We first analyzed the expression of EVER
genes in a panel of freshly collected murine tissues, and in murine and human lymphocyte subsets. Using both classical RT-PCR (data not shown) and quantitative RT-PCR (qRT-PCR), we observed that EVER1
were clearly expressed in spleen and thymus (), as well as in purified murine () and human (Fig. S1A, B
) lymphocyte populations. EVER1
, and to a lesser extent EVER2,
were also expressed in brain, heart, kidney, liver and skin (). The identity of the amplified bands was confirmed by direct sequencing of the RT-PCR products or by cloning and sequencing of the qRT-PCR products (data not shown). It is noteworthy that the highest expression of EVER1
genes was not found in the skin but rather in lymphoid organs (). The expression of EVER1
did not differ significantly between B and T cells (), or between CD8+ and CD4+ T cells (, Fig. S1B
). Moreover, the amount of EVER1
transcripts, as assessed by qRT-PCR, was similar in CD8+ and CD4+ T cell subsets. The mean EVER2/EVER1 expression ratio was 0.96 (n
4) and 0.77 (n
6) in CD8+ and CD4+ T cells, respectively. Interestingly, a strikingly different pattern of EVER
gene expression was observed in the skin where EVER1
was preferentially expressed whereas EVER2
was barely detectable (mean EVER2/EVER1 expression ratio 0.07; n
EVER1 and EVER2 are expressed in lymphocytes.
EVER1- or EVER2-deficiency leads to a clinically identical EV phenotype in humans 
. This observation has two important implications. First, it suggests that EVER1 and EVER2 have non-redundant functions. Indeed, the absence of either one of the EVER proteins leads to the dysfunction of the whole EVER complex resulting in zinc imbalance 
. Second, the cell type(s)/tissue(s) where both genes are expressed, and not those with a selective expression of either EVER1 or EVER2, are most probably essential in EV pathogenesis. In this context, it is tempting to speculate that the lymphoid tissue, which is characterized by a high expression of both EVER1
genes, rather than the skin, where the constitutive expression of EVER2 is extremely low (), could be most relevant to EV pathogenesis.
We next verified by western blot that EVER proteins were expressed in spleen and freshly purified CD4+ or CD8+ T cells. A single band of approximately 100 kDa was detected using a commercially available antibody against murine EVER1 (). To ensure that the antibody used was specific, and that it could discriminate between the two related EVER1 and EVER2 proteins, we employed two complementary strategies. First, the expression of EVER1 was confirmed with a second antibody recognizing a distinct epitope of EVER1 (Osenses; OSR00223W). A single band of approximately 100 kDa was detected with this antibody in spleen and in CD8+ T cell lysates (data not shown). Second, we tested whether these two anti-EVER1 antibodies specifically recognized EVER1. To this end, 293 T cells were transfected with plasmids encoding either EVER1-FLAG or EVER2-FLAG fusion proteins. The fusion proteins were subsequently precipitated using the anti-FLAG antibody and western blot analyses were performed. We were able to detect both EVER1-FLAG and EVER2-FLAG fusion proteins with the anti-FLAG antibody. However, the two anti-EVER1 antibodies recognized a single band (corresponding to MW ≈ 85 kDa) in cells expressing EVER1-FLAG, but not in cells expressing EVER2-FLAG (), demonstrating their specificity. We attempted to verify by the same approach the expression of the EVER2 protein in T cells but the anti-EVER2 antibody tested (Abcam, ab70002) did not work in our hands (data not shown).
Since EVER proteins form a complex with ZnT-1 
, a well-characterized zinc transporter whose expression is enhanced by Zn2+
ions through the MTF-1 transcription factor 
, we evaluated whether transcription of EVER1
in T cells could also be regulated by zinc. Whereas expression of two zinc-inducible genes (MT-2
) was clearly increased 2 or 6 hours after exposition to a non-toxic concentration of zinc (100 µM), expression of EVER
genes remained stable following 6- and even 24-hour exposition to zinc ().
EVER1 and EVER2 expression is strongly down-regulated in T cells following TCR-mediated activation.
