Primary Normal Human Epidermal Keratinocytes (NHEK) (Promocell C-12001, Heidelberg, Germany; these cells are tested by the supplier for cell morphology, adherence rate, and cell viability. Immunohistochemical tests for specific cytokeratins are carried out for each lot. Growth performance is tested through multiple passages up to 15 population doublings (PD) under culture conditions without antibiotics or antimycotics. Finally, cells are tested for the absence of HIV-1, HIV-2, HBV, HCV, HTLV-1, HTLV-2, fungi, bacteria, and mycoplasma.) were cultured at 37 °C in keratinocyte basal medium (Promocell C-20216, Heidelberg, Germany) supplemented with 0.06 mM Ca${}^{2+}$ (Promocell C-34005, Heidelberg, Germany) and a keratinocyte-specific supplement mix (Promocell C-39016, Heidelberg, Germany), under a humidified atmosphere containing 5% CO${}_2$, so as to reach 40–60% confluency at the time of irradiation. The culture medium was renewed just before irradiation, and the cells were exposed to 5G bandwidth radiation using a N5172B EXG X-Series RF Vector Signal Generator (Keysight Technologies, Santa Rosa, CA, USA), placed 9.5 cm above the cell culture, at 6 GHz, 30 dBm, 20 mW/cm${}^2$ (based on the recommended exposure limits for currently widespread 5G equipment types in guidelines quoted above: the guidelines on electromagnetic field exposure by ICNIRP [4] specify a maximum power density of 10 mw/cm${}^2$ for occupational exposure, averaged over an area of 4 cm${}^2$; when averaging for smaller areas on the order of 1 cm${}^2$, such as for small diameter/focal beam exposure as is the case for the irradiated area in this study, the maximum power density is 20 mw/cm${}^2$), for 1 h, 2 h, 4 h, 6 h or 24 h. A control leg was left untreated except for parallel renewal of the culture medium. The cells were then fixed with methanol for 5 minutes at –20 °C, and frozen.

The cells were blocked with PBS (Sigma Aldrich P3813-10PAK, St Louis, MO, USA) containing 0.001% Tween-20 (Sigma Aldrich P1379, St Louis, MO, USA), 1% BSA (Sigma Aldrich A9085, St Louis, MO, USA) and 10% goat serum (Sigma Aldrich, G9023, St. Louis, MO, USA) during 1 h and then incubated overnight at 4 °C with: anti-TIMP1 mouse monoclonal antibody (Invitrogen MA1-773, Waltham, MA, USA, diluted 1/2000 in blocking solution); anti-ANGPTL4 rabbit monoclonal antibody (Invitrogen 701155, Waltham, MA, USA; diluted to 1/500 in blocking solution); or anti-S100A9 rabbit polyclonal antibody (Invitrogen PA5-79949, Waltham, MA, USA; diluted to 1/500 in blocking solution). Secondary antibodies (AlexaFluor® AF488 and AF594, Thermo Fisher, Waltham, MA, USA) were used for immunofluorescence staining, which was assessed by microscopical observation. Manual scaling was carried out per field, taking into account all cells in the field. IL1-$\alpha$ was quantified by enzyme-linked immunosorbent assay (ELISA) (human IL1-$\alpha$ kit, Cayman Chemical 583301, Ann Arbor, MI, USA). Four independent biological replicates (4 sampled wells) were included for each set of conditions.

Data analysis was performed using GraphPad Prism 9 (GraphPad Software, Inc., Boston, MA, USA). Statistical tests were selected on the basis of normality and lognormality tests: a parametric test was applied when the normality was positive, and a non-parametric test was used when the normality was negative. For this study, an unpaired Student t-test was applied. Data in the figures is shown as mean +/- Standard Deviation, with p-values appearing on brackets.

When primary Normal Human Epidermal Keratinocytes (NHEK) were exposed to 5G bandwidth radiation, clear effects of 5G radiation on the culture of human epidermal keratinocytes were observable after a remarkably short 1-h exposure time, in several indicators of keratinocyte function (Figs. 3,4), including: a strong reduction in collagenase inhibitor TIMP1; a strong increase in wound healing and epidermal differentiation facilitator ANGPLT4; a marked increase in S100A9, involved in immune recruitment during injury; and, an increase in key inflammatory cytokine IL1-$\alpha$ (whose level was tripled vs. non-irradiated control, albeit with borderline statistical significance).

