CD34+DNAM-1brightCXCR4+ haemopoietic precursors circulate after chemotherapy, seed lung tissue and generate functional innate-like T cells and NK cells

Background There is little information on the trajectory and developmental fate of Lin-CD34+DNAM-1bright CXCR4+ progenitors exiting bone marrow during systemic inflammation. Objective To study Lin-CD34+DNAM-1bright CXCR4+ cell circulation in cancer patients, to characterize their entry into involved lung tissue and to characterize their progenies. Methods Flow cytometric analysis of PBMC from 18 patients with lung cancer on samples collected immediately before the first and the second treatment was performed to study Lin-CD34+DNAM-1bright CXCR4+ precursors. Precursors were purified (>99%) and cultured in vitro from all patients. Paired PBMC and tissue samples from patients undergoing tumor resection were analyzed by flow cytometry to assess tissue entry and compare phenotype and developmental potential of Lin-CD34+DNAM-1bright CXCR4+ cells in both compartments. Results Significant circulation of Lin-CD34+DNAM-1bright CXCR4+ precursors was observed 20d after the first treatment. Precursors express CXC3CR1, CXCR3, CXCR1 consistent with travel towards inflamed tissues. Flowcytometric analysis of lung tissue samples showed precursor presence in all patients in tumor and neighboring uninvolved areas. Successful purification and in vitro culture from both blood and lung tissue generates a minor proportion of maturing NK cells (<10%) and a predominant proportion (>85%) of α/β T-progenies with innate-like phenotype expressing NKG2D,NKp30,DNAM-1. Innate-like maturing T-cells in vitro are cytotoxic, can be triggered via NKR/TCR co-stimulation and display broad spectrum Th1,Th2 and Th1/Th17 cytokine production. Conclusion In advanced stage lung cancer CD34+DNAM-1brightCXCR4+ inflammatory precursors increase upon treatment, enter involved tissues, generate functional progenies and may thus represent an additional player contributing to immune balance in the highly SDF-1/CXCR4-biased pro-metastatic tumor microenvironment.


Introduction
During the course of inflammatory conditions including HIV, HCV, MTB and COPD, an unconventional CD34 + DNAM-1 bright CXCR4 + hemopoietic stem cell (HSC) exits the bone marrow (BM) and circulates in peripheral blood (PB) (1).These inflammatory HSCs are common lymphoid precursors (CLP) that express increased proportions of the chemokine receptors CXCR1 and CX3CR1 when compared to PB or cord blood CD34 + CXCR4 - DNAM-1 -HSCs, and lower proportions of CCR7 and CD62L (1).Accordingly, it has been proposed that they may seed inflamed peripheral tissues where fractalkine/CX3CL1 Eotaxin-3/CCL26 and IL-8/CXCL8 are abundant (2)while still maintaining the ability to seed secondary lymphoid tissues through CCR7 and CD62L signaling.
Different from conventional CD34 + CXCR4 -precursors, inflammatory CD34 + DNAM-1 bright CXCR4 + cells very rapidly generate in vitro NK cell and T cell progenies with predominant (2:1) frequency of maturing NK cells (1).In the presence of an HLA-E rich milieu these precursors generate in vitro highly functional memory-like NKG2C + NK cell producing IFN-g and controlling CMV replication (1,10).
The level of lymphoid infiltration in cancer tissues represents an attempt of the immune response to control the growth of transformed cells in the tumor microenvironment and contributes to the control of tumor spreading (11,12).Lymphocyte infiltration density correlates inversely to the risk of recurrence in colon cancer (13) and to the response rate to chemotherapy (14).Indeed, tumors with less lymphoid infiltration and with reduced transcript activation with an overall "frozen" transcriptional landscape have worst chances to be controlled by the immune system and by treatments involving checkpoint inhibition (15).In these instances, peripheral blood lymphocyte subsets may also be associated with disease response to immunotherapy (16).
The tumor itself and altered tumor-associated cells may contribute to the production of countermeasures that inactivate infiltrating lymphocyte functions, reduce or dampen inflammation, and divert the tumor infiltrate favoring immune tolerance.
Stem cell circulation in blood after exiting the BM with a trajectory towards peripheral tissues has been well documented in animal models for endothelial- (17), skeletal muscle- (18) and skeleton- (19) tissue-specific stem cells and also for hemopoietic stem cells (20).In humans, tissue seeding of lymphocyte eosinophil progenitors has relevant implications in asthmatic patients during allergen challenge, where they increase in BM, in peripheral blood and in bronchial mucosa (21)(22)(23)(24)(25).
There is little information on CD34 + DNAM-1 bright CXCR4 + Inflammatory precursor entry in inflamed tissues and particularly into cancer tissue, while trafficking and developmental trajectory towards secondary lymphoid organs or thymus of conventional CD34 + DNAM-1 -CXCR4 + cells is well characterized.In order to address this question, we here studied if and when CD34 + DNAM-1 bright CXCR4 + cells may be recovered from peripheral blood mononuclear cells (PBMC) of patients with lung tumor and whether they generate progenies with comparable characteristics.
The present work provides insight in timing of inflammatory HSC release from BM in patients with NSCLC, their entry into involved inflamed tissues, and in the characteristics of their progenies.

