Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Mar 15.
Published in final edited form as: J Neurosci Methods. 2015 Dec 29;262:32–40. doi: 10.1016/j.jneumeth.2015.12.012

Controlled microfluidics to examine growth-factor induced migration of neural progenitors in the Drosophila visual system

Cade Beck a,1, Tanya Singh a,1, Angela Farooqi b, Tadmiri Venkatesh b, Maribel Vazquez a,*
PMCID: PMC4775339  NIHMSID: NIHMS755527  PMID: 26738658

Abstract

Background

The developing visual system in Drosophila melanogaster provides an excellent model with which to examine the effects of changing microenvironments on neural cell migration via microfluidics, because the combined experimental system enables direct genetic manipulation, in vivo observation, and in vitro imaging of cells, post-embryo. Exogenous signaling from ligands such as fibroblast growth factor (FGF) is well-known to control glia differentiation, cell migration, and axonal wrapping central to vision.

New method

The current study employs a microfluidic device to examine how controlled concentration gradient fields of FGF are able to regulate the migration of vision-critical glia cells with and without cellular contact with neuronal progenitors.

Results

Our findings quantitatively illustrate a concentration-gradient dependent chemotaxis toward FGF, and further demonstrate that glia require collective and coordinated neuronal locomotion to achieve directionality, sustain motility, and propagate long cell distances in the visual system.

Comparison with existing method(s)

Conventional assays are unable to examine concentration- and gradient-dependent migration. Our data illustrate quantitative correlations between ligand concentration/gradient and glial cell distance traveled, independent or in contact with neurons.

Conclusions

Microfluidic systems in combination with a genetically-amenable experimental system empowers researchers to dissect the signaling pathways that underlie cellular migration during nervous system development. Our findings illustrate the need for coordinated neuron-glia migration in the Drosophila visual system, as only glia within heterogeneous populations exhibited increasing motility along distances that increased with increasing FGF concentration. Such coordinated migration and chemotactic dependence can be manipulated for potential therapeutic avenues for NS repair and/or disease treatment.

Keywords: Microfluidics, Drosophila, Glia, FGF, Visual system

1. Introduction

Neuronal-glia interactions are critical for the development and function of the nervous system (NS) of both vertebrates and invertebrates. Glia and neurons are well-known to migrate to specific target locations during NS development, but the molecular signals that orchestrate this migration remain incompletely understood. While concentration gradients of a variety of growth factors are well-known to determine the positioning and fate of embryonic stem cells (Gomez-Marin and Louis, 2012; Green and Smith, 1991; Kornberg and Guha, 2007), the gradients that mediate the coordinated migration of cells of neuronal-glial lineage are largely unexplored. An understanding of the mechanisms that regulate such coordinated neuronal-glia migration may provide insights into possible therapeutic approaches for the myriad of neurological diseases caused by aberrant migration (e.g. lissencephaly, polymicrogyria, gliomas), as well as advance potential regenerative strategies to aid in NS repair.

Neurological study has increasingly embraced microfluidics-based technologies, with numerous systems used to examine varied aspects of embryogenesis (Levario et al., 2013), neural regeneration and growth (Brunello et al., 2013) and intracellular synaptic activity (Chotard and Salecker, 2004; Kim et al., 2010; Tabata, 2015), among others (Cimetta and Vunjak-Novakovic, 2014; Kim et al., 2015). Previous work from our group has developed microdevices to analyze NS behaviors in response to controlled environments generated without the use of forced flow (Able et al., 2012; Kong et al., 2010; Mishra et al., 2015; Rico-Varela et al., 2015). Such a system is distinctive for study of NS cell migration, as the external pumps and micro-valving routinely used to stream biochemical agents in many existing devices are well-known to alter the responses of cells unaccustomed to physiological flow (Huh et al., 2013; Kim and Tarbell, 1994). The developing visual system in Drosophila melanogaster provides an excellent model with which to examine neuronal-glial migration via microfluidics, as the combined experimental system enables direct genetic manipulation, in vivo observation, and in vitro imaging of cells, post-embryo.

