Solid phase cytometry for detection of rare events.

We describe a simple and rapid laser scanning device for solid phase cytometry. This system can detect and count fluorescent cells over a 22 mm diameter surface (membrane or glass circle) as well as quantify the fluorescence that they emit. A comprehensive discrimination package includes optical and software parameters and accurately distinguishes between valid signals (e.g., labelled cells) and nonspecific signals (e.g., auto-fluorescent particles and debris). Any event detected may also be easily confirmed by visual observation after transfer of the sample to an epifluorescence microscope fitted with a motorized stage driven by the system. We show a linear relationship between the amount of fluorescein coupled to the cells and the fluorescence signal of the cells detected. This approach is not destructive and further characterization of the sample may be carried out. We have been able to detect rare cellular events at a frequency of 10(-7) in 3 min. Potential applications include monitoring of residual disease in oncology and detection of virus-infected cells circulating at very low frequencies.

Solid phase cytometry allows rapid in situ quantification of human papilloma virus infection in biopsy material.

Solid phase cytometry would be an asset for many histological and cytological studies. Current microscope-based cytometers and image analysis systems are too slow to analyze specimens several millimeters wide. We have recently shown that a rapid wide area laser scanning device that operates on solid supports has a linear response. We assess it here for solid phase cytometry. Each cell detected by the cytometer can be automatically positioned for visual observation in the field of an epifluorescence microscope (conventional or confocal) in which the stage is driven by the instrument's computer. We were able to detect and map human papillomavirus-infected cells labeled by fluorescent in situ hybridization in cervical condyloma biopsies. We could quantify the fluorescence emitted by these cells and show differences of up to 35-fold in fluorescence intensity between individual cells. These differences in intensity might reflect differences in viral copy number. The potential of the system to provide fast, reliable and reproducible analyses of solid tissue samples is discussed.

Induction of experimental autoimmune thyroiditis (EAT) by heat-denaturated thyroglobulin (Tg).

Experimental Autoimmune Thyroiditis (EAT) is characterized by autoreactive T and B cell responses, assessed by a marked lymphocytic infiltration of the thyroid gland by T cells and the occurrence of circulating autoantibodies (AAb) to thyroglobulin (Tg). It was recently reported that administration of denaturated exogenous antigens primes class I-restricted cytotoxic T cells in vivo. Since cytotoxic T cells are involved in EAT development, porcine Tg (pTg) was heat-denaturated, i.v. injected into CBA/J mice and features of EAT evaluated. Simultaneously, mice were immunized with pTg and adjuvants and evaluation of EAT performed. We found that heat-denaturated pTg (hdpTg) induced EAT in recipient mice similar to native pTg/adjuvants. Surprisingly, whereas Tg-specific cytotoxic T cells were regularly found in lymph node cells from hdpTg or native pTg immunized mice, proliferative responses were only detected using T cells from native pTg immunized mice. Autoantibodies to pTg were decreased by a factor 30 in sera from mice immunized with hdpTg. These data further emphasized the role of Tg-specific cytotoxic T cells in EAT.

B cell superstimulatory influenza virus activates peritoneal B cells.

We evaluated the potential of B cell "superstimulatory" influenza viruses to activate peritoneal B cells (PBC) from BALB/c mice containing both CD5+ and CD5- "sister" cells. Like conventional B cells, PBCs responded to influenza viruses in a hemagglutinin glycoprotein (HA) subtype-specific manner with proliferation and vigorous Ig synthesis. However, a number of HA subtypes that are highly stimulatory for conventional B cells failed to induce significant responses of PBC. Isotype-determination revealed a high predominance of IgM and only very low production of IgA and IgG. HA-activated CD5+ B cells showed a hyperexpression of various activation markers, including MHC class II, intercellular adhesion molecule 1 (CD54), and B7-1 molecules. In contrast to conventional B cells, where activation by HA is antagonized by phorbol esters (PMA), HA and PMA acted synergistically on PBC, suggesting differential activation requirements of B-2 cells vs PBC in response to HA. Like HA stimulation of B-2 cells, virus-triggered proliferation of PBC was abrogated by a simultaneous treatment with F(ab')2 fragments of anti-Ig Ab and exhibited synergistic effects with LPS stimulation. HA-mediated proliferative responses of PBC, but not of B-2 cells, were positively controlled by various cytokines, including IL-4 and IL-10, and to a lesser extent by IL-6. In conclusion, our data present the first example of a stimulation of peritoneal B cells by a polyclonal-activating virus, findings that call for considering infections with polyclonal B cell-stimulatory viruses as a means of expanding the pool of potentially autoreactive CD5+ B cells.