Patients
For initial analysis of inflammatory precursors, peripheral blood mononuclear cells (PBMC) from 18 cancer patients with progressive disease, encompassing lymphoma, non-small cell lung cancer (NSCLC), and Kaposi's sarcoma (KS) were collected and compared with healthy donors (n= 18).Additionally, we included samples from HIV patients (n= 15) and COVID patients (n= 28) previously reported in the literature.
Subsequently, we investigated patients with advanced lung cancer scheduled to undergo concurrent chemo-immunotherapy (CT/IT) at baseline, just prior to the initial CT/IT administration (Time 0, T0), and at day 21, immediately before the second CT/IT cycle (Time 1, T1).
For the evaluation of tissue penetration, we analyzed paired blood/tissue samples from five NSCLC patients undergoing surgery according to best practice guidelines.Cancer tissue (C-T) samples and distant uninvolved tissue samples here defined "uninvolved tissue" (U-T) were collected.
Patient characteristics and treatments are indicated in Table 1.Sampling was approved by the local ethics committee (registry number: P.R. 191REG2015) and all samples provided informed consent to the observational study.

Cell isolation and culture
PBMC were obtained from peripheral blood by density gradient centrifugation.Samples were cryopreserved until processed.Where applicable, paired samples from the same patient were processed at the same time.Tissue samples were obtained immediately after surgical resection for diagnostic purposes and were selected from Pathology Department of San Martino Hospital, Genoa and Pathology Unit of Sacro Cuore Hospital, Negrar, Verona.Neoplastic and non-neoplastic tissues were mechanically dissociated and cell suspensions obtained were filtered through a 40mm cell strainer (Jet Biofil, Guangzhou, China).Cell suspensions were separated by density gradient centrifugation to obtain PBMCs (Ficoll-Hypaque).Cells were either directly analyzed by flow cytometry or cryopreserved at -86°C for further analyses.Highly purified Lin -CD34 + DNAM-1 bright and CD3 -CD4 -CD19 - CD56 -CD16 + cell populations (>99% purity) were obtained using FACSAria III (BD Biosciences) cell sorter.Purified cells were cultured in limiting dilution conditions in Myelocult medium (StemCell Technologies, Vancouver, British Columbia, Canada) supplemented with 10% human AB serum (ICN Pharmaceuticals Italy, Milano, Italy), 5% FCS and purified recombinant human rhIL-15, rhIL-7, SCF, FLT3-L (PeproThec, London, UK) at the final concentration of 20 ng ml -1 with irradiated Feeder cells, for 30 days.

Immunofluorescence analysis
Cells were analyzed by multi-parameter flow cytofluorometry.Direct staining was performed incubating cells with fluorochromeconjugated monoclonal antibodies (mAbs) for 15 minutes at 4°C.Cells were then washed and the flow cytometric analysis was performed (FACSFortessa, BD, Mountain View, CA, USA).Mean fluorescence intensity ratios are calculated as follows: MFI sample/ MFI negative control and mean fluorescence intensity absolute are calculated as follows: MFI sample-MFI negative control.Data were analyzed using FlowJo (Tree Star, Inc, BD, Ashland, Ore) and FCS Express 7 (De Novo Software, Pasadena, Calif).