During the development of the Drosophila visual system, retinal basal glia (RBG) are generated in the developing brain and migrate into the developing eye imaginal disk. The RBG have been implicated in the targeting of the photoreceptor cell axons to the optic ganglia. While the genes involved and the cellular interactions regulating the development of the photoreceptor R-cells and RBG have been well studied for Drosophila, the environmental cues that initiate the coordinated and/or collective migration of these vision-critical cells remain incompletely understood (Choi and Benzer, 1994; Rangarajan et al., 1999). Genetic studies have illustrated that conserved signaling pathways, such as Notch-Delta, Epidermal growth factor receptor (EGFR) and Fibroblast Growth Factor Receptor (FGFR), are central to cell fate specification and pattern formation in the developing visual system (Huang and Stern, 2005; Shamloula et al., 2002). In particular, our group has demonstrated that Notch loss of function disrupts RBG migration (Shamloula et al., 2002), while exogenous EGF signaling has been shown to activate long-distance migration of retinal progenitor cells (Unachukwu et al., 2013). In addition, FGF ligands are known to control glia differentiation, migration, and axonal wrapping central to vision (Silies et al., 2010; Tulin and Stathopoulos, 2010a,b). FGF signaling molecules are believed to be produced at a localized source and then dispersed into targeted tissue. This dispersion occurs via the transient formation of concentration gradients that then dictate the expression of different genes to lead to tissue patterning and morphogenesis (Yu et al., 2009).

In the present study, we examined the extent to which FGF concentration profiles and gradients play a role in regulating critical neuronal-glia migration during development of the visual system. We employed microfluidics to investigate the chemotaxis of RBG cells derived from the third larval instar stage of Drosophila melanogaster. Our results are among the first to illustrate that RBG exhibit concentration-gradient dependent chemotaxis to FGF signaling, and more significantly, these glial cells require coordinated and/or collective neuronal locomotion to achieve directionality, sustain motility and advance large migration distances. These results highlight the importance of coordinated migration between neuronal-glial cells in the developing visual system, as well as reinforce the advantages of microfluidics-based systems for the study of integrated neural cell response.

2. Materials and methods

2.1. Drosophila stock, neural cell dissociation, culture and viability

Drosophila were cultured on the standard corn meal agar medium and reared at 23 °C. To specifically label and mark glia with GFP we used the GAL4-UAS system (Brand and Perrimon, 1994). Briefly, repo-Gal4/Tb flies were crossed to UAS-GFP flies and the larvae form the progeny of the genotype repo-Gal4; UAS-GFP was used for all series of experiments in this study. Flies of this stock express GFP+ glial cells that then fluoresce green in the visual system. Flies were mated for 3 days, and subsequent F1 offspring produced 12 days later were dissected in the 3rd instar larvae stage. The procedure for culturing glia was adapted from a widely-accepted method established to culture neural cells from Drosophila larvae (Crank, 1979). In brief, third instar larvae were surface sterilized with 70% ethanol and placed in PBS (Mediatech Inc., Herndon, VA). Brain complexes were then dissected out using fine #5 stainless steel tweezers and placed in a solution of 0.5 mg/mL collagenase (Gibco, Grand Island, NY) in phosphate buffered saline (PBS). Brain tissue was then finely chopped, covered and incubated at 25 °C room temperature for 1 h with slight manual agitation. The cell suspension was then centrifuged and re-suspended in culture medium consisting of 90% revised Schneider’s Drosophila medium (Lonza, Walkersville, MD) and 10% heat-inactivated FBS (Mediatech Inc., Manassas, VA) with 50 mg/mL penicillin (Mediatech Inc.) and 50 μg/mL streptomycin sulfate (Mediatech Inc.), as shown in Fig. 1. Cells were incubated at room temperature at atmospheric CO2 levels. The brain complexes of 50–75 files were used per experimental condition. Cells were cultured for 5–6 days and examined via MTT assay to verify >93% viability prior to experiments.

Fig. 1.

Fig. 1

Open in a new tab

Derivation of Drosophila-derived neural cell populations. (A) Illustration of GAL4-UAS system. Repo-Gal4 flies are mated with UAS-GFP. Offspring express GFP in glia only. Drosophila were bred and grown in house and brain complexes of those in the larva stage were extracted for cell testing. (B) Confocal image of Drosophila whole brain complex, confirming presence of GFP+ glial cells.

2.2. Fluorescent activated cell sorting (FACS)

Experiments were performed using glia from two groups of cells extracted from the drosophila NS and sorted via FACS: (A) A heterogeneous neural cell population derived from dissociated brain complexes, which contained cells of both neural and glial lineages; and (B) A homogenous population of GFP+ cells of glial lineage, only. FACS was performed using a BD FACS Aria II (BD Biosciences, San Jose CA) with accompanying FACSDiva 6.1.3 software (BD Biosciences, San Jose CA). A 488-nm wavelength laser through a 530/30 filter was used to detect fluorescent signature of GFP (541 nm). The 633 nm laser through a 660/20 filter was used to detect To-pro-3 Iodide. 10 μL of To-pro-3 Iodide (Invitrogen, Eugene OR) at 25 μg/mL was added to the sample and incubated for 15 min.

