM DTT) containing 40μL of magnetic SNAP-capture beads and rotated at 4°C for 1 hour. Beads were washed 3 times with 1mL lysis buffer before boiling in 100μL 2× loading buffer. Boiled lysates were vortexed and centrifuged, lysate (~90μL) was carefully removed from beads and used for western blotting.

The same method was used for calcineurin-FK506 and mTOR-rapamycin pull-downs after transfection of 10ug FKBP12-SNAP cDNA only.

FKBP Knockout Lines

Jurkat T and Hek293T cells were transfected as previously described with all-in-one CRISPR/Cas9 (mCherry tagged) plasmids containing guide-RNAs for FKBP12, FKBP51, or FKBP52. After 48 hours, cells were sorted for mCherry fluorescence (650nm laser) into 96-well plates (1 cell/well). After 2 weeks of culture, single clones were validated by western blotting.

BMP and NFAT Pathway Reporters

Jurkat T cells transfected with BRE-Luc plasmid were split into a 96-well plate (80 μL/well of 0.5×106 cells/mL)) and treated with previously stated compounds/proteins (20 μL of 5× stock in RPMI, 0.5% DMSO) for 18 h. rBMP-4 and rTGF-β1 were used as positive and negative controls, respectively. Plates were centrifuged at 3000rpm for 10min, then carefully aspirated. Cells were re-suspended in 100 μL lysis buffer (per well) and placed on a plate-shaker for 30 min. An aliquot of 80 μL of lysate was transferred to a white-walled 96-well plate, and luminescence was recorded 2 seconds after automated injection of luciferase substrate. Luminescence values were background subtracted (lysis buffer + substrate) and normalized to DMSO control values.

FKBP12KO Jurkat T cells transfected with FKBP12-SNAP plasmid were selected with 1200 μg/mL G418 for seven days before use in BRE-Luc assays.

Jurkat T cells transfected with NFAT-Luc plasmid were split into a 96-well plate (80 μL/well of 0.5×106 cells/mL) and treated with indicated compounds (20 μL of 5× stock in RPMI, 0.5% DMSO) 30 min before activation with phorbol myristate acetate (PMA) and ionomycin (40 nM/1 μM). After 6 h, wells were aspirated, lysed, and measured for luminescence as previously described. FK506 and CsA served as positive control while DMSO and non-activated wells gave negative control values. With the exception of knockout cell experiments, Jurkat cells used for each experiment were transfected at the same time and cultured together overnight until plating and treatment the following day.

FKVP Synthesis and Formulation for Animal Studies

To a solution of FK506 (100mg, 0.120 mmol) and 40mol% Zhan1b catalyst RuCl2[C21H26N2][C12H17NO3S], in 3mL anhydrous DCE was added 4-vinylpyridine (14.2 μL, 0.132 mmol). The mixture was stirred for 30sec before microwave irradiation at 120°C for 20 mins. The mixture was then purified using flash chromatography (0–25% MeOH in DCM), preparative-TLC (9:1 DCM:MeOH), and reverse-phase HPLC (45–85% ACN in H2O). Conversion=25%, Purified Yield=8%. LC-MS and 1H-NMR experiments were used to confirm the synthesized compound was >99% free of the parent compound (FK506). Product was characterized using Hi-Resolution MS (Figure S5A) and 1H-NMR (Figure S5B), then dissolved into DMSO or used in formulation for animal experiments.

For animal experiments, FKVP powder was dissolved into 80% EtOH/20% Cremophor RH60 solution at 5mg/mL. This stock was diluted 1:50 into saline before subcutaneous injection.

In Vivo Excisional Wound Model

Full-thickness wounds were created in the dorsal skin of rats with a sterile disposable biopsy punch (8 mm in diameter). The animals were injected subcutaneously with saline, AMD3100 (1mg/kg) plus FK506 (0.1mg/kg) or AMD3100 (1mg/kg) plus FKVP (0.1mg/kg) immediately after wounding and every other day until complete healing, defined as complete re-epithelialization of the wound area. For assessing the role of BMP signaling, animals were injected intraperitoneally (i.p.) with LDN-193189 (2mg/kg/day) in addition to standard saline or AF treatment. Wounds were evaluated daily according to the method described previously (Lin et al., 2014).