Cytotoxicity assay
The cytotoxic activity of in vitro growing progenies was determined using a PKH-26 and TO-PRO-3 (Sigma-Aldrich and Invitrogen, respectively) assay as previously described (26).FcgR+ P815 mouse mastocytoma and A549 human lung carcinoma cell lines were used as target cells and were labeled with PKH-26 (26).A549 PKH-26 + cells were incubated with cell progenies at a 1:1 effector:target (E:T) ratio.Cultures were incubated for 6 h at 37°C in 5% CO2 in complete medium and then placed on ice until flow cytometric analysis.
Cytotoxic activity against P815 cells was tested in a reverse ADCC (Ab-dependent' cell-mediated cytotoxicity) at a 1:1 effector: target (E:T) ratio in complete medium in the absence or presence of mAbs (0.1 mg/mL).P815-PKH-26 + cells were incubated with effector cells for 6 h at 37°C in 5% CO2 in complete medium and then placed on ice until flow cytometric analysis.Spontaneous and maximal target cell deaths were determined by PKH-26 labeling of cells cultured alone and permeabilized with BD Cytofix/Cytoperm reagent (BD Pharmingen), respectively.To identify dead cells, 5 ml of a 10 mM stock solution of TO-PRO-3 was added to each tube immediately before analysis.Cells were analyzed by FACSCFortessa (BD), and 10.000 events were collected.Specific lysis was calculated by use of the following formula for dye-labeled cells: (sample − spontaneous)/(total − spontaneous) × 100.

Statistical analysis
Statistical analysis was performed using the Mann-Whitney U test for unpaired datasets for comparisons.Analysis was performed using JMP 10.0 (SAS) if not otherwise stated.Significant differences (two-tailed) are reported in the text and figures, while nonsignificant differences are reported in the figure legends only.

Results
Normal baseline and increased CD34 + DNAM-1 bright CXCR4 + precursor circulation after chemotherapy/ immunotherapy in lung cancer patients We first verified whether CD34 + DNAM-1 bright CXCR4 + "inflammatory" CLP that are detected in PBMC of patients with acute or chronic infections (1,27,28) could also be detected in cancer patients.We investigated PBMC of sequential patients on treatment for non-Hodgkin lymphoma, NSCLC, Kaposi's sarcoma (KS).
In order to study the dynamic evolution of inflammatory precursor circulation in cancer patients, we next studied PBMC of a new cohort of 18 patients with newly diagnosed advanced lung cancer by flowcytometry with sampling at diagnosis before any systemic treatment and after 21 days just before the second treatment.Demographic and clinical characteristics are reported in Table 1.
In order to determine whether the observed increase was part of a generalized mobilization of CD34 cells from the BM, or rather an event restricted to inflammatory precursors, the fractional frequency of conventional and of inflammatory CD34 + cells over total CD34 + precursors were considered.Interestingly, total circulating CD34 + cell frequencies were not increased following CT/IT (Figure 2C), while the proportion of CD34 + DNAM-1 bright cells significantly increased after CT/IT (15,31 ± 10,8 vs. 29,36 ± 19,9; T0 vs. T1; p=0.005; Figure 2D).These findings indicating a selective BM exit of Lin -CD34 + DNAM-1 bright precursors (Figure 2B) was also confirmed by a higher fold-change in their frequency after CT/IT compared to classical CD34 + DNAM-1 - (p=0.03; Figure 2E).
Overall, therefore, these data indicate that in patients with lung tumor only CT/IT induces selective increase of specialized CD34 + DNAM-1 bright CXCR4 + precursors.

