The requirement of new kind of frequency for technological development are increasing. Among all radiofrequencies, millimeter waves that are used in the 5G/6G wireless networks, are a range of frequencies between 30 GHz and 300 GHz. Their peculiar physical properties make them of interest for telecommunication considering their short latency (<1 ms), high speed data transfer (multi-Gbps), and reduced size of devices (Bautista et al., 2019). Even if in the 5G deployment these frequencies are not yet widely deployed, the future technological development will surely use these peculiar frequencies. Question on their potential adverse associated risks are nevertheless still pending (Simkó and Mattsson, 2019). Up to now, the main link with pathology comes from epidemiologic studies that have relied the occurrence of acoustic neurinome and some brain cancer with large use of mobile phone, however these results concern only lower part of radiofrequencies. After intense debate on results (Hardell et al., 2013; INTERPHONE Study Group, 2010, INTERPHONE Study Group, 2011), IARC classified electromagnetic waves as possibly carcinogenic (Risk 2B) (IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2013). However, no convergent information and no mechanisms are really available on MMW and cancer promotion particularly.

Among all possible mechanisms potentially disturbed by the MMW, we have intensively explored the gene expression. Due to their specific physical properties MMW have low penetration depth (< 1 mm), which means that the main potential target of these frequencies are keratinocyte that are the main constituent of the first layer of skin (Orlacchio et al., 2019; Zhadobov et al., 2011). In MMW's studies several experiments of microarrays were performed to evaluate the effect of these frequencies (30–300 GHZ) on gene expression (Habauzit et al., 2014a; Le Quément et al., 2012; Millenbaugh et al., 2008; Soubere Mahamoud et al., 2016). A pioneering study was performed to analyse the heat-associated effect of high power MMW exposure on rat skin and compare to conventional heat stroke. This experiment evidenced that the main effect was the heat effect associated to the MMW (Millenbaugh et al., 2008). Some other experiments have tried to discriminate between thermal and non-thermal effects. First of all, these approaches did not evidenced specific action of these frequencies on cellular stress derivate functions (Le Quément et al., 2014). Microarray analysis of primary cultures of keratinocyte exposed to non-thermal 60 GHz millimeter wave (incident power density (IPD) of 1.8 mW/cm2) has enabled to identify very few differentially expressed genes (Le Quément et al., 2012). No effects were observed under non-thermal exposure even with higher IPD (Habauzit et al., 2014b; Soubere Mahamoud et al., 2016). If MMW does not have any direct effect, the question of co-exposure was to be explored. At the former ICNIRP (Ahlbom et al., 1998) IPD limit (20 mW/cm2), our group have previously showed that some slight effects could be evidenced in co-exposure scenario (Habauzit et al., 2014a; Le Quément et al., 2012; Soubere Mahamoud et al., 2016). Indeed, when MMW exposure is associated with another stressor agent, such as a heat shock or as a metabolic stress, around 1% of the all genes modified by the stressor agent are targeted by MMW. Nevertheless, when the co-stressor agent is removed, effects were no more observed to the short-term in in vitro models (Habauzit et al., 2014a; Soubere Mahamoud et al., 2016), but also to the long-term exposure in animal models (Habauzit et al., 2020). Among the 1% genes modified when MMW are applied with the co-stressor agent heat shock, 3 genes (ADAMTS6, IL7R and Nog) were evidenced as reproducibly differentially expressed in a primary culture of keratinocytes (Habauzit et al., 2014b). It is important to note that these results were not confirmed by using three other different primary cultures and one HaCaT cell line (Martin et al., 2020). Taken together, these results showed that the transcriptional response to MMW was of very low amplitude and could be variable from one cellular model to another.

In order to get deeper insight into transcriptional changes induced by millimeter-wave and to identify potential biomarkers of exposure, we performed a gene expression analysis of three distinct models of human keratinocytes exposed to 60-GHz millimeter-waves by using Bulk RNA barcoding sequencing (BRB-Seq). These experiments focus on the molecular aspect of the cellular response to MMW, so we have chosen in vitro models which are perfectly suit to this purpose and which have the advantage to fit the 3R principle and replace animal model (Russell and Burch, 1959).