Immunohistochemistry

Cut sections of 5 μm were prepared from frozen tissue for immunohistochemistry staining. Frozen sections were fixed with acetone at −20°C for 10 min and dried for 1 h at room temperature. After inactivation of endogenous peroxidase and blocking of nonspecific antibody binding, the specimens were incubated with anti-CD133 (Table S1) at 4°C overnight. The tissue sections were then subsequently incubated with biotin-conjugated goat anti-rabbit IgG (Table S1) for 30 minutes at room temperature. The VectaStain Elite ABC kit (HRP) (Vector Laboratories, Burlingame, CA) was used to increase the sensitivity of the staining. Diaminobenzidine tetrahydrochloride (5 min) was used as the chromogen, and Mayer’s Hematoxylin (30 s) was used for counterstaining.

QUANTIFICATION AND STATISTICAL ANALYSIS

Applied Software

GraphPad Prism 6 software was used for in vivo statistical analysis and in vitro standard deviation calculations.

In Vitro Statistics

For in vitro cell assays, n refers to number of replicate wells analyzed for each treatment group. Error bars for in vitro assay figures represent standard deviation from mean.

In Vivo Statistics

For animal models, n refers to the number of rats used in each treatment group. Four wounds per animal were separately measured by days to re-epithelialization for use as the main parameter. The one-way analysis of variance (ANOVA) was used to determine the statistical difference in wound healing among AF, AF+LDN, Saline and S+LDN groups or between AF, AV and Saline groups when comparing days of wound healing. Bonferroni-Holm post-hoc procedure was used for p value adjustment. p<0.05 is considered significantly different. Error bars for in vivo assay figures represent Standard Error of Mean (SEM).

DATA AND SOFTWARE AVAILABILITY

None reported.

Supplementary Material

S1

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
(See Table S1)
Bacterial and Virus Strains
DH5α competent cells ThermoFisher 18265017
Chemicals, Peptides, and Recombinant Proteins
FK506 (Tacrolimus) ApexBio B2143
4-Vinylpyridine Sigma V3204
Zhan-1b ruthenium catalyst Strem 44–0082
LDN-193189 Adooq Bioscience A11478
AMD3100 Sigma A5602
Resazurin Sigma R7017
BMP-4 R&D Systems 314-BP
TGF-B1 R&D Systems 240-B
Noggin R&D Systems 6057-NG
Diaminobenzidine tetrahydrochloride Sigma D4293
Mayer’s Hematoxylin Dako S3309
Critical Commercial Assays
SNAP Magnetic Beads New England Bio S9145S
Experimental Models: Cell Lines
Jurkat Clone E6.1 (Male, Human T Lymphocyte) ATCC TIB-152
HUVEC (Donor pooled, Human Endothelial) Lonza C2519A
Hek293T/17 (Fetal, Human Kidney Epithelial) ATCC CRL-11268
Experimental Models: Organisms/Strains
Goto-Kakazaki rats (Male and Female) Taconic GK/MolTac
Recombinant DNA
BRE-Luc Addgene 45126
NFAT-Luc Promega E848A
pSNAPf New England Bio N9183S
FKBP12-YFP Addgene 20175
pEF-DEST51 ThermoFisher 12285011
Alk1 (pDONR223-ACVRL1) Addgene 23873
Alk2-HA Addgene 80870
Alk3-HA Addgene 80873
Alk6-HA Addgene 80882
FKBP12KO Plasmid Genecopeia HCP267023-CG01–3-B
FKBP51KO Plasmid Genecopeia HCP257374-CG01–3-B
FKBP52KO Plasmid Genecopeia HCP205551-CG01–3-B
Software and Algorithms
Prism 6 GraphPad https://www.graphpad.com/scientific-software/prism/
Open in a new tab

SIGNIFICANCE.

We have demonstrated the utility of non-immunosuppressive FK506 analogs as therapeutic leads for tissue repair and regeneration. In addition, we have shown that FKBP12 inhibition alone is sufficient for BMP pathway activation by FK506 or FKVP. BMP dysfunction has been reported to cause disease as well as cancer progression, and compounds such as FKVP represent a class of FKBP-selective, small-molecule BMP agonists with modest potency. When combined with AMD3100, FKVP reduces average healing time in diabetic rats by over 20% after full-thickness excision, and systemic inhibition of BMP receptor kinases with LDN-193189 (LDN) abolishes combination treatment effects. Further study revealed that wounds of rats treated with LDN contained significantly fewer stem cells, suggesting that BMP signaling may be involved in the recruitment of endogenous stem and progenitor cells to injured tissues. Research into BMP pathway products and signaling components within the context of these cells may improve aspects of stem cell therapy, as BMP stimulation may also improve homing of exogenous cells to sites of injury.

Highlights.