Identification of Lin -CD34 + DNAM-1 bright CXCR4 + cells in lung tissue samples from NSCLC patients
Circulating Lin -CD34 + DNAM-1 bright cells have been shown to have a different expression of chemokine receptors when compared to canonical CD34 + DNAM-1 -cells that would support their migration into inflamed tissues (1).
In this regard, although DNAM-1 is known as an activating receptor constitutively expressed by NK cells, T cells, macrophages, and DCs [(30) specifically recognizing two cell ligands (PVR, CD155 and Nectin 2, CD112] (31), it also is crucially involved in transendothelial cell migration (32).Expression of high DNAM-1 densities on inflammatory precursors therefore could contribute to transendothelial cell migration of Lin -CD34 + DNAM-1 bright CXCR4 + cells out of BM niches and also into inflamed tissues possibly leading to is down-expression upon tissue entry.Identification of inflammatory precursors in PB of cancer patients by flow cytometry.(A) inflammatory precursors (Lin − CD34 + DNAM-1 bright CXCR4 + and Lin -CD56 -CD16 + CD7 -CXCR4 + are recovered from PBMC of patients with different tumors (representative of 3 patients each cancer type).Flow cytometric gating strategy to identify and characterize CXCR4+ CLPs in cancer PBMC.Among Lineage -(CD3 -CD14 -CD19 -CD20 -CD56 -) (Lin -) gated cells Lin − CD34 + DNAM-1 bright and Lin -CD56 -CD16 + CD7 -cells are shown.To address these questions we therefore analyzed parallel samples of tumor tissue, uninvolved lung tissue and PBMC from 5 patients undergoing surgery for lung cancer CD34 + DNAM-1 bright CXCR4 + cells could be identified, in addition to PBMC, in all tissue samples.Analysis of DNAM-1 MFI in CD34 + DNAM-1 bright CXCR4 + cells (Figure 3A) showed a higher mean DNAM-1 density in peripheral blood vs. cancer-tissue precursors (p=0.03; Figure 3B).When analyzing MFI with negative control correction (MFIr), the difference in DNAM-1 molecule density was more comparable in blood and in tissue samples.(Figure 3C).

Peripheral inflammatory precursors in cancer patients are oriented towards inflamed tissues according to chemokine receptor expression
We next analyzed the frequency of Lin -CD34 + DNAM-1 bright CXCR4 + cells recovered in blood, tumor and uninvolved tissue samples in order to provide an estimate of their tissue trafficking.The frequency of peripheral blood Lin -CD34 + DNAM-1 bright CXCR4 + cells was higher compared to tissue Lin - CD34 + DNAM-1 bright CXCR4 + cells, while no difference was observed between involved and uninvolved tissue (Figure 4A).Cell migration into tissues follows chemokine gradients and chemokine receptor expression governs lymphocyte trafficking into tissues.Accordingly, different chemokine receptor expression by these cells could associate with differences in observed PB vs. tissue frequencies.
Thus, when considering these CLPs in PB, expression of CX3CR1 and CCR7 predominates on Lin -CD34 + DNAM-1 bright CXCR4 + cells with low levels of CXCR1, CXCR3, CD62L, and CCR7.On the other hand, Lin -CD56 -CD16 + CD7 -cells display a different chemokine receptor signature, with predominant expression of CX3CR1 and CD62L and lower expression of CXCR1 and (Figure 4B and Supplementary Figures 1A, B).Overall, the frequency of CCR7 and CD62L expression on Lin - CD34 + DNAM-1 bright CXCR4 + in PBMC (40-50%) suggests that this proportion could enter secondary lymphoid organs (Figure 4B), thus accounting for a reduced fractional recovery in tissues compared to PB.
To verify this possibility, chemokine receptor expression was studied by flow cytometry also on common lymphocyte precursors recovered from tissues.Both Lin -CD34 + DNAM-1 bright cells and Lin-CD56 -CD16 + CD7 -tissue cells were recovered in patient tissues and were enriched for the expression of chemokine receptors driving towards inflamed tissues.Indeed, their combined mean expression of CCR7 or CD62L accounted for only 24% and 14% respectively, with correspondingly increased frequencies of CX3CR1, CXCR3 and CXCR1 expression (Figure 4B).Thus inflammatory precursors from PBMC expressing CXCR1, CXCR3 and CX3CR1 selectively enter inflamed tumor tissues.