Fig. 3.

Representative images of immunofluorescence staining revealing effects of 5G exposure on keratinocytes. Typical images of staining for TIMP metallopeptidase inhibitor 1 (TIMP1), Angiopoietin-like protein 4 (ANGPLT4), and S100 calcium-binding protein A9 (S100A9), in a culture of normal human dermal keratinocytes exposed to 5G radiation for 1–24 h. For each timepoint, comparing unirradiated control (Left) with 5G irradiation for the time indicated (Right). Scale bar = 50 μm. Trend evaluation (visual) symbol key: Decrease: $\downarrow$$\downarrow$$\downarrow$$\downarrow$ : clear; $\downarrow$$\downarrow$$\downarrow$: fairly clear; $\downarrow$$\downarrow$: moderate; Increase: $\uparrow$$\uparrow$$\uparrow$$\uparrow$: clear; $\uparrow$$\uparrow$$\uparrow$: fairly clear; $\uparrow$$\uparrow$: moderate; $\uparrow$: slight. $\leftrightarrow$: no change.

Fig. 4.

Interleukin 1-alpha release in a culture of normal human dermal keratinocytes exposed to 5G radiation. Enzyme-linked immunosorbent assay (ELISA) measurement of IL1-$\alpha$ after 1–24 h 5G radiation exposure (black: unirradiated control; gray: after irradiation for the time indicated). Data shown as mean +/- SD; brackets indicate p-values (unpaired Student t-test). Four independent biological replicates (4 sampled wells) were included for each set of conditions.

Interestingly, the effects on all four markers became less marked with longer exposure times (although, generally, some effect was still detectable). The TIMP1 reduction became less intense from 4 h exposure, though it remained noticeable. The increase in ANGPTL4 weakened steadily with increasing exposure time, until at 24 h there was only a slight increase vs. the unexposed control. The increase in S100A9 similarly weakened with increased exposure duration, until there was no noticeable change at 6 h exposure vs. unexposed control, and only a slight change at 24 h exposure. The increase in IL1-$\alpha$ continued to be at least a detectable trend through 6 h exposure, although with lower statistical significance vs. the unexposed control at most time points; however, a steady increase in baseline IL1-$\alpha$ levels was also observed in the unexposed control leg, reaching levels high enough that by 24 h the 5G exposure had essentially no effect (the level in the exposed leg at 24 h was actually somewhat lower vs. the unexposed control, but the difference was not statistically significant). It should be noted that, as culture media were left in continuous contact with the cell culture for the duration of the irradiation, the increase observed in the IL1-$\alpha$ levels likely results at least in part from accumulation into the medium over this period.

Using the same exposure model, we exposed epidermal keratinocyte cultures to 5G radiation following 24 h pre-treatment with one of several botanical extracts, including: a Cistus incanus extract, a Trifolium pratense extract, and an Alteromonas ferment extract (in addition to an untreated control leg) (Fig. 5). Here again, after 1h of 5G irradiation, the data show a clear increase in IL1-$\alpha$ release in our keratinocyte culture with irradiation (+68% vs. unirradiated control, from 52.0 to 87.5 pg/mL with p = 0.00089). Although the intensity of this increase is less than that observed in the initial study described above, this result still shows an influence of the 5G radiation on the cell culture (and, its statistical significance is actually clearer). On the other hand, we also observe clear protective effects for the three botanical extracts. Two of the three extracts reduce IL1-$\alpha$ release to levels below that of the unirradiated baseline, most strongly so in the case of the Cistus incanus extract, which almost totally extinguishes IL1-$\alpha$ release (to 6.3 pg/mL, or –93% vs. irradiated, untreated control, with p = 0.00000002; and –88% vs. unirradiated baseline, p = 0.00021) (Fig. 5). The Trifolium pratense extract reduced IL1-$\alpha$ to 31.7 pg/mL (–64% vs. irradiated control, with p = 0.00001; though this level is still somewhat below that of the unirradiated control, the difference does not reach full statistical significance, with p = 0.07858). Finally, the Alteromonas ferment extract reduced IL1-$\alpha$ to 48.7 pg/mL (–44% vs. irradiated control, with p = 00071), returning IL1-$\alpha$ levels to the vicinity of the unirradiated baseline (statistically identical levels, with p = 0.72338).