  • Accelerated wound healing promoted by FK506 is independent of calcineurin inhibition

  • A non-immunosuppressive FK506 analog activates BMP pathway signaling in human cells

  • Blockade of BMP signaling abolishes acceleration of healing by combination treatment

  • BMP receptor inhibition prevents stem cell recruitment at wound sites in vivo

ACKNOWLEDGMENTS

This work was supported by the Liu Lab Discretionary Fund, FAMRI, NCI (P30CA006973) (J.O.L.), NIH-funded Chemistry-Biology Interface Program at Johns Hopkins University (T32 GM080189) (B.P.), and a grant from MedRegen (Z.S.).

Footnotes

SUPPLEMENTAL INFORMATION

Supplemental Information includes five figures and one table and can be found with this article online at https://doi.org/10.1016/j.chembiol.2019.01.011.

DECLARATION OF INTERESTS

The authors declare no conflict of interest.

REFERENCES

  1. Abraham RT, and Wiederrecht GJ (1996). Immunopharmacology of rapamycin. Annu. Rev. Immunol 14, 483–510. [DOI] [PubMed] [Google Scholar]
  2. Albiñana V, Sanz-Rodríguez F, Recio-Poveda L, Bernabéu C, and Botella LM (2011). Immunosuppressor FK506 increases endoglin and activin receptor-like kinase 1 expression and modulates transforming growth factor-β1 signaling in endothelial cells. Mol. Pharmacol 79, 833–843. [DOI] [PubMed] [Google Scholar]
  3. Balaji S, King A, Crombleholme TM, and Keswani SG (2013). The role of endothelial progenitor cells in postnatal vasculogenesis: implications for therapeutic neovascularization and wound healing. Adv. Wound Care (New Rochelle) 2, 283–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bechstein WO (2000). Neurotoxicity of calcineurin inhibitors: impact and clinical management. Transpl. Int 13, 313–326. [DOI] [PubMed] [Google Scholar]
  5. Bennett J, Cassidy H, Slattery C, Ryan MP, and McMorrow T (2016). Tacrolimus modulates TGF-β signaling to induce epithelial-mesenchymal transition in human renal proximal tubule epithelial cells. J. Clin. Med 5, E50, 10.3390/jcm5050050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bueno OF, Brandt EB, Rothenberg ME, and Molkentin JD (2002). Defective T cell development and function in calcineurin Abeta-deficient mice. Proc. Natl. Acad. Sci. U S A 99, 9398–9403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cameron AM, Steiner JP, Sabatini DM, Kaplin AI, Walensky LD, and Snyder SH (1995). Immunophilin FK506 binding protein associated with inositol 1,4,5-trisphosphate receptor modulates calcium flux. Proc. Natl. Acad. Sci. U S A 92, 1784–1788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cameron AM, Wesson RN, Ahmadi AR, Singer AL, Hu X, Okabayashi T, Wang Y, Shigoka M, Fu Y, Gao W, et al. (2016). Chimeric allografts induced by short-term treatment with stem cell mobilizing agents result in long-term kidney transplant survival without immunosuppression: II, study in miniature swine. Am. J. Transpl 16, 2066–2076. [DOI] [PubMed] [Google Scholar]
  9. Chaikuad A, Alfano I, Kerr G, Sanvitale CE, Boergermann JH, Triffitt JT, von Delft F, Knapp S, Knaus P, and Bullock AN (2012). Structure of the bone morphogenetic protein receptor ALK2 and implications for fibro-dysplasia ossificans progressiva. J. Biol. Chem 287, 36990–36998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen YG, Liu F, and Massague J (1997). Mechanism of TGFbeta receptor inhibition by FKBP12. EMBO J. 16, 3866–3876. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Clemons PA, Gladstone BG, Seth A, Chao ED, Foley MA, and Schreiber SL (2002). Synthesis of calcineurin-resistant derivatives of FK506 and selection of compensatory receptors. Chem. Biol 9, 49–61. [DOI] [PubMed] [Google Scholar]