Purification, culture and progeny characterization of CLP from tissues and peripheral blood in cancer patients
In patients with chronic infections, the precursors identified as Lin -CD34 + DNAM-1 bright and Lin -CD56 -CD16 + CD7 -both generate in vitro NK cell progenies with a minor (25%) frequency of T cell progenies (10).However, no information has been available on the behavior of these precursors when they migrate into tissues.Therefore, following the isolation of these inflammatory precursors from both peripheral blood (PB) and lung tissue in patients with NSCLC, we conducted a comparative characterization of their progenies.Tissue mononuclear cell preparations were obtained from both uninvolved lung tissue (U-T) and cancer tissue (C-T) samples derived from seven different patients, along with corresponding PB samples.Highly purified cells (Figure 5A B C A In vitro culture of highly purified Lin -CD34 + DNAM-1 bright and Lin -CD56 -CD16 + cells from peripheral blood (PB), cancer tissue (C-T) and uninvolved tissue (UI-T) of lung cancer patients generates NK and T progenies.Frontiers in Immunology frontiersin.orgdetails the gating strategy) were then cultured in vitro under limiting dilution conditions.Progenies were identified optically for growth after 16-20days from purification and seeding, and subsequently analyzed after 30 ± 1days (mean ± SEM).Quantitatively sufficient growing cultures were obtained from all patients from purified Lin -CD34 + DNAM-1 bright including 45 cultures from C-T, 43 cultures from U-T and 59 cultures from PB.Similarly, from highly purified Lin -CD56 - CD16 + CD7 -precursors we obtained 33 cultures from C-T, 45 from U-T and 61 from PB. Flow cytometric analysis showed that precursor cultures were composed of lymphoid cells that were in majority CD56 -CD3 + (85-90%) with lower frequency of CD56 + CD3 -(<10%) and a minor representation of CD56 + CD3 + cells (4%) which were not further studied (Figure 5B).
Additionally, T-cell progenies predominantly expressed the ab T cell receptor (TCRab), with only a minor frequency of the gd TCR (<2-3%) (Figures 7A, B).Analysis of CD45RA expression on CD4 + cells revealed a bimodal distribution, with a higher proportion of CD4 + CD45RA -cells and a consistent but smaller fraction of CD4 + CD45RA + cells across all samples and patients (Figure 7C).
To further characterize the T helper (Th) differentiation potential of CD4 + CD45RA -progenies, we employed flow cytometry to assess their expression of Th1, Th2, and Th17associated markers, including CXCR3, CCR4, CCR6, and CCR10 according to a previously described method with a gating strategy depicted in Figure 7D.Accordingly, we successfully distinguished various T helper (Th) subsets, including also Th1*.These Th1* cells, as defined by Sallusto represent a unique subset of Th cells that exhibit a dual phenotype, combining characteristics of both Th1 and Th17 cells.Notably, Th1* cells express CXCR3 and produce IFN-g, two hallmarks of Th1 cells, while also expressing CCR6, a marker typically associated with Th17 cells.However, in contrast to Th17 cells, Th1* cells do not produce IL-17 (33, 34).

Maturing progenies from inflammatory CLP migrated to tumor tissue are functionally comparable to those of CLP purified from PBMC in NSCLC patients
In light of the diverse array of T helper cell (Th) subsets identified among CD4+ T cell progenies based on their chemokine receptor expression, we assessed the actual cytokine secretion capabilities of these progenies.Culture supernatants were evaluated either without additional stimulus beyond culture conditions ("unstimulated") or after additional activation with PMA+ionomycin ("stimulated").
As depicted in Figure 8, progenies exhibited a broad range of cytokine production in basal conditions, which could be further enhanced upon stimulation, aligning with the functional characteristics of Th1, Th1*, and Th17 cells.Notably, the supernatants demonstrated production of IL-2, TNFa, IFN-g, IL-5, IL-13, IL-9, and IL-17a (Figure 8).
To additionally evaluate the functionality of T progenies derived in vitro from their highly purified precursors, a cytotoxicity assay was performed using a lung cancer target cell line A549 or a mouse FcgR + P815 cells in the presence of mAbs specific for triggering receptors, in a redirected killing assay (reverse ADCC).Double staining with PKH-26 and TO-PRO-3 was used to detect killed target cells that appear PKH-26 + TO-PRO-3 + (Figure 9A).T progenies exhibited cytotoxicity towards A549 targets at an effector-to-target ratio of 1:1.Those generated in vitro from Lin - CD34 + DNAM-1 bright cells purified from cancer tissue displayed significantly greater cytotoxicity compared to those derived from tissue Lin -CD56 -CD16 + cells and from peripheral blood (PB) (Figure 9B).In the redirected killing assay, T cell progenies from tissue-derived CLPs exhibited potent direct cytotoxic activity upon activation via CD3 and NKp30, outperforming those derived from PBMCs (Figure 9C).
Taken together these results indicated that T cell progenies derived from inflammatory precursors purified from blood or from inflamed tissues are functional, produce cytokines consistent with different functional polarization and may be cytotoxic by engagement of their activating receptors.