2.3. μLane microfluidic system

Experiments used our established microfluidic system, called the bridged μLane, whose design and operation has been described previously in greater detail (Able et al., 2012; Dudu et al., 2012; Kong et al., 2010; Rico-Varela et al., 2015). As shown in Fig. 2, the system is fabricated via elastomeric molding of two layers of conventionally-used Poly dimethyl siloxane (PDMS,) that are bonded to one another, and to a boron silicate glass microscope slide. The first layer includes a 150-μm diameter by 15-mm long microchannel, as well as a source and sink reservoir, which have volumes of 9-μL each. The second layer consists of 170-μL source and sink chambers, and a larger hemispherical bridge channel connecting them. During operation, the biochemical agent of interest, in this case FGF, is inserted into the source chamber, while the remainder of the system is filled with PBS or media. The bridge channel is filled with media last, as its function is to precisely balances the liquid levels in the source and sink chambers to minimize the hydrostatic pressure difference between them (Kong et al., 2010). Transport of FGF via so-called convective diffusion (Bird and Hassager, 1987; Crank, 1979; Crowe et al., 2000) then establishes a concentration gradient between the source and the sink reservoir that spans several orders of magnitude. This distribution of FGF concentration can be maintained for several days and is modeled computationally and verified experimentally. Concentrations of 10-ng/mL, 100-ng/mL, and 1000-ng/mL of FGF were used to generate concentration profiles for testing of cell migration within the μLane. For control experiments, media was added into the μLane system in lieu of FGF.

Fig. 2.

Fig. 2

Open in a new tab

Description of Bridged μLane System. (A) Schematic of microfluidic system used in the current study. The reservoirs hold a 9 μl volume and both chambers hold a 170 μl volume each. Images of μLane system. (B) First layer of PDMS: Two reservoirs connected by a 15-mm-long microchannel. (C) Second layer of PDMS: Sink and source chambers connected by a bridged channel. (D) GFP+ glial cells within the microchannel at 0, 24, and 48 h, respectively. (E) FGF distribution within laminin-coated μLane system, predicted analytically and measured experimentally. Experimental data confirms that a steady-state FGF distribution occurs within the entire length of the channel before 24 h.

To mimic concentrated biological ECM microenvironments in the NS (Branco et al., 2009; Brandl et al., 2010; Singh et al., 2014; Taylor-Weiner et al., 2013), μLane systems were filled with 50 μg/mL of Laminin (BD Biosciences, Bedford, MA, Cat. No. 354232) and incubated at 37 °C for 1 h. The source of laminin is from Engelbreth-Holm-Swarm murine sarcoma basement membrane. Broadie et al. describe the breakdown of the Drosophila ECM molecules specifically two well conserved laminins, Laminin A and a minor Laminin W both sharing common β and γ chains with distinct α, with different tissue distributions (Broadie et al., 2011). Laminin was allowed to gel within the μLane for 1 h at room temperature to generate 2D microenvironments. Laminin was then aspirated from the channel interstitial spaces leaving the protein adhered to the channel surfaces, only. During experiments, the microchannel, sink reservoir, sink chamber, and bridge channel were first filled with PBS solution. As performed previously by our group, fluorescently-labeled FITC-Dextran (20 kDa, Sigma–Aldrich, St. Louis, MO, Cat. No. FD20S-100MG) was used to model FGF-8 due to similar molecular weight. For 2D experiments, Dextran at a concentration of 40 μg/mL was added drop wise to the source chamber until the reagent made contact with the PBS in the bridge channel, initiating the experimental system.