Fig. 5.

Protective effect of three botanical extracts. IL1-$\alpha$ measurement by ELISA on a culture of normal human dermal keratinocytes exposed to 5G radiation for 1h, as compared to untreated, unirradiated and untreated, irradiated controls. Data shown as mean $\pm$ SD; brackets indicate p-values (one-way ANOVA followed by Holm-Šídák’s multiple comparisons test). Six independent biological replicates (6 sampled wells) were included for each set of conditions.

Although the results presented above (especially the results of the immunofluorescence stainings) are mostly qualitative indications, it appears that, after a short exposure time to 5G radiation (1 h), a (negative) influence of this radiation on the exposed epidermal keratinocyte culture can be observed. This includes indications of elevated inflammation status (IL1-$\alpha$), and the activation of homeostasis and damage/injury response mechanisms. A reduction in collagenase inhibitor TIMP1, for one, may lead to increased collagenase activity in the exposed tissue, leading to extracellular matrix protein degradation and/or its eventual replacement with freshly synthesized material. Increased ANGPLT4, given its role in wound healing and epidermal differentiation, may be symptomatic of the activation of homeostasis mechanisms aiming to repair injury caused by the 5G radiation exposure. The observed increase in S100A9, involved in immune recruitment during injury, would seem to support the same indication. These results are in accordance with the implications of the literature discussed above.

The effects observed with longer, continuous exposure times are somewhat more ambiguous. Indeed, with continued exposure the effects described above seemed to weaken considerably—although in most cases they did not disappear entirely. One could suggest as a possible explanation for these effects the idea that, beyond a relatively short initial shock, represented by the data collected after 1 h 5G radiation exposure, we in fact observing injury response adaptive mechanisms coming into action to mitigate the damage resulting from this exposure, and seek to restore homeostasis. It cannot, of course, be ruled out that other effects may also be in play; the observed increase in baseline IL1-$\alpha$ in the control cultures, for example, may simply be due to accumulation in the medium (left continuously in contact with the cell culture over the duration of exposure), or it may also hint at aging effects in the culture itself—in which case the weakening in the effects observed with 5G irradiation may actually reflect a reduced response and repair ability on the part of the cell culture.

In follow-up studies, it could be very valuable to confirm these directions by increasing exposure times beyond the 24-h mark, with the aim of elucidating whether the cell culture is indeed capable of mitigating the damage caused beyond an initial short-term shock. Further, should this turn out to be the case, a study examining the effect of repeated short-term exposures, seeking to determine whether this type of exposure results in accumulated damage (or, conversely, whether homeostasis mechanisms come into play here also), could be of very high relevance to the continued investigation of the real-world effects of chronic exposure of the population to 5G emissions. Within the bounds of possibility, these studies should use quantitative measures of the effects observed. Finally, additional irradiation parameters such as frequency, intensity, and related others, should be investigated so as to expand the relevance and applicability of this work.

It should be noted that mechanisms such as inflammation and protease activation, both of which our results to be possibly activated by 5G radiation exposure, are widely recognized to lead to premature skin aging, notably via extracellular matrix (ECM) degradation [48]. It thus seems possible that extended and/or chronic exposure to 5G radiation may lead to the premature appearance or worsening of signs of skin aging such as wrinkles, loss of elasticity, pigmentation irregularities, and more.

As shown above, when we pre-treated our keratinocyte culture with each one of the three botanical extracts presented here, we observed a significantly reduced increase in IL1-$\alpha$ following 1 h exposure to 5G radiation. While keeping in mind the preliminary nature of the data (assessment based on the single IL1-$\alpha$ marker, imperfect reproducibility between studies), and while the specificity of this effect to mitigation of 5G-induced inflammation (vs. general anti-inflammation effects) is as yet unclear, it still appears possible that these results point at least in part to a mitigation of the inflammatory processes triggered by this radiation exposure, and thus indicate that these extracts may have significant potential to ameliorate the effects of 5G radiation on epidermal keratinocytes.