Discussion
In the present work we provide evidence that in patients with advanced stage NSCLC, baseline circulation of inflammatory CLP released from BM (1){Bozzano, 2015 #8} is low and increases following the initiation of CT/IT.In addition, these cells enter peripheral tissues including cancer tissue where they are found after chemokine-selective migration and give rise in vitro to a majority of functionally active T cells co-expressing activating NK cell receptors.
The finding of a normal baseline circulation of inflammatory precursors in PBMC of patients with advanced stage NSCLC and their increase following CT/IT are in line with the notion that the inflammatory effect of the tumor on the surrounding tissue milieu favors tumor growth and prevents or dampens effective immune responses.Chronic inflammatory conditions may increase the risk of developing cancer (35,36) and of cancer progression (37).The inflammation orchestrated by the tumor is aberrant and promotes the recruitment and/or the induction of cells that, besides having a role in the direct promotion of the tumor progression, are also endowed with immunosuppressive properties (38).Indeed, patients with NSCLC and little inflammation at baseline before CT/IT have higher progression-free survival (39).
The present observation of a selective increase of CD34 + DNAM -1 bright CXCR4 + after CT/IT, is in line with the inflammatory effects of CT/IT leading to recruitment of inflammatory precursors to the inflamed peripheral tissues and with previous reports where lymphoid precursors including CD34 + HSCs increase in the bone marrow, circulate at higher frequency and seed into peripheral tissues (17)(18)(19)(20) to replenish immune cells upon increased peripheral turnover (21-25, 40, 41).
Tissue seeding of inflammatory precursors in this case is supported by recovery of precursors in all samples, by chemokine receptor enrichment in tissues favoring CXCR3 + , CXCR1 + and CXCR3 + cell entry, by their consistent expression of DNAM-1 and CXCR4 in tissues, and by the similar pattern of progeny generation with characteristics of a T-cell-skewed common lymphocyte precursor both in PBMC and in lung tissues.
This view is also suggested by the difference in CD34 + DNAM-1 bright CXCR4 + cell progeny phenotype detected in lung tumor (85-90% T-cells) compared to the considerably lower one (20-30%) reported in patients with infection (CMV, HIV, HCV) (1, 10).These observations could be explained by the hypothesis that different subsets of inflammatory CD34 + DNAM-1 bright CXCR4 + exist in the BM, that different inflammatory signatures (e.g.tumor vs. infection, acute vs-chronic inflammation) may induce a different upstream precursor development and/or their skewed exit from the BM.Clearly, additional work is needed to further address these points and verify this hypothesis.
With regard to the progenies derived from inflammatory precursors entering lung tissue, also in this case we confirmed previous reports of absent growth of myelomonocytic cells (1,10) under the same culture conditions that allow growth of NK-and myelomonocitic progenies using cord-blood derived CD34 + cells (42,43).Analysis of maturing T cells in vitro after purification of tissue precursors showed that these cells have the ability to produce Th1, Th17 and Th2 cytokines in vitro, with a predominant representation of cells with the potential to produce IFNg (Th1 and Th1*).Interestingly, characterization of these T cell progenies revealed their consistent expression of NK cell activating (NKG2D, DNAM-1, NKp30 that appear to confer HLA-independent function when crosslinked.Although the present analysis was limited in T cell clonality and specificity, according to NK cell receptor expression and function and to Th1 and Th17 cytokine production it appears that these T cell progenies could represent innate-T cells that may be triggered in a TCR-independent fashion.This may be relevant particularly in tumor areas and in patients in whom tumor antigens are elusive or with low HLA-Class I molecule expression. The extent to which the present in vitro differentiation system accurately replicates the diverse conditions of tissues or blood, including the variable levels of local cytokines, chemokines, and stromal factors, remains to be determined.When