Two-dimensional numerical simulations of the bridged μLane system were performed in order to model the transport within the entire microsystem, i.e. the microchannel, both the SRR and SKR reservoirs in the 1st layer PDMS, the bridge channel, and both the SRC and SKC chambers in the 2nd layer PDMS. The mass transport within the entire microsystem was modeled using the 2-D continuity equation, convective-diffusion equation, momentum equation, and hydrostatic equation, as described previously in detail by our group. Free solution diffusivities and effective diffusivities of growth factor proteins were determined in the microsystem by modeling the transport within the system microchannel using convective-diffusion, shown below (Yuan et al., 1991):

∂C∂t+VB∂C∂x=D∂2C∂x2 (1)

where C (μg/mL) is concentration, t (s) is time, D (m2/s) is diffusivity, VB (m/s) is bulk velocity, and x (m) is position. The time-evolving solution to this equation was modeled computationally via finite-element-analysis (FEM) in Matlab 7.7 (MathWorks, Natick, MA). The boundary conditions fixed the source reservoir (X0 = 0) at 40 μg/mL and the sink reservoir (XL = 15 mm) at 0 μg/mL for all experimental times. The system initial condition was set such that the concentration along the entire length of the microchannel was 0 μg/mL at t = 0. In order to determine bulk flow velocity, VB, 1.9-μm-diameter fluorescent beads (Duke Scientific, Palo Alto, CA, Cat. No. G0200) were injected in the system and visualized via fluorescence microscopy, as done previously. Note that Equation 1 reduces to Fickian diffusion in the absence of convective flow.

2.4. Well plates

Experiments utilized 96 well plates (Corning Inc., Corning, NY) coated with Laminin as described above, and 100-μL of cell solutions per well to image migration of both cell populations to defined concentrations of FGF. Concentrations of 1-ng/mL, 10-ng/mL, and 100-ng/mL of FGF were used for testing, while no FGF was used for controls.

2.5. Microscopy and imaging

Dissected eye brain complexes from larvae were cleaned and mounted with Vectashield medium (Vector Laboratories, Burlingame, CA). Images were taken with a Zeiss CLSM-510 (Zeiss, Jena, Germany) and a 40× objective. An argon laser with excitation of 488 nm was used activate the GFP+ cells. Images were obtained as Z-stacks and processed with LSM software, which converted the Z stacks to a projection. Imaging within the μLane system was performed using a Nikon TE2000 inverted microscope with a 20x objective and a cooled CCD camera (CoolSNAP EZ, Photometrics, Tucson, AZ) with Nikon software (Nikon Instrument Element 2.30 with 6D module, Morrell Instrument Company Inc., Melville, NY).

Fluorescent images of the μLane were taken every hour for 72 h at the points x = 4 mm, x = 8 mm, and x = 12 mm, where the origin was defined as the point where the source reservoir abutted the microchannel. Imaging was performed using a Nikon TE2000 inverted microscope with a 20× objective and a cooled CCD camera (CoolSNAP EZ, Photometrics, Tucson, AZ) with Nikon software (Nikon Instrument Element 2.30 with 6D module, Morrell Instrument Company Inc., Melville, NY).

2.6. Data analysis and statistics

Data were analyzed using the Chemotaxis and Migration Tool (Ibidi, Verona, WI) in ImageJ (NIH). Cells were individually tracked using the manual tracking plug-in. Cell trajectories were plotted, along with accumulated distance and motility. In addition, the Chemotactic index, CI, was used as the measure of directed cell migration, as defined in Eq. (2) (Kong et al., 2011):

CI=xd (2)

where x is distance toward gradient and d is total accumulated distance. Values approach 1 as cells move more directly toward the gradient, are negative for cells moving away from the gradient. CI approaches zero in the absence of a gradient.

2.7. Statistical analysis

Statistical significance between groups was evaluated using a student’s t-test. In the gradient experiments, the Raleigh test was used to assess uniformity of the cell distribution.

3. Results

This study examined how the chemical cues from the microenvironment influence coordinated migration of cells of glial and neuronal lineages. In general, during Drosophila neural development glia have been shown to provide both attractive and repulsive guidance cues to NS neurons as well as require their associated axons for trophic support. Parker and Auld (Parker and Auld, 2006) have shown specifically that the midline glia are crucial for the proper formation of horizontal commissural axon tracts in the Drosophila CNS. In a review by Lemke (Lemke, 2001), reciprocal interactions between differentiating glial cells and neurons define the course of nervous system development even before the point at which these two cell types become definitively recognizable. Glial cells control the survival of associated neurons in both Drosophila and mammals, but this control is dependent on the prior neuronal triggering of glial cell fate commitment and trophic factor expression. The first set of experiments used FACS to generate two groups of cells from the drosophila CNS: (A) A heterogeneous population derived from dissociated brain complexes, which contained cells of both neural and glial lineages; and (B) A homogenous population of GFP+ cells of glial lineage, only. As seen in Fig. 3A, Drosophila brain complexes are imaged after utilizing the Gap4-Repo system to genetically introduce GFP+ glia into live Drosophila. As shown, glia are seen to migrate from the optic stalk (~250 μm in cross-section) to prescribed locations in the eye upon being ushered by concentration gradients of key growth factors (Silies et al., 2010). These eye complexes were then removed and FACS sorted for GFP+ glia, as seen in Fig. 3B. Here, analysis of FACS revealed less than 10% of the viable cells derived from larvae NS presented as GFP+ glia, requiring increased sample sizes to insure that both the heterogeneous and homogeneous populations retained recovery of 105 cells for each experimental test. Furthermore, viability tests were performed on all FACS sorted populations to insure that live cells (>90%) were used for migration testing.

