Granulocytes and vascularization regulate uterine bleeding and tissue remodeling in a mouse menstruation model.

Menstruation-associated disorders negatively interfere with the quality of life of many women. However, mechanisms underlying pathogenesis of menstrual disorders remain poorly investigated up to date. Among others, this is based on a lack of appropriate pre-clinical animal models. We here employ a mouse menstruation model induced by priming mice with gonadal hormones and application of a physical stimulus into the uterus followed by progesterone removal. As in women, these events are accompanied by menstrual-like bleeding and tissue remodeling processes, i.e. disintegration of decidualized endometrium, as well as subsequent repair. We demonstrate that the onset of bleeding coincides with strong upregulation of inflammatory mediators and massive granulocyte influx into the uterus. Uterine granulocytes play a central role in regulating local tissue remodeling since depletion of these cells results in dysregulated expression of matrix modifying enzymes. As described here for the first time, uterine blood loss can be quantified by help of tampon-like cotton pads. Using this novel technique, we reveal that blood loss is strongly reduced upon inhibition of endometrial vascularization and thus, is a key regulator of menstrual bleeding. Taken together, we here identify angiogenesis and infiltrating granulocytes as critical determinants of uterine bleeding and tissue remodeling in a mouse menstruation model. Importantly, our study provides a technical and scientific basis allowing quantification of uterine blood loss in mice and thus, assessment of therapeutic intervention, proving great potential for future use in basic research and drug discovery.

Induction of overt menstruation in intact mice.

The complex tissue remodeling process of menstruation is experienced by humans and some primates, whereas most placental mammals, including mice, go through an estrous cycle. How menstruation and the underlying mechanisms evolved is still unknown. Here we demonstrate that the process of menstruation is not just species-specific but also depends on factors which can be induced experimentally. In intact female mice endogenous progesterone levels were raised by the induction of pseudopregnancy. Following an intrauterine oil injection, the decidualization of the endometrium was reliably induced as a prerequisite for menstruation. The natural drop of endogenous progesterone led to spontaneous breakdown of endometrial tissue within an average of 3 days post induction of decidualization. Interestingly, morphological changes such as breakdown and repair of the endometrial layer occurred in parallel in the same uterine horn. Most importantly, endometrial breakdown was accompanied by vaginally visible (overt) bleeding and flushing out of shed tissue comparable to human menstruation. Real-time PCR data clearly showed temporal changes in the expression of multiple factors participating in inflammation, angiogenesis, tissue modulation, proliferation, and apoptosis, as has been described for human menstruating endometrium. In conclusion, human menstruation can be mimicked in terms of extravaginally visible bleeding, tissue remodeling, and gene regulation in naturally non-menstruating species such as intact female mice without the need for an exogenous hormone supply.

Accumulation and local proliferation of antigen-specific CD4+ T cells in antigen-bearing tissue.

Although activation and subsequent expansion of naive CD4(+) T cells within lymph nodes is well characterized, the fate of T effector cells activated within peripheral tissues during secondary reactions is poorly defined. Therefore, we studied the recruitment, proliferation and egress of antigen-specific Th1 effector cells in comparison with nonspecific Th1 cells throughout a delayed-type hypersensitivity reaction (DTH). Although we observed a high turnover of Th1 effector cells with unspecific high-rate recruitment and CCR7-dependent egress from the inflamed tissue in the early, acute DTH phase, a strong, selective accumulation of antigen-specific T cells occurred during the chronic, late DTH phase. This was mainly based on local proliferation of CD4(+) effector cells within the DTH tissue and concomitant retention. Considering the strong CCR7-dependent Th cell egress found in this model, the reduced CCR7 expression on antigen-specific T cells isolated from late-phase DTH tissue most likely contributes to the retention of these cells within the tissue. Thus, peripheral tissues can support not only the proliferation of CD8(+) T cells, as recently shown, but also that of CD4(+) T effector cells, forming a pool of tissue-resident T cells.

Persistence of effector memory Th1 cells is regulated by Hopx.

Th1 cells are prominent in inflamed tissue, survive conventional immunosuppression, and are believed to play a pivotal role in driving chronic inflammation. Here, we identify homeobox only protein (Hopx) as a critical and selective regulator of the survival of Th1 effector/memory cells, both in vitro and in vivo. Expression of Hopx is induced by T-bet and increases upon repeated antigenic restimulation of Th1 cells. Accordingly, the expression of Hopx is low in peripheral, naïve Th cells, but highly up-regulated in terminally differentiated effector/memory Th1 cells of healthy human donors. In murine Th1 cells, Hopx regulates the expression of genes involved in regulation of apoptosis and survival and makes them refractory to Fas-induced apoptosis. In vivo, adoptively transferred Hopx-deficient murine Th1 cells do not persist. Consequently, they cannot induce chronic inflammation in murine models of transfer-induced colitis and arthritis, demonstrating a key role of Hopx for Th1-mediated immunopathology.