  12. Csiszar A, Ahmad M, Smith KE, Labinskyy N, Gao Q, Kaley G, Edwards JG, Wolin MS, and Ungvari Z (2006). Bone morphogenetic protein-2 induces proinflammatory endothelial phenotype. Am. J. Pathol 168, 629–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cuny GD, Yu PB, Laha JK, Xing X, Liu J-F, Lai CS, Deng DY, Sachidanandan C, Bloch KD, and Peterson RT (2008). Structure-activity relationship study of bone morphogenetic protein (BMP) signaling inhibitors. Bioorg. Med. Chem. Lett 18, 4388–4392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fife CE, and Carter MJ (2012). Wound care outcomes and associated cost among patients treated in US outpatient wound centers: data from the US wound registry. Wounds 24, 10–17. [PubMed] [Google Scholar]
  15. Fung JJ (2004). Tacrolimus and transplantation: a decade in review. Transplantation 77, S41–S43. [DOI] [PubMed] [Google Scholar]
  16. Giordano A, Romano S, Mallardo M, D’Angelillo A, Calì G, Corcione N, Ferraro P, and Romano MF (2008). FK506 can activate transforming growth factor-beta signalling in vascular smooth muscle cells and promote proliferation. Cardiovasc. Res 79, 519–526. [DOI] [PubMed] [Google Scholar]
  17. Gold BG (1999). FK506 and the role of the immunophilin FKBP-52 in nerve regeneration. Drug Metab. Rev 31, 649–663. [DOI] [PubMed] [Google Scholar]
  18. Gold BG, Densmore V, Shou W, Matzuk MM, and Gordon HS (1999). Immunophilin FK506-binding protein 52 (not FK506-binding protein 12) mediates the neurotrophic action of FK506. J. Pharmacol. Exp. Ther 289, 1202–1210. [PubMed] [Google Scholar]
  19. Griffith JP, Kim JL, Kim EE, Sintchak MD, Thomson JA, Fitzgibbon MJ, Fleming MA, Caron PR, Hsiao K, and Navia MA (1995). X-ray structure of calcineurin inhibited by the immunophilin-immunosuppressant FKBP12-FK506 complex. Cell 82, 507–522. [DOI] [PubMed] [Google Scholar]
  20. Hatse S, Princen K, Bridger G, De Clercq E, and Schols D (2002). Chemokine receptor inhibition by AMD3100 is strictly confined to CXCR4. FEBS Lett. 527, 255–262. [DOI] [PubMed] [Google Scholar]
  21. Hoorn EJ, Walsh SB, McCormick JA, Zietse R, Unwin RJ, and Ellison DH (2012). Pathogenesis of calcineurin inhibitor-induced hypertension. J. Nephrol 25, 269–275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hu X, Okabayashi T, Cameron AM, Wang Y, Hisada M, Li J, Raccusen LC, Zheng Q, Montgomery RA, Williams GM, et al. (2016). Chimeric allografts induced by short-term treatment with stem cell-mobilizing agents result in long-term kidney transplant survival without immunosuppression: a study in rats. Am. J. Transpl 16, 2055–2065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jayaraman T, Brillantes AM, Timerman AP, Fleischer S, Erdjument-Bromage H, Tempst P, and Marks AR (1992). FK506 binding protein associated with the calcium release channel (ryanodine receptor). J. Biol. Chem 267, 9474–9477. [PubMed] [Google Scholar]
  24. Jujo K, Hamada H, Iwakura A, Thorne T, Sekiguchi H, Clarke T, Ito A, Misener S, Tanaka T, Klyachko E, et al. (2010). CXCR4 blockade augments bone marrow progenitor cell recruitment to the neovasculature and reduces mortality after myocardial infarction. Proc. Natl. Acad. Sci. U S A 107, 11008–11013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kissinger CR, Parge HE, Knighton DR, Lewis CT, Pelletier LA, Tempczyk A, Kalish VJ, Tucker KD, Showalter RE, and Moomaw EW (1995). Crystal structures of human calcineurin and the human FKBP12-FK506-calcineurin complex. Nature 378, 641–644. [DOI] [PubMed] [Google Scholar]
  26. Kozany C, März A, Kress C, and Hausch F (2009). Fluorescent probes to characterise FK506-binding proteins. Chembiochem 10, 1402–1410. [DOI] [PubMed] [Google Scholar]
  27. Lewis CJ, Mardaryev AN, Poterlowicz K, Sharova TY, Aziz A, Sharpe DT, Botchkareva NV, and Sharov AA (2014). Bone morphogenetic protein signaling suppresses wound-induced skin repair by inhibiting keratinocyte proliferation and migration. J. Invest. Dermatol 134, 827–837. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liles WC, Broxmeyer HE, Rodger E, Wood B, Hübel K, Cooper S, Hangoc G, Bridger GJ, Henson GW, Calandra G, et al. (2003). Mobilization of hematopoietic progenitor cells in healthy volunteers by AMD3100, a CXCR4 antagonist. Blood 102, 2728–2730. [DOI] [PubMed] [Google Scholar]