We next tested the impact of in vitro
T-cell activation and differentiation on EVER expression. EVER1
were constitutively expressed in purified naive CD8+ () and CD4+ () T cells. Interestingly, activation of these T cells using anti-CD3 and anti-CD28 antibodies led to a rapid and profound decrease in EVER1
mRNA levels. The drop in EVER1
expression was detected as early as 3 hours after activation (). This decreased expression of EVER
genes was maintained for at least 3 days (). A similar EVER1
mRNA down-regulation was found in human T cells following TCR/CD28–dependent activation (Fig. S2A
, B). The TCR and CD28 exerted a synergistic effect on the down-modulation of EVER
genes, as triggering of either TCR or CD28 alone had little or no effect (). We then validated these data in an antigen-driven T-cell activation system, using naive CD8+ T cells specific for an influenza
virus hemagglutinin peptide (HA512–520
; IYSTVASSL), originating from CL4-TCR transgenic mice 
. Activation of purified TCR-transgenic CD8+ T cells with syngeneic antigen-presenting cells loaded with the cognate HA512–520
peptide was also associated with a clear and long-lasting decrease in the expression of both EVER
The exact mechanism of the down-regulation of EVER expression in response to T-cell activation remains unknown. Interestingly, the reduction in the amount of EVER transcripts is rapid, suggesting that EVER mRNA might be relatively unstable. Probably, due to its rapid turnover, the suppression of the transcription of the corresponding gene would result in such a rapid reduction in the mRNA quantity. Regardless of the exact molecular basis of this observation, the role of EVER1 and EVER2 in T cells and the functional significance of the modulation of their expression constitute an important unanswered question.
Cellular zinc homeostasis is tightly controlled by an elaborate system composed of different zinc sensors (like MTF-1), transporters (ZnT and ZIP families), and buffering proteins such as metallothioneins 
. It has been suggested that ZnT-1 and the EVER/ZnT-1 complex, like the other members of the ZnT family, could be involved in the transport of Zn2+
ions out of the cytoplasm, in order to maintain the low Zn2+
cytoplasmic level in spite of the large gradient of this metal across the plasma membrane 
. Based on our data, we propose that the EVER complex is involved in the regulation of zinc balance in T cells, and that down-modulation of the EVER
expression during T-cell activation could impact zinc homeostasis. In the course of CD8+ and CD4+ T-cell activation, we observed a decreased expression of EVER1
2 as well as of ZnT-1
(data not shown), with concomitant up-regulation of ZIP10
() and ZIP6
, B) expression. ZIP10 and ZIP6 are plasma membrane transporters responsible for Zn2+
influx into the cell, and in particular in T cells 
. This coordinated reorganization of zinc transporter expression is compatible with a program of zinc accumulation in the cytosol of activated T cells. Indeed, we confirmed that the total amount of free Zn2+
ions per cell was significantly increased 24 h after TCR- and CD28-dependent activation (), and maintained at this level for at least 72 h (data not shown). The mechanism and functional significance of such an accumulation remain uncertain given the very limited knowledge on the physiological role of free Zn2+
ions in T cells. However, this zinc accumulation could be related to the massive changes in lymphocyte volume (about 10 times, based on the changes of the forward side scatter in flow cytometry) and metabolic activity during blast formation. Thus, at least two non-mutually exclusive hypotheses could be formulated. First, zinc accumulation could be a response to the increased needs of this metal for the newly synthesized zinc-binding proteins. Second, it might constitute a compensatory mechanism in order to maintain a constant free Zn2+
ion concentration during the massive enlargement of the stimulated T cells.
Recent findings, showing that Zn2+
ions play an important role in signal transduction and can even serve as a classical second messenger 
, highlight the need for tight control of the intracellular free zinc pool. Indeed, we observed that the aforementioned modulation in zinc transporter expression in activated CD4+ and CD8+ T cells is accompanied by massive transcription of MT-2
(), encoding for the major zinc-buffering metallothionein 
. Finally, despite the increase in the total zinc content per cell following TCR/CD28 triggering (), the intracellular concentration of free zinc remained remarkably stable (). In this context, it is tempting to speculate that the lack of either EVER proteins could disrupt the EVER/ZnT-1 complex and consequently impose zinc imbalance in T cells as previously demonstrated in keratinocytes 
. To assess the impact of the EVER-deficiency on zinc homeostasis in lymphocytes, we first measured the concentrations of free Zn2+
in EBV-transformed B cells from an EV patient (EVER1
Del275A) and from a healthy relative using FluoZin-3, a zinc ion-specific fluorescent probe, and a flow cytometric assay. In five independent measurements, the zinc concentration was higher in EVER-deficient than in control cell lines (). These results suggest that EVERs could be involved in the regulation of cellular zinc homeostasis.
Increased cellular zinc concentration in lymphocytes from EVER2-deficient patients.
