*3.3. Cytokine Secretion Dynamics of Immune Cells Co-Cultured with Nasopharyngeal Cells*

In the measurements of the cytokine dynamics, immune cells (1 × 105 cells/mL) and nasopharyngeal cells (1 × <sup>10</sup><sup>5</sup> cells/mL) were cocultured in the culture chamber, with either undifferentiated THP-1 cells or differentiated macrophages applied as the immune cells, and either EBV-positive nasopharyngeal carcinoma (NPC43) cells or EBV-negative nasopharyngeal (NP460) cells applied as the nasopharyngeal cells. Macrophage cells were differentiated from the THP-1 cells by PMA with a concentration of 50 ng/mL for 24 h before the measurements. The cytokine measurements of TNF and IL-12-p70 were performed prior to the culture and at 3, 4, 5, 6, 7 and 8 h of culture. The first three hours of culture could offer a stable environment for cell adaptation.

In the measurements as shown in Figure 4a, we have considered both NPC43 and NP460 in three conditions for different purposes: (1) cocultured with undifferentiated THP-1 on a flat surface as a control case, (2) cocultured with THP-1-differentiated macrophages on a flat surface for estimating the NPC43/NP460-stimulated cytokine secretion, and (3) cocultured with THP-1-derived macrophages on microgratings for revealing the topographyinduced cytokine responses.

For the TNF measurements, our results show that the differentiated macrophages secrete a much higher level of the pro-inflammatory cytokine TNF than the undifferentiated THP-1 monocytic cells for both cocultured cases (NPC43 and NP460). The TNF secretions of the cell cocultures on microgratings have an increment of ~20% on average, compared to the cells on flat surfaces, implicating an underlying mechanism related to the microtopographic factors (Figure 4a). Nevertheless, the cytokine secretions of both macrophages and monocytic cells cocultured with NPC43 are less than those cocultured with NP460, suggesting that NPC43 may suppress the TNF secretion of immune cells. One possible explanation is that EBV can suppress TNF-α synthesis from lipopolysaccharide-treated monocytes at both protein and transcriptional levels as reported previously [33].

For the IL-12p70 measurements, it is noteworthy that a distinct secretion profile of IL-12p70 was observed compared to that of TNF (Figure 4b). Though similar trends as the TNF secretion cases have been shown that the IL-12p70 levels are, in general, higher (1) in the macrophages cocultures than in the monocyte cocultures and (2) on the microgratings than on the flat surface for the macrophage cocultures; the coculture with NPC43 does not necessarily induce higher IL-12p70 levels than the coculture with NP460. Furthermore, IL-12p70 secretions of the macrophage-NPC43 cocultures have increasing cytokine levels in the initial stage (0–6 h), reaching the maximum levels at ~6 h and gradually decreasing afterward. Notably, the coefficient of variation (CV), the ratio of the standard deviation to the mean, was investigated for repeatability, which was less than 15% for all data.

**Figure 4.** (**a**) The concentration of TNF secreted from cocultured undifferentiated THP-1 and THP-1-differentiated macrophages with NPC43 cells or NP460 cells, with and without a parallel grating array. (**b**) The concentration of IL-12p70 secreted from THP-1 and macrophages co-cultured with NPC43 cells or NP460 cells, with and without parallel grating array (*n* = 4 for each point). Error bars are the standard errors in all plots. Asterisk represents a *p*-value of <0.05.

These results reveal some insightful observations. Interestingly, the macrophage-NPC43 cocultures exhibit suppressed IL-12p70 expressions after 6 h of coculture, implicating an underlying related mechanism between macrophages and nasopharyngeal cancer cells. This agrees with previous clinical studies that nasopharyngeal carcinoma patients have a reduced level of IL-12p70 in their serum [9]. This indicates the suppression of IL-12 secretion, which can be caused by the induced MCP-1 expression of the EBV-infected macrophages [7]. EBV-infection of microphages can also promote polarization to the M2 macrophages [34], leaving a smaller portion of IL-12p70 secreting M1 macrophages [35]. On the other hand, our results show that the microgratings as a microtopographic factor can induce the cytokine secretions of TNF and IL-12p70 on top of the molecular interactions between the immune and nasopharyngeal cells. One possible explanation is that cells adhering to the parallel gratings have different cell behaviors, e.g., morphology and migration [23], which can then affect the immune-cancer cell interaction to some extents. Therefore, it is worthwhile to further apply the reported microfluidic immunoassay to investigate the cell-microenvironment dependency through the simultaneous monitoring of cell behaviors during the coculture periods.

#### *3.4. Cell Migration*

We further investigated the migration behaviors of NP460 and NPC43 single cell co-cultured with or without THP-1 derived macrophages on grating platforms. We seeded NP460/MPC43 cells at a density of 5 × <sup>10</sup><sup>2</sup> cells/cm2 and each of the co-cultured cells at a density of 2.5 × 102 cells/cm2, followed by culturing the cells for 8 h and monitoring their migration under a microscope. Migration trajectories of NP460 and NPC43 cells growing on the planar/microgratings substrates (along 90◦/270◦) are shown in Figure 5a. Clearly, the cells on planar surfaces display a random migration trajectory, whereas the cells on microgratings migrate with a direction along the microgratings.