considering the impact of tissue-specific conditioning on CLP differentiation, two important considerations should be taken into account.Firstly, all cells analyzed were rigorously purified from distinct tissues after a prolonged period of exposure to their local environment.This suggests that the purified precursors had ample opportunity to be conditioned by local cytokines/factors.However, despite this, their extraction and subsequent in vitro culture yielded similar progenies in both tissue and peripheral blood mononuclear cells (PBMCs).Furthermore, utilizing the same cytokine cocktail employed for NK cell development from CD34 + hematopoietic stem cells in vitro resulted in the predominant differentiation of T cells, suggesting that the ability of inflammatory CLPs to follow a specific developmental program may be determined elsewhere (e.g., bone marrow) and that tumor-derived or tissue stimuli may only be able to partially influence their functional trajectory.Future studies are needed to investigate whether and under which experimental conditions different cytokines and stimuli present in tumor tissues might influence CLP development.
Finally, a particularly intriguing question arising from this study concerns the potential relationship between the mobilization and dissemination of inflammatory CLPs to tumor tissues, the extent of tumor lymphocyte infiltration (TIL), and subsequent disease progression.Unfortunately, the current study was not designed to directly assess the association between inflammatory CLPs, TIL, immunoscore, and disease progression in lung cancer patients.Additionally, the limited sample size precluded the detection of associations between inflammatory precursor levels and disease progression.Indeed, information regarding the dissemination of inflammatory CLPs in tumor patients constituted one of the study's objectives.Consequently, further investigations are warranted to address these questions.
In conclusion, our findings unveil a coordinated inflammatory response to CT/IT that triggers the mobilization and deployment of CXCR4+ precursors to tissues, resulting in the generation of highly functional innate immune cells.This observation provides a new perspective for interpreting the current findings.Tumor microenvironments in the lung favor inflammation by exploiting the SDF/CXCR4 axis, leading to angiogenesis, tumor progression, and metastasis.The SDF/CXCR4 axis plays a crucial role in several mechanisms that promote tumor growth and metastasis (44,45).Additionally, it contributes to the establishment of immunosuppressive tumor microenvironments, favoring the infiltration of myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs) into tumors (46,47).Accordingly, the circulation and tissue infiltration of CXCR4 + inflammatory precursors following chemotherapy/immunotherapy (CT/IT) in lung cancer patients represent an additional factor that could be exploited to enhance cancer monitoring and immunotherapy strategies by regulating the immune response against tumors.
(A) Flow cytometric purification strategy and culture of Lin -CD34 + DNAM-1 bright and Lin -CD56-CD16+ cells isolated from PB, C-T and UI-T.(B) Characterization of in vitro cultures from highly purified Lin -CD34 + DNAM-1 bright (black columns) and Lin -CD56 -CD16 + (grey columns) cells from PB, C-T and UI-T.Bars express the proportion of cells recovered in culture.Maturing T cell progenies are predominant in in vitro cultures derived from both CLPs.Histograms (mean ± sem) show 18 different cultures derived from each of the two CLP populations isolated from 4 different PB, 7 different C-T and 7 different UI-T samples.Mann-Whitney U-test analysis is shown.(C) Expression of CD4 and CD8 surface molecules on CD56 -CD3 + T cell progenies derived from highly purified Lin -CD34 + DNAM-1 bright (black columns) and Lin -CD56 -CD16 + (grey columns) cells from PB, C-T and UI-T.Histograms (mean ± sem) show 18 different cultures derived from each of the two CLP populations isolated from 4 different PB, 7 different C-T and 7 different UI-T patients.(no symbol corresponds to: p=n.s.; Mann-Whitney U-test, all comparisons).SSC-A, side scatter area."**" indicates a p value < 0.01, "*" indicates a p value < 0.05.Perrone et al. 10.3389/fimmu.2024.1332781

TABLE 1
Patient characteristics.