Lymphocytes represent about 20-40% of white blood cells (i.e., 70-90% of PBMCs) and are responsible for the antigen-specific (B and T cells) and innate (NK cells) characteristics of immune response. In Additional file
1 we summarized the studies conducted so far evaluating the impact of f-CNTs, graphenes, and carbon nanohorns in these cells[
13‐
39]. One of the first work on CNTs, functionalized by 1,3-dipolar cycloaddition reaction and by oxidation/amidation treatment, was conducted by Dumortier
et al. in 2006[
16]. The study demonstrated the non toxicity of well functionalized CNTs on primary mouse T and B cells. A year later, the group of Dai, while studying a specific siRNA delivery system against CXCR4 (a critical receptor used by HIV to infect target cells), revealed the non toxicity of pegylated CNTs on human T cell lines and primary T cells[
17]. Delogu
et al.[
21] focusing on the use of CNTs for drug delivery to knock-down PTPN22, a gene related to type 1 diabetes[
40], confirmed the non toxicity of PEGylated CNTs on T cells. Similar results were also found by Cato
et al.[
18]. A systematic genotoxic study[
23] reported the effect of SWCNTs, MWCNTs, and amidated SWCNTs on cultured human lymphocytes. This investigation showed that, in contrast to non functionalized CNTs, the amidation of SWCNTs did not disturb the T cell proliferation, therefore preserving T cells from the observed genotoxicity. The studies discussed above highlights that different type of functionalizations such as 1,3-dipolar cycloaddition, adsorption with phospholipids and polyethylene glycol, and amidation are fundamental for preserving viability of treated cells. Furthermore, other than addressing biocompatibility issues, Fadel
et al. suggested that f-CNTs can be potentially used in the context of immunotherapy[
19]. They proved that anti-CD3 adsorbed onto SWNTs bundles stimulate T cell lines more effectively than equivalent concentrations of soluble anti-CD3. However, the stimuli was more efficient with functionalized versus non functionalized nanotubes[
25]. The groups of Bianco and Gennaro demonstrated that two different conjugates f-CNTs and amphotericin B can achieve an antifungal activity comparable to, or better than, amphotericin alone in T cell lines (jurkat cells)[
27]. A large number of studies took advantage of the easiness of culturing cell lines such as jurkat cells[
13‐
15,
18,
21,
26‐
28,
39], a line established from the peripheral blood of a 14 year old child affected by T cell leukemia. The clear advantages in using this cell model (highly reproducible experiments, low cost and favorable culture conditions) are somehow in contrast with its neoplastic characteristics that may not always reflect the
in vivo physiological condition of T cell behavior. To avoid this concern, in a recent study, we explored the impact of different functionalized MWCNTs on PBMCs from healthy donors[
31]. Two types of multi-walled carbon nanotubes (of low, 9.5 nm, and big, 20–30 nm, diameter) were initially oxidized and further modified by 1,3-dipolar cycloaddition reaction to obtain ammonium-functionalized nanotubes. After having excluded
in vitro toxicity and confirmed their dose-depending uptake, we observed that f-CNTs with smaller diameter were internalized more efficiently. Moreover, by analyzing CD4+ and CD8+ T cells, B cells, NK cells, and monocytes, we surprisingly noticed that f-CNTs have a cell specific effect. In fact, none of them induced T cell expansion or activation (evaluated by CD25, and CD69 markers), while all of them were able to activate NK cells, which upregulated CD69 and CD161 activation markers following f-CNTs treatment. Our study[
31], which was the first one exploring the impact of nanotubes on natural killer cells, suggests that the use of these f-CNTs could be explored as tool for NK expansion, for example in pre-clinical model of NK-based adoptive therapy. Similarly, the up-regulation of CD25 and the production of IL-6 by CD14+ monocytes upon treatment with ammonium-functionalized CNTsrevealed the activatory properties of these nanotubes (see discussion in the following section) on innate immune cells. Few research groups have started to look at the impact of graphene and GO on the immune system. The group of Liu showed that the complex between GO and anti-IL10 receptor antibodies elicited LPS-stimulated CD8 T cell responses, which makes this functionalized graphene suitable for reprogramming suppressive tumor environment in experimental models[
34]. The first extensive investigation on pristine and GO to assess hemocompatibility was performed by Sasidharan
et al.[
35]. The authors observed a significant release of IL-8 and IL-6 in pristine graphene treated PBMC, while GO was able to determine only a moderate release of IL-8. No modulation of the other inflammatory cytokines (IL1B, IL1, TNF, and IL12) was observed. Importantly, the authors showed an excellent compatibility of both types of graphene in multiple assays assessing hemolysis, platelet aggregation and coagulation. The group of Cui addressed whether the coating of GO with polyvinylpyrrolidone (PVP) could increase biocompatibility GO with human immune cells, including T lymphocytes[
38]. Besides confirming the GO dose-dependent toxicity on T cells, the authors showed that the induction of apoptosis in these cells was much less pronounced when PVP-coated GO was used[
38].