**Figure 5.** (**a**) Migration trajectories of single NP460 and NPC43 cells on flat/grating substrates. (**b**) Migration speed of single NP460, single NPC43 and those co-cultured with THP-1 derived macrophages on platforms with/without grating. *n* > 35 for all cases. Error bars represent the standard errors. Asterisk represents a *p*-value of <0.05.

Our results (Figure 5b) further indicate that NPC43 cells migrate faster than NP460 cells on both planar and micrograting surfaces. NPC43 cells can migrate even faster when they are cultured on micro-grating substrates than on planar substrates. On the other hand, NP460 cells migrate slower on microgratings than they do on planar substrates, which suggests the different responses of NP460 and NPC43 cells upon the microgratings topography. Furthermore, the co-cultures of macrophages with NP460/NPC43 cells on planar substrates induce faster cell migration, suggesting that molecular secretions of macrophages can promote cell migration. In fact, it has been reported that the MCP-1 secreted by macrophages can promote cell migration [36]. Furthermore, our results indicate that >5% of macrophages and NPC43 cells adhered together on microgratings without further migration, whereas the other cells appearing as single cells without noticeable cell-cell contact can still maintain at a faster migration speed. Interestingly, the cytokine and migration measurements exhibit that nasopharyngeal cancer cells can stay on the microgratings with suppressed IL-12 secretion of the contacting macrophages, supporting the higher tendency of nasopharyngeal cancer spreading to the grating-like pterygoid muscles [37]. Together, the microgratings can affect cell migration behaviors and possibly the intracellular interactions of nasopharyngeal cells

(NP460 and NPC43) and immune cells. For example, it is worthwhile to further examine the correlation between the direct NPC43/macrophage contact and the suppressed IL-12 secretion, and the underlying mechanism.

There are several limitations in the current study. For instance, the effects of interference components such as cell debris in detection samples would be eliminated by integrating a porous membrane filter [38] for sample pretreatments before cytokine detection. Integrating the current microfluidic immunoassay with non-washing cytokine detection strategy such as AlphaLISA [39] may further improve the detection sensitivity and specificity. Simultaneous monitoring of cell behaviors and highly multiplex cytokine detection during the coculture periods using the developed immunoassay would provide valuable insights into the comprehensive and dynamic immune status in solid tumors and during inflammatory states that result in heterogeneous tumor microenvironmental features for precision medicine. Moreover, standardized and automated fabrication setup such as cost-efficient and multilayer PDMS aligner [40] should be further developed for large-scale and high throughput fabrication of the developed microfluidic immunoassay.

#### **4. Conclusions**

In conclusion, we have reported a multifunctional microfluidic immunoassay by integrating microtopographic cell-culture substrates with a microbeads-based immunofluorescence assay that enables parallel detection of different immune biomarkers and intercellular behaviors in a rapid, sensitive, and easy-to-implement manner. The developed assay exhibits the advantages of the simultaneous investigation of different cytokines and cell migration behaviors on flat/grating ECM substrates, requiring a low-volume sample (0.5 μL) and short assay time (30 min) but a sensitive performance in a wide range of cytokine concentrations (5–5000 pg/mL). Secretions of TNF and IL-12p70 were successfully monitored throughout the co-culture period to evaluate the different immunological states of undifferentiated THP-1 monocytic cells or PMA-differentiated THP-1 macrophages cocultured with immortal cells NP460/NPC43 on flat and micrograting surfaces. We believe that the reported immunoassay is a promising approach to allow continuous, broad-range and precise on-chip characterization of cytokine and intercellular interactions on different topographical substrates, and thus provides clinical significance for early tumor diagnosis and treatment.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/bios12110963/s1, Figure S1: Fabrication process of the integrated microfluidic immunoassay; Figure S2. A microscope system integrated with a confining shield (a) and computer-controlled compressed air supply manifolds (b); Video S1: Device operation for extracting culture media from the cell culture chamber; Video S2: Cytokine detection procedures in one detection unit integrated with an active micromixer.

**Author Contributions:** Conceptualization, X.C. and R.H.W.L.; Data curation, X.C., L.L., J.L., Y.L. (Yi Liu) and D.H.; Formal analysis, L.L., J.L., D.H., R.Z., S.H., Z.J. and Y.W.; Funding acquisition, X.C. and R.H.W.L.; Investigation, X.C., S.W.P. and R.H.W.L.; Methodology, X.C., L.L. and S.W.P.; Project administration, S.W.P.; Resources, Y.L. (Yi Liu), Y.L. (Ya Liu), R.Z., S.H., Y.Q., S.W.P. and R.H.W.L.; Software, J.L. and Y.Q.; Supervision, R.H.W.L.; Validation, L.L., J.L., Y.L. (Yi Liu), Y.L. (Ya Liu) and Y.W.; Visualization, J.L., Y.L. (Ya Liu), Z.J. and Y.Q.; Writing—original draft, X.C., L.L. and J.L.; Writing—review & editing, X.C. and R.H.W.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the General Research Grant (11215619 and 11216220) of Hong Kong and Natural Science Foundation of Guangdong Province (2021A1515011167 and 2020A1515010332).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.