Curative and protective effects of IL-10 in experimental autoimmune thyroiditis (EAT). Evidence for IL-10-enhanced cell death in EAT.

We studied the effects of in vivo administration of rhIL-10 in two models of experimental autoimmune thyroiditis (EAT): 1)-in EAT induced by injection of mTg emulsified in adjuvant, and 2) in EAT induced by adoptive transfer of mTg-specific T lymphocytes. Furthermore, we tried to assess both the protective and curative potential of IL-10 in EAT, by administering rhIL-10 either at the time of priming and challenge with mTg, or only at the time of challenge. We demonstrated that proliferative and cytotoxic responses of splenic cells to mTg were markedly reduced by in vivo rhIL-10 treatment. Cell surface marker studies revealed a 40 to 45% reduction in CD4+ and in CD8+ lymphoblastoid spleen cells from mice treated with rhIL-10 either in the early or in the late EAT. The severity of EAT was significantly reduced in mice treated with high-dose rhIL-10, whereas levels of autoantibodies to mTg were not altered. Furthermore, when analyzing purified T lymphocytes from rhIL-10-treated animals, an increase of cells undergoing apoptotic cell death became evident in the rhIL-10 treated group, as compared with controls. This IL-10-mediated enhancement of activation-induced cell death critically depended on the applied therapeutic dose of rhIL-10. Thus, IL-10 exerts beneficial effects on the development and course of EAT through a mechanism that could imply an IL-10-mediated enhancement of activation-induced cell death in T lymphocytes, findings that call for considering IL-10 in the immunotherapy of early-phase and likewise of already established autoimmune thyroiditis.

Molecular heterogeneity of antigen- or idiotype-induced anti-thyroglobulin monoclonal autoantibodies.

To define the molecular basis of the cognitive interaction in experimental autoimmune thyroiditis (EAT), we sequenced the variable regions of monoclonal autoantibodies to thyroglobulin (Tg), specific or not for the F40D peptide, a Tg peptide capable of inducing EAT in CBA/J mice. Three MoAbs were obtained by immunization with syngeneic Tg of CBA/J (3B8G9, 2F6F2) or C57Bl/6 (4D11F4) mice. 3B8G9 was specific for F40D peptide, whereas 2F6F2 and 4D11F4 were not. Two others were raised in CBA/J mice by manipulation of idiotypic pathways: B12 resulted from the immunization with one Ab2 beta, bearing the internal image of one F40D epitope, and TA2 from the immunization with F40D-specific cytotoxic HTC2 T cells. B12 and TA2 were both specific for F40D. All hybridomas expressed different members of the J558 VH family, except 3B8G9 which expressed a Q52 VH gene segment. These data led us to hypothesize that regulatory anti-id autoantibodies used members of one VH family located in the 5'-end of the VH locus, whereas EAT-associated autoantibodies used a member of one of the most D-proximal VH family. As expected, no homologies were found when anti-F40D monoclonal autoantibodies were compared with two other monoclonal autoantibodies displaying a different epitopic specificity. Among the anti-F40D monoclonal autoantibodies, one histidine residue located in position 35 of the CDR1 region was constantly found. Moreover, TA2 and B12 exhibited two common amino acids in their CDR3 regions, one glycine and one tyrosine, in positions 98 and 99, respectively. Striking homologies were found between TA2 and one anti-polyGAT MoAb, and between 3B8G9 and some anti-phenyloxazolone (phOx) monoclonal autoantibodies. Lastly, the VK sequence from 4D11F4 was identical at the amino acid level to the VK sequence from another monoclonal autoantibody, 81B1, which was previously raised towards syngeneic Tg in CBA/J mice. Our data imply that anti-idiotypic regulatory circuits in EAT might be generated by a heterogeneous population of B cells rather than obtained by a single dominant B cell population.