Retinoic acid-induced gut tropism improves the protective capacity of Treg in acute but not in chronic gut inflammation.

Treg are endowed with immunosuppressive activities and have been proposed as promising targets for the therapy of autoimmune diseases. As the suppressive capacity of Treg depends on their migration into the affected tissues, we tested here whether modulation of Treg homing would enhance their capacity to suppress inflammation in mouse models of inflammatory bowel disease. Retinoic acid (RA) was used to induce the gut-specific homing receptor alpha(4)beta(7) efficiently and, to some extent, the chemokine receptor CCR9 on in vitro expanded Treg. Upon transfer, RA-treated Treg were indeed more potent suppressors in an acute, small intestinal inflammation model, compared with Treg stimulated without RA. By contrast, the efficacy of Treg to resolve an established, chronic inflammation of the colon in the transfer colitis model was not affected by RA-treatment. In the latter model, a rapid loss of RA-induced alpha(4)beta(7) expression and de novo induction of alpha(4)beta(7) on previously negative cells was observed on transferred Treg, which implies that Treg acquire gut-seeking properties in vivo under inflammatory and/or lymphopenic conditions. Together, our data show that the induction of appropriate homing properties prior to transfer increases the protective potential of adoptively transferred Treg in acute, but not in chronic, inflammatory disorders of the gut.

Generation and application of a fluorescein-specific single chain antibody.

A recombinant single chain antibody fragment (designated scDE1) of the murine monoclonal anti-fluorescein antibody B13-DE1 was generated using the original hybridoma cells as source for the variable antibody heavy and light chain (VH and VL) genes. After cloning the variable genes into a phage vector a functional antibody fragment was selected by phage display panning. Recombinant antibody could be expressed as phage antibody and as soluble single chain antibody in Escherichia coli. High yield of scDE1 could also be detected in bacterial culture supernatant. The scDE1 showed the same binding specificity as the parental monoclonal antibody, i.e. it bound fluorescein, fluorescein derivatives and a fluorescein peptide mimotope. Surface plasmon resonance revealed a K(D) of 19 nM for the scDE1 compared to 0.7 nM for the monoclonal antibody. The isolated soluble scDE1 could easily be conjugated to horseradish peroxidase which allowed the use of the conjugate as universal indicator for the detection of fluorescein-labelled proteins in different immunoassays. Detection of hCG in urine was performed as a model system using scDE1. In addition to E. coli the scFv genes could also be transferred and expressed in eukaryotic cells. Finally, we generated HEK293 cells expressing the scDE1 at the cell surface.

Distinctive role of CCR7 in migration and functional activity of naive- and effector/memory-like Treg subsets.

Foxp3+CD25+CD4+ Treg play a fundamental role in the maintenance of self tolerance and the control of inflammatory reactions. Previous data demonstrated a division of labor between naive- and effector/memory-like Treg subsets, which is largely based on their lymph node-recirculating and inflammation-seeking migration behavior, respectively. The chemokine receptor CCR7 is expressed on both types of Treg subsets, albeit at different levels. Whether it fulfills similar or distinct roles in these subsets has not been studied so far. We here show that the recirculation of naive-like Treg through LN and, to some extent, the gut is dependent on CCR7. Lack of CCR7 not only prevents recirculation, but also almost completely abolishes the ability of naive-like Treg to control the priming phase of an immune response. In contrast, CCR7 deficiency in effector/memory-like Treg promotes their accumulation in inflamed sites, compatible with a role of CCR7 for exit from the tissue. Local Treg accumulation was accompanied by an enhanced suppression of inflammation. Together, our findings provide conclusive evidence that CCR7 expression on Treg differentially controls in vivo function of the naive- and effector/memory-like subsets.

Induction of organ-selective CD4+ regulatory T cell homing.

Compelling evidence suggests that Foxp3(+)CD25(+)CD4(+) Treg play a fundamental role in immunoregulation. We have previously demonstrated that Treg have to enter peripheral tissues to suppress ongoing inflammation. However, relatively little is known about how Treg acquire the expression of homing receptors required for tissue- or inflammation-specific migration. Migratory properties of conventional naïve T cells are shaped by the tissue microenvironment and organ-specific dendritic cells during priming. Here, we show that this basic concept also holds true for CD25(+)CD4(+) Treg: Priming of Treg within peripheral LN led to the expression of selectin ligands, which facilitate migration into inflamed skin, whereas activation within mesenteric LN led to induction of the integrin alpha(4)beta(7), which is required for migration into mucosal tissues. Furthermore, we could establish in vitro culture systems containing either dendritic cells from mesenteric and peripheral LN, or retinoic acid and IL-12 as polarizing compounds to induce mucosa- and skin-seeking Treg, respectively. Together, our results demonstrate that Treg, similarly to conventional T cells, can be configured with organ-selective homing properties allowing efficient targeting into distinct tissues.