Fig. 3.

Fig. 3

Open in a new tab

A Derivation of Drosophila-derived neural cell populations. Drosophila larvae. Panel (A) shows an early third instar developing eye imaginal disk attached to the brain lobe. Arrow points to the glia migrating from the brain through the optic stalk into the developing eye. Panel (B) shows a late third instar eye imaginal disk with glia migration in to the eye disk complete. The glia migrate and stop at the Morphogenetic Furrow (MF) [Arrow]. Ahead of the MF dividing undifferentiated cells can be seen labeled with the mitotic marker, anti-Phospho-Histone 3 antibody. The developing photoreceptors are not labeled. (B) Selection of homogeneous and heterogeneous neural cell samples. FACS analysis of cells derived from brain tissue of third instar larvae drosophila. (A) A four quadrant distribution (Q1–Q4) of cells stained with a To-pro-3 Iodide as a viability marker (APC-A) against a GFP signature (FITC-A). Cells in Q3 and Q4 are considered viable cells, while cells in Q2 and Q4 are denoted as GFP positive. Points in Q4 represent cells that are both alive and GFP positive. (B) A representative FACS sorting for cells that are GFP positive/negative only. Points denoted in P3 denote cells that are GFP+.

Once cell samples were sorted into two populations, their migratory responses to exogenous FGF signaling at different concentrations were examined using 96 well plates. As seen in Fig. 4, glia from homogeneous populations migrated less than 20 μm in response to increasing FGF concentrations of 1-, 10-, and 100-ng/mL. By contrast, glia of heterogeneous populations migrated steadily larger distances of up to 3-fold higher than that of homogeneous glia when exposed to the same FGF concentration fields. Furthermore, Fig. 4 illustrates that migration of glia from both populations was non-directional and radially semi-uniform within each plate at all concentrations examined. In addition, experiments measured the average accumulated distances traveled and average motilities of glia from both the heterogeneous and homogeneous populations when exposed to different concentrations of FGF, shown in Fig. 5. As seen, glia from heterogeneous populations migrated statistically-significant larger distances in response to increasing FGF concentrations, and demonstrated increased average motilities in response to signaling from increasing concentrations of FGF. By contrast, populations of homogeneous glia illustrated no statistically-significant differences in migration distances or average motilities in response to signaling from increasing FGF concentrations.

Fig. 4.

Fig. 4

Open in a new tab

Trajectories of homogeneous (sorted) and heterogeneous (unsorted) neural cell samples in response to FGF signaling. Plots illustrate representative trajectories from cells of homogenous population exposed to (A1) media only (control), (A2) 1-ng/ml (A2) 10-ng/ml and (A3) 100 ng/ml of FGF. Plots of representative trajectories of cells from heterogeneous populations of (B1) media only (control), (B2) 1-ng/ml (B2) 10-ng/ml and (A3) 100 ng/ml of FGF. Each plot contains 10 paths, and each path contains 1 data point every 2 h for 48 h. Concentrations shown are uniform; 10 cells were tracked for each concentration.

Fig. 5.

Fig. 5

Open in a new tab

Accumulated distances and motilities of heterogeneous and homogeneous neural cell populations. (A) Average accumulated distance traveled and (B) average motility of cells from both populations. Values reported are mean with error bars (+/−) standard deviation. *p < 0.05 compared to the control.