Distinctive modulation by IL-4 and IL-10 of the effector function of murine thyroglobulin-primed cells in "transfer-experimental autoimmune thyroiditis".

Experimental autoimmune thyroiditis (EAT) is characterized by autoreactive T and B cell responses, a marked lymphocytic infiltration of the thyroid gland, and the occurrence of circulating autoantibodies to thyroglobulin. Direct evidence for the involvement of lymphocytes stems from the observation that EAT can be induced in naive, irradiated CBA/J mice by transfer of in vitro restimulated effector spleen cells obtained from murine thyroglobulin (mTg)-immunized donors. Using this transfer-EAT (tEAT) model, we have investigated whether addition of recombinant murine IL-4 (rIL-4) or human IL-10 (rIL-10) during the in vitro restimulation by mTg would affect the subsequent induction of the disease. To determine the modification(s) induced during the secondary in vitro incubation with mTg and cytokine, proliferative and cytotoxic responses to mTg were studied. MTg-activated cells cultured with mTg and rIL-4 exhibited only slightly decreased proliferative responses to mTg and increased cytotoxic responses toward mTg-pulsed macrophages compared to mTg-activated cells cultured in the absence of cytokine. In contrast, proliferative and cytotoxic responses to mTg were diminished by approximately 45 and 85%, respectively, when cells were cultured with mTg and rIL-10. The injection of mTg-activated spleen cells, cultured in the presence of rIL-10, into irradiated CBA/J mice induced a significant decrease (P = 0.02) in lymphocytic infiltrations of the recipient thyroid glands compared to injection of irradiated hosts with mTg-activated cells cultured without cytokines, but no reduction in anti-mTg autoantibody production in vivo. In contrast, when mice were injected with mTg-activated cells cultured with mTg and rIL-4, the lymphocytic infiltrations of the recipient thyroid glands were similar to controls, but circulating anti-mTg antibody were surprisingly significantly reduced. These results show that two typical "Th2 cytokines," IL-4 and IL-10, when added in vitro to mTg-specific spleen cells under identical experimental conditions, can have rather diverse effects in terms of an exacerbation or weakening of cytotoxic mTg-specific T cell reactivity and a maintenance or attenuation of subsequent tEAT severity.

Superantigens induce primary T cell responses to soluble autoantigens by a non-V beta-specific mechanism of bystander activation.

Superantigens have been suggested to act as powerful TCR V beta-specific inducers of T cell reactivity in autoimmune diseases. We have investigated the capacity of staphylococcal enterotoxins (SE) to prime autoreactive T cell responses in naive animals in the Lewis rat model of experimental autoimmune encephalomyelitis (EAE), where myelin basic protein (MBP)-specific CD4+ effector T cells express almost exclusively V beta 8.2 TCR elements. By taking advantage of the reactivity of V beta 8.2+ MBP-specific T cells to SEE but not to other SEs in vitro, we estimated the potential of different SEs (SEA, SEB, and SEE) to induce a primary T cell response to soluble MBP in vivo. Upon immunization of naive rats with soluble MBP alone or MBP and SEB (which is only a very weak superantigen for rat T cells), no MBP-responses could be retrieved. Similarly, when coimmunizing naive rats with MBP and V beta 8.2-activating SEE, no autoreactivity was inducible. By contrast, coimmunization of animals with soluble MBP and the superantigen SEA that is strongly activating various T cell subpopulations in Lewis rats but not V beta 8.2+ (i.e., potentially MBP reactive) T cells led to a significant primary MBP-specific T cell autoreactivity. These SEA-induced MBP-reactive T cells expressed V beta 8.2 TCRs at levels similar to those seen in autoreactive T cells conventionally induced by immunization with MBP administered in complete Freund's adjuvant (CFA) and could induce disease in a transfer model of EAE. Thus, our results are consistent with the notion that superantigens are able to induce primary T cell responses to soluble autoantigens by a non-V beta specific mechanism of bystander priming.

Immunization with thyroglobulin-specific cytotoxic T cell hybridoma induces anti-thyroglobulin antibodies: characteristics of monoclonal anti-thyroglobulin auto-antibody.