  29. Lin Q, Wesson RN, Maeda H, Wang Y, Cui Z, Liu JO, Cameron AM, Gao B, Montgomery RA, Williams GM, et al. (2014). Pharmacological mobilization of endogenous stem cells significantly promotes skin regeneration after full-thickness excision: the synergistic activity of AMD3100 and tacrolimus. J. Invest. Dermatol 134, 2458–2468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu J, Farmer JD, Lane WS, Friedman J, Weissman I, and Schreiber SL (1991). Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes. Cell 66, 807–815. [DOI] [PubMed] [Google Scholar]
  31. Naesens M, Kuypers DRJ, and Sarwal M (2009). Calcineurin inhibitor nephrotoxicity. Clin. J. Am. Soc. Nephrol 4, 481–508. [DOI] [PubMed] [Google Scholar]
  32. Okabayashi T, Cameron AM, Hisada M, Montgomery RA, Williams GM, and Sun Z (2011). Mobilization of host stem cells enables long-term liver transplant acceptance in a strongly rejecting rat strain combination. Am. J. Transplant 11, 2046–2056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Plikus MV, Guerrero-Juarez CF, Ito M, Li YR, Dedhia PH, Zheng Y, Shao M, Gay DL, Ramos R, Hsi T-C, et al. (2017). Regeneration of fat cells from myofibroblasts during wound healing. Science 355, 748–752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rao A, Luo C, and Hogan PG (1997). Transcription factors of the NFAT family: regulation and function. Annu. Rev. Immunol 15, 707–747. [DOI] [PubMed] [Google Scholar]
  35. Schäffer MR, Fuchs N, Proksch B, Bongartz M, Beiter T, and Becker HD (1998). Tacrolimus impairs wound healing: a possible role of decreased nitric oxide synthesis. Transplantation 65, 813–818. [DOI] [PubMed] [Google Scholar]
  36. Shin K, Lim A, Zhao C, Sahoo D, Pan Y, Spiekerkoetter E, Liao JC, and Beachy PA (2014). Hedgehog signaling restrains bladder cancer progression by eliciting stromal production of urothelial differentiation factors. Cancer Cell 26, 521–533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Spiekerkoetter E, Tian X, Cai J, Hopper RK, Sudheendra D, Li CG, El-Bizri N, Sawada H, Haghighat R, Chan R, et al. (2013). FK506 activates BMPR2, rescues endothelial dysfunction, and reverses pulmonary hypertension. J. Clin. Invest 123, 3600–3613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sun CC, Vaja V, Chen S, Theurl I, Stepanek A, Brown DE, Cappellini MD, Weiss G, Hong CC, Lin HY, et al. (2013). A hepcidin lowering agent mobilizes iron for incorporation into red blood cells in an adenine-induced kidney disease model of anemia in rats. Nephrol. Dial. Transplant 28, 1733–1743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Tanaka H, Kuroda A, Marusawa H, Hatanaka H, Kino T, Goto T, Hashimoto M, and Taga T (1987). Structure of FK506, a novel immunosuppressant isolated from streptomyces. J. Am. Chem. Soc 109, 5031–5033. [Google Scholar]
  40. Tateishi K, Higuchi C, Ando W, Nakata K, Hashimoto J, Hart DA, Yoshikawa H, and Nakamura N (2007). The immunosuppressant FK506 promotes development of the chondrogenic phenotype in human synovial stromal cells via modulation of the Smad signaling pathway. Osteoarthritis Cartilage 15, 709–718. [DOI] [PubMed] [Google Scholar]
  41. Wang T, Li BY, Danielson PD, Shah PC, Rockwell S, Lechleider RJ, Martin J, Manganaro T, and Donahoe PK (1996). The immunophilin FKBP12 functions as a common inhibitor of the TGF beta family type I receptors. Cell 86, 435–444. [DOI] [PubMed] [Google Scholar]
  42. Young K, Conley B, Romero D, Tweedie E, O’Neill C, Pinz I, Brogan L, Lindner V, Liaw L, and Vary CPH (2012). BMP9 regulates endoglin-dependent chemokine responses in endothelial cells. Blood 120, 4263–4273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhai R, Wang Y, Qi L, Williams GM, Gao B, Song G, Burdick JF, and Sun Z (2018). Pharmacological mobilization of endogenous bone marrow stem cells promotes liver regeneration after extensive liver resection in rats. Sci. Rep 8, 3587. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1
NIHMS1068956-supplement-S1.pdf (980KB, pdf)

Data Availability Statement

None reported.

RESOURCES