The final set of experiments used the μLane system to examine migratory responses from glia of both homogenous and heterogeneous cell populations to signaling from defined concentration gradients of FGF. Here, the molecular transport within the microchannel was, first, validated experimentally to characterize the concentration distribution within the system. Bulk flow within the system was examined using fluorescent beads with and without the use of the bridge channel, as described previously (Kong et al., 2010). Beads were seen to migrate with different velocities in the microsystem with and without the bridge channel. An average bulk velocity of Vavg = 92.3 ± 1.4 μm/s was measured within the μLane system in the absence of the bridge channel (n = 12 channels), but an average velocity of Vavg = 0.27 ± 0.06 μm/s was measured within the bridged μLane system (n = 16 channels). In addition, no bulk flow was measured within the bridge channel when beads were used, confirming that transport within the bridge was diffusion-dominated and did not significantly affect transport within the microchannel nor alter concentrations of the large volume SRC and SKC chambers over experimental time scales of up to 72 h.

The spatial and temporal patterns of FGF profiles generated within this system were rigorously analyzed by solving the convective–diffusion flux between the system reservoirs and channels, exactly. As seen in Fig. 2, the μLane was able to generate multiple concentration gradient fields within one experimental setting. The spatial distribution of ligand within the microsystem generated a wide range of gradients along the 12-mm-long channel length that were sustained for 5–7 days. Here, some gradient fields were generated via small changes in absolute concentration of attractant, and were therefore shallow and approximately linear, while other gradient fields (Rico-Varela et al., 2015) were created via much greater changes in attractant concentration, and were thus steeper and highly non-linear. Lastly, as seen in Fig. 2, the experimentally-obtained data were within 2% of the analytically-predicted value.

With a validated μLane system, heterogeneous and homogenous populations of neural-derived cells were then inserted into the μLane to examine glia viability, morphology and migration in response to different concentration gradients of FGF. Raw data images illustrate that glia do not modify their rounded morphology within any portion of the channel when exposed to varied FGF concentrations and gradients. Further, continued GFP expression from glia denoted that cells remained viable throughout the experimental test period of 72 h. Fig. 6 illustrates sample trajectories of glia derived from both homogenous and heterogeneous populations in response to signaling from different concentration gradient fields of FGF. As seen, glia from homogeneous samples did not exhibit statistically-significant differences in accumulated distance or motility, with values of each remaining approximately constant for the different FGF profiles studied. By contrast, the trajectories of glia from heterogeneous populations illustrate that cells increased their migration distance up to 6-fold when responding to concentration gradients of increasing FGF concentration. Further, glia from heterogeneous populations illustrated optimal CI values at gradient fields of intermediate FGF concentration of 10-ng/ml. Lastly, average motility of these cells was seen to increase in a statistically significant manner in response to gradient fields of increasing FGF concentration.

Fig. 6.

Fig. 6

Open in a new tab

Migration of heterogeneous cell populations to concentration gradients of FGF. (A) Plot of cell trajectories normalized to the origin. Each plot contains 10 trajectories generated using data points gathered every 2 h. Gradient increases from left to right as indicated by the arrows. Each line on the grid is 20 μm. (B) Values of chemotactic index for all experimental conditions. (C) Values of average motility for all experimental conditions. Reported values are mean data with standard deviation error bars. An * indicates p < 0.05 compared to control.

4. Discussion

Coordinated neuronal-glia interactions are known to be critical for the development, function and response of the visual system. Yet, how the collective but finely-tuned migration of these cell populations is regulated by their microenvironment remains incompletely understood. Results of the current study illustrated distinct migratory behaviors between homogenous populations of glia, and glia from heterogeneous populations containing cells of both glial and neuronal lineages. Using conventional well technology, glia of the homogeneous populations were seen to migrate radially, and with constant motility and distance independent of the FGF concentration used. By contrast, glia within heterogeneous populations exhibited increasing motility along migration distances that increased with increasing FGF concentration. We propose that these differences can be attributed to glia-neuronal interactions that are absent in the homogeneous glial population. We note that while FACS is known to influence cell viability (Arun et al., 2005; Freer and Rindi, 2013; Herzenberg et al., 2002) glia from both populations expressed comparable intensity levels of GFP and their migration remained constant over multiple experiments for several days. Our migration data is significant because RBG glia were initially only believed able to migrate into the eye disk via physical mechanisms of haptotaxis, i.e. crawling along a substrate (Haeger et al., 2015; Ricoult et al., 2015). However, more recent work suggested abilities of RBG to migrate toward differentiating photoreceptors without using a continuous physical substratum (Rangarajan et al., 1999). The data from our study support this later phenomenon, which point to short-range diffusible signals as regulators of RBG-neuron interactions, in addition to hypothesized contact-mediated processes that require physical interactions between cell types. Such neuronal-glia coordinated migration has been previously unstudied, and points to the importance of examining so-called collective migration (Kopf, 2015; Kumar et al., 2015; Pocha and Montell, 2014; Riahi et al., 2015; Stonko et al., 2015) for potential therapeutic avenues for NS repair and/or disease treatment.