We have produced five monoclonal autoantibodies (mA-Abs) to thyroglobulin (Tg) and more precisely to one epitope located within the < 10-kDa pTg tryptic fragment suspension capable of inducing experimental autoimmune thyroiditis (EAT). They were selected from spleen cells from CBA/J mouse immunized with the syngeneic cytotoxic T cell hybridoma HTC2. HTC2 cells are specific for one Tg epitope located within the EAT inducer pTg tryptic fragments and are able to prevent EAT induction by pTg. The restricted specificity of the humoral response previously observed in vivo was further demonstrated and defined in vitro at the single cell level. Competitive studies for binding to pTg or to the < 10-kDa pTg tryptic fragments demonstrated that HTC2-induced anti-Tg mA-Abs recognized an epitope(s) located in the < 10-kDa pTg tryptic fragment (as did 3B8G9, one conventional anti-Tg mA-Ab we selected). We ruled out the possibility that HTC2-induced anti-Tg A-Abs belong to the group of the natural A-Abs due to the lack of recognition of actin, dsDNA, TNP-ovalbumin, tubulin, their isotypes (IgG1 or Ig2a), and their affinities (in the 10(-7) M order of magnitude). The results strengthen the hypothesis that T and B cells sharing the same specificity can express similar idiotopes on their respective receptors for antigen. They also demonstrate the existence of a regulatory idiotypic network that could explain the protection from EAT after injection of inactivated HTC2 cells or its anti-clonotypic mAb.

Ig VH gene family usage in spleen cells of CBA/J mice immunized with experimental autoimmune thyroiditis (EAT) inducer antigens.

Experimental autoimmune thyroiditis (EAT) is an autoimmune disorder of the thyroid gland induced in susceptible strains of mice by thyroglobulin (Tg). We recently showed that low Mr (< 10 kDa) Tg tryptic fragments and a 40 amino-acid peptide (F40D) from Tg could induce EAT as well as native Tg. Because it has been reported that autoantibodies (A-Abs) express VH families preferentially located in the D-proximal VH gene segment, we investigated whether A-Abs specific for one pathogenic peptide from Tg were also skewed towards D proximal VH gene segment. In that respect, we immunized CBA/J mice with EAT inducer antigens of decreasing sizes: Tg (660 M(r)), < 10 kDa Tg trypic fragments or F40D peptide (4.9 kDa M(r)) from Tg. The VH gene segments utilized by immune spleen cells were determined by hybridization to total spleen cell RNA previously deposited onto nylon membranes and densitometric scans. This study was conducted on days 7 and 9 after determination of the maximum amounts of mRNA coding for immunoglobulins and on day 28 when A-Ab levels are the highest. Results were compared to VH gene segment expression both in normal and adjuvant-injected mice. We found that immunization of CBA/J mice with EAT inducer antigens stimulate B cells the restriction of which, in terms of VH family usage, depends on the size of the immunizing antigen: the larger the antigen, the higher the numbers of VH families used. Moreover, we found that B cell stimulation consecutive to immunization with the peptidic antigen inducing EAT occurs in VH Q52 family, a VH encoded by D-proximal gene segment.

The effects of a monoclonal antibody to interferon-gamma on experimental autoimmune thyroiditis (EAT): prevention of disease and decrease of EAT-specific T cells.

CBA/J mice immunized with thyroglobulin (Tg) develop an experimental autoimmune thyroiditis (EAT) with lymphocytic infiltration of the thyroid glands, autoantibodies to Tg and occurrence of EAT-specific T cells. When these mice were treated for 4 weeks after immunization with 1 mg/week of a monoclonal antibody (mAb) that neutralizes the activity of interferon-gamma (IFN) a beneficial effect on the onset of EAT was observed. Characteristic features of EAT were significantly reduced, including the lymphocytic infiltrations of the thyroid glands and the serum levels of autoantibodies to Tg. Moreover, in lymphoid organs, mAb to IFN-gamma significantly reduced the percentages of Tg-specific CD8+ cells, labeled by the anti-clonotypic mAb AG7. These Tg-specific T cells seem responsible for thyroid damages and disease development, since EAT was simultaneously abrogated. These results show that IFN-gamma plays an essential role in the pathophysiology of EAT and suggest the possibility to treat autoimmune thyroid diseases with mAb to IFN-gamma or drugs able to antagonize the production and/or the action of this cytokine.