To then address the contribution of EVER proteins for zinc balance in primary lymphocytes, untransformed T cell lines were generated from PBMC of two healthy individuals and two patients, one with full-blown EV with homozygous EVER2
T150fdX3 mutation and one heterozygous for the EVER2
T150fdX3 mutation. From a clinical point of view, this heterozygous patient clearly presented transiently EV-like eruptions and was HPV3 and HPV5 positive in the lesions (see Materials and Methods
). Interestingly, high HPV5 viral load was recently described in non-symptomatic family members of an EV patient with heterozygous EVER2 13-nucleotide deletion 
. These data suggest that heterozygous EVER2 mutation, and the resulting decrease in protein expression, favors HPV5 replication. This may explain the transient EV-like clinical presentation of this heterozygous patient. Using the FluoZin-3 based cytometric assay, two independent measurements of free zinc in CD4+ T cells were performed, side-by-side for all the lines. The mean free Zn2+
concentration in EV-derived CD4+ T cells was higher than in CD4+ T cells from healthy donors (). This higher concentration of free zinc in EVER-deficient CD4+ T cells was maintained following overnight incubation of the cells with a nontoxic zinc concentration (100 µM) (). Although derived from a small number of patients, these data suggest that EVER proteins might be involved in the control of zinc homeostasis in lymphocytes. At least two different types of intracellular zinc signaling occur in T cells. One happens extremely rapidly after TCR triggering as a result of zinc transporter activity. The other depends on transcriptional changes in expression of zinc importers and exporters that increase the cellular zinc content in a delayed and durable manner. In this study, we have monitored the second type of zinc signal but it would be most interesting to assess whether there is any impact of EVER-deficiency on the early and focal T cell zinc fluxes upon TCR activation. Further studies on a larger cohort of EV patients and/or on EVER
knock-out mice should help settle this issue.
Since the concentration of free Zn2+
is tightly controlled during lymphocyte activation () and EVER-deficiency might increase the concentration of unbound zinc ions (), one burning question relates to the biological consequence of the zinc imbalance in T cells. During T-cell activation zinc acts at different levels. Zn2+
participates in the interaction between Lck and CD4 and CD8 co-receptors, stabilizes Lck dimerization, and is likely involved in the activation of protein kinase C 
. It has been shown that Zn2+
is released from lysosomes upon T-cell activation 
and that it participates to IL-2R-mediated signaling through the regulation of ERK activity 
. Moreover, zinc influx from extracellular sources is induced within minutes of TCR triggering in primary T cells, with compartmentalized high zinc concentration under the zone of the T cell/APC contact. This local increase in cytoplasmic zinc influences proximal TCR signaling resulting in enhanced proliferative responses to suboptimal stimuli 
. On the other hand, Tanaka and colleagues have reported that excess of zinc suppresses mitogen-activated IL-2 production in lymphocytes 
. Collectively, the current data indicate that Zn2+
ions play a clear role in signal transduction and T-cell activation, and any deviation in free zinc concentration is likely to impair these processes 
. This, in turn, highlights the importance of the system that protects the cells from the high concentration of Zn2+
in the extra-cellular milieu. Interestingly, when we disrupted this zinc homeostasis by incubating T cells with non-toxic concentrations of a zinc ionophore (pyrithione), the CD3/CD28-dependent proliferation was inhibited, as demonstrated by thymidine incorporation () and CFSE dilution assay (data not shown). Naive T cells appear more sensitive to zinc than activated T cells (), even though this inhibition is only temporary lasting cells enter into mitosis 24 hours after exposure to pyrithione (data not shown). The reason of the higher sensitivity to zinc of naive T cells remains to be elucidated. It is likely that activated T cells are more resistant to excess of zinc, since they have already reorganized the system controlling zinc homeostasis, and notably exhibit increased expression of zinc buffering proteins (metallothioneins). Taking into account the known properties of ZnT-1 as a zinc transporter, and our observations of increased concentration of Zn2+
in EVER-deficient lymphocytes (), it seems probable that the ZnT-1/EVER complex participates in the maintenance of low levels of free zinc in lymphocytes. It is tempting to speculate that EVER-deficiency, by shifting zinc homeostasis, could have an impact on T-cell activation, especially in naive T cells, in which EVER expression is the highest. Certainly, the clinical manifestations of EVER-deficiency suggest that such an impact does not interfere with crucial steps of T-cell activation, as EV patients respond normally to pathogens other than HPV. Indeed, we did not observe any striking differences in the rate of culture expansion during preparation of the primary T cell lines from EV patients and healthy controls (data not shown).
In conclusion, we have demonstrated that EVER
genes are expressed at high levels in T and B cells. Their expression in T cells is sharply decreased upon TCR/CD28 stimulation, as a part of the global activation-dependent reorganization of the molecular machinery controlling cellular zinc homeostasis. Indeed, the elevated intracellular amount of free Zn2+
ions is associated with decreased expression of ZnT-1
and increased expression of Zip10
in activated T cells. The preliminary data from EV patient-derived cell lines suggest that EVER might be involved in maintaining a constant low level of free zinc in lymphocytes. Altogether, these results support our previously formulated hypothesis 
that the natural anti-HPV barrier, of which EVER proteins constitute an essential part, might comprise at least two arms: keratinocytes and lymphocytes. Although still speculative, we suggest that EVER-deficiency in T cells may contribute to susceptibility to HPV infection, possibly in combination with EVER-deficiency in keratinocytes. However, the exact mechanism of the impact of EVER-deficiency on T cell function as well as the reason for the extreme selectivity of this barrier towards HPV remains to be elucidated.