We next utilized a microfluidics system, the μLane, to examine the migration of RBG in the presence of defined concentration gradients of FGF. While traditional benchtop technologies such as boyden chambers and agarose assays have contributed greatly toward identifying combinations of cytokines that generate directed cell movement (Reviewed in Hulkower and Herber, 2011; Kramer et al., 2013; Okada, 2012; Sackmann et al., 2014), these systems cannot be used to measure mechanistic parameters of migration, such as motility and directionality. By contrast, our quantitative and tunable μLane system empowers researchers to examine the migration of cell populations as a function of the ligand(s), concentration(s), and concentration gradient field(s) used. While it is well-established that no in vitro system can physiologically replace the in vivo environment, microfluidics offer unique advantages of biological scale that enable closer approximation to in vivo conditions. For instance, the μLane device creates a microenvironment for RBG that is 100 μm in diameter, which approaches dimensions of the in vivo optic stalk (Edwards et al., 2012). This creates an experimental test system where cells experience diffusible signals, matrix interactions, and cell to cell contacts on physiological scales. Thus, in this work, we used FGF as a diffusible signal because of its well-established role in signaling transduction critical to eye development (Silies et al., 2010), laminin as ECM because it is ubiquitous in the developing visual system of Drosophila (Singh et al., 2014), and neuronal-glia cell densities that approach physiological numbers based on in vivo spacing and dimensions (Li et al., 2015).

Using FGF as a model cytokine in this work, the μLane was able to generate multiple concentration gradient fields within one experimental setting that were sustained for 5–7 days. This exploits the analytical nature of microfluidic devices as the μLane can examine which ranges of concentration and gradient (i.e. concentration distribution over specified distances) stimulate migratory responses from cells. Such a tunable system facilitated the experimental tracking of 25–50 distinct cells for each FGF gradient field. Three different concentrations of FGF were used to generate gradient fields within the μLane. Glia of the homogenous population were not observed to migrate appreciable distances (less than 1 cell diameter) in our system, although cells remained viable within the device for several days as evidenced by their GFP expression and unchanged morphology. Amazingly, glial cells of the heterogeneous population were seen to migrate in highly directed trajectories along increasing FGF gradient, and with increasing motility. These results point to an obvious chemotactic dependence on FGF concentration gradient, as cell trajectories and distances traveled were much larger for cells exposed to signaling from FGF gradient fields than from concentration alone. Our results further underscore the presence and significance of coordinated neuronal-glia migration in the development of the Drosophila visual system. Cells of the homogeneous glial population did not exhibit measurable movement, while glia of the heterogeneous population illustrated directed migration within gradient fields, as well as sensitivity to specific FGF concentration. These data underscore the impact of the coordinated migration of neuronal-glia cells requisite for development of the visual system.

5. Conclusion

Our studies presented here illustrate the efficacy and power of microfluidic systems in combination with a genetically amenable experimental system to dissect the signaling pathways that underlie cellular migration during nervous system development. Taken together, these findings illustrate the significance of the coordinated neuron-glia interaction for migration in the development of the Drosophila visual system. Our results show that glia within heterogeneous populations exhibited increasing motility along migration distances that increased with increasing FGF concentration. Such coordinated migration and chemotactic dependence of these cells can be manipulated for potential therapeutic avenues for NS repair and/or disease treatment.

HIGHLIGHTS.

  • A microfluidic system examines migration of neural progenitors from visual system.

  • Glia chemotaxis is measured to defined concentration gradients of FGF.

  • Data suggest collective glial-neuronal migration mechanisms towards FGF signaling.

  • Chemotactic dependence evoke potential therapeutic avenues of visual system repair.

Acknowledgments

This work was supported by the National Science Foundation (CBET 0939511), a RCMI grant from the NIH (8G12MD007603-29) and Professional Staff Congress of CUNY PSCREG-36-893.