The effects of EM/radiofrequency radiation are largely recognized as proceeding at least in significant part via the increase of oxidative stress (possibly through the generation of Reactive Oxidative Species (ROS)) and the triggering of inflammatory states in living tissue [8, 49]. In our own work, as described above, 5G irradiation seems to trigger inflammatory processes in the keratinocyte culture (as evidenced by increased IL1-$\alpha$). In this light, it could be expected that antioxidant and anti-inflammatory substances may offer some protection against the negative effects of 5G exposure. At present we are only aware of a single directly comparable study in the scientific literature, by Kappler et al. [50], demonstrating the protective effects of botanical extracts (in this case, an unspecified extract containing tannins, complex polyphenolic compounds with recognized anti- oxidative properties) against the effects of 5G/radiofrequency irradiation on skin. It is worth noting that the protective effects described by Kappler et al. [50] seem to rest mainly on anti-inflammatory and/or anti-oxidant effects (evaluated in their work via effects on IL1-$\beta$ and ROS production, respectively).

The protective effects observed here may thus be considered quite logical given some of the extracts’ known composition and biological properties. The Cistus incanus extract, which showed the strongest protective effect in our work, has been shown to possess strong anti-inflammatory effects, mediated at least in part via NF-kB [38]. Cistus incanus extracts have also been shown to contain polyphenolic compounds (including myricitrin) and possess antioxidant properties [39]. The latter have also been shown in vivo with aqueous extracts of the aerial parts of this plant, possibly attributable to their high polyphenol content [51]. It seems likely that antioxidant properties also participate in the observed protective effect. The same logic applies to the Trifolium pratense flower extract, which possesses documented antioxidant and anti-inflammatory properties, due at least in part to the presence of flavonoids such as biochanin A [40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53]. For its part, the Alteromonas ferment extract has been shown to enhance keratinocyte renewal in-vitro, and while the exact mechanisms in play have not been completely elucidated, it is possible that they are linked, directly or indirectly, to the effect we observe here [47]. Some Alteromonas ferment extracts have also been shown to possess antioxidant and anti-inflammatory properties [52, 53]. Finally, it is notable that $\beta$-glucan polysaccharides have been reported as showing protective effects against oxidative injury caused by EM radiation at 2.45 GHz (bordering on the energy ranges relevant to 5G) in rat skin tissue [54].

The specificity of the effect of the extracts tested here is not yet clear, and may vary from extract to extract. In the case of the Cistus incanus extract, which reduced IL1-$\alpha$ production to levels well below those of the unirradiated control leg, it is likely that the effect extends to more than just countering the effects of the irradiation itself, and at least part of the efficacy observed is linked to the extract’s known anti-inflammatory effects (discussed above). Treatment with the other two extracts returned IL1-$\alpha$ levels to the vicinity of that observed in the unirradiated control leg—leaving open the possibility that in those cases the effect is more specific to IL1-$\alpha$ production induced by the irradiation (though of course this does not definitely prove this point on its own, and we could also be observing a weaker, nonspecific effect). It would likely be valuable to include treated, un-irradiated cultures in future studies, in an attempt to clarify this point.

In follow-up studies, it would likely be advisable to improve the repeatability of this model, seeking to overcome some of the variability observed between our two studies. The consistency of the primary cells used in the cultures could be a relevant area of focus, for instance—or conversely, it could be interesting to explore the effects of this irradiation on cells from different provenances (e.g., covering a range of donor ages, among other individual characteristics which could influence the results), as well as on additional skin cell types. It may also be valuable to explore the effect of these extracts, and possibly others, on other processes and biological markers which have been shown, here and elsewhere, to be impacted by exposure to 5G radiation. Also, as noted above, it would be interesting to evaluate the protective effect of these and other botanical extracts on a keratinocyte culture exposed to 5G radiation for longer durations (up to, and beyond, 24 h), as well as for short but repeated durations. Additional irradiation parameters such as frequency and intensity should be varied in the context of treatment with botanical extracts, in order to demonstrate protective effects over a wider range of real-life conditions. As suggested by Kostoff et al. [17], it may be useful to examine signal pulsation and modulation, to more closely mimic actual exposure.