References

  1. Able RA, Jr, Ngnabeuye C, Beck C, Holland EC, Vazquez M. Low concentration microenvironments enhance the migration of neonatal cells of glial lineage. Cell Mol Bioeng. 2012;5 doi: 10.1007/s12195-012-0226-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arun KH, Kaul CL, Ramarao P. Green fluorescent proteins in receptor research: an emerging tool for drug discovery. J Pharm Toxicol Methods. 2005;51:1–23. doi: 10.1016/j.vascn.2004.07.006. [DOI] [PubMed] [Google Scholar]
  3. Bird RB, Hassager O. Dynamics of polymeric liquids: fluid mechanics. Wiley; 1987. [Google Scholar]
  4. Branco MC, Pochan DJ, Wagner NJ, Schneider JP. Macromolecular diffusion and release from self-assembled beta-hairpin peptide hydrogels. Biomaterials. 2009;30:1339–47. doi: 10.1016/j.biomaterials.2008.11.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brand AH, Perrimon N. Raf acts downstream of the EGF receptor to determine dorsoventral polarity during Drosophila oogenesis. Genes Dev. 1994;8:629–39. doi: 10.1101/gad.8.5.629. [DOI] [PubMed] [Google Scholar]
  6. Brandl F, Kastner F, Gschwind RM, Blunk T, Tessmar J, Gopferich A. Hydrogel-based drug delivery systems: comparison of drug diffusivity and release kinetics. J Control Release. 2010;142:221–8. doi: 10.1016/j.jconrel.2009.10.030. [DOI] [PubMed] [Google Scholar]
  7. Broadie K, Baumgartner S, Prokop A. Extracellular matrix and its receptors in Drosophila neural development. Dev Neurobiol. 2011;71:1102–30. doi: 10.1002/dneu.20935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Brunello CA, Jokinen V, Sakha P, Terazono H, Nomura F, Kaneko T, et al. Microtechnologies to fuel neurobiological research with nanometer precision. J Nanobiotechnol. 2013;11:11. doi: 10.1186/1477-3155-11-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Choi KW, Benzer S. Migration of glia along photoreceptor axons in the developing Drosophila eye. Neuron. 1994;12:423–31. doi: 10.1016/0896-6273(94)90282-8. [DOI] [PubMed] [Google Scholar]
  10. Chotard C, Salecker I. Neurons and glia: team players in axon guidance. Trends Neurosci. 2004;27:655–61. doi: 10.1016/j.tins.2004.09.001. [DOI] [PubMed] [Google Scholar]
  11. Cimetta E, Vunjak-Novakovic G. Microscale technologies for regulating human stem cell differentiation. Exp Biol Med. 2014;239:1255–63. doi: 10.1177/1535370214530369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Crank J. The mathematics of diffusion. Clarendon Press; 1979. [Google Scholar]
  13. Crowe CT, Elger DF, Roberson JA. Engineering fluid mechanics. Wiley; 2000. [Google Scholar]
  14. Dudu V, Able RA, Jr, Rotari V, Kong Q, Vazquez M. Role of epidermal growth factor-triggered PI3K/Akt signaling in the migration of medulloblastoma-derived cells. Cell Mol Bioeng. 2012;5:402–13. doi: 10.1007/s12195-012-0253-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Edwards TN, Nuschke AC, Nern A, Meinertzhagen IA. Organization and metamorphosis of glia in the Drosophila visual system. J Comp Neurol. 2012;520:2067–85. doi: 10.1002/cne.23071. [DOI] [PubMed] [Google Scholar]
  16. Freer G, Rindi L. Intracellular cytokine detection by fluorescence-activated flow cytometry: basic principles and recent advances. Methods. 2013;61:30–8. doi: 10.1016/j.ymeth.2013.03.035. [DOI] [PubMed] [Google Scholar]
  17. Gomez-Marin A, Louis M. Active sensation during orientation behavior in the Drosophila larva: more sense than luck. Curr Opin Neurobiol. 2012;22:208–15. doi: 10.1016/j.conb.2011.11.008. [DOI] [PubMed] [Google Scholar]
  18. Green JB, Smith JC. Growth factors as morphogens: do gradients and thresholds establish body plan. Trends Genet: TIG. 1991;7:245–50. doi: 10.1016/0168-9525(91)90323-I. [DOI] [PubMed] [Google Scholar]
  19. Haeger A, Wolf K, Zegers MM, Friedl P. Collective cell migration: guidance principles and hierarchies. Trends Cell Biol. 2015 doi: 10.1016/j.tcb.2015.06.003. [DOI] [PubMed] [Google Scholar]
  20. Herzenberg LA, Parks D, Sahaf B, Perez O, Roederer M, Herzenberg LA. The history and future of the fluorescence activated cell sorter and flow cytometry: a view from Stanford. Clin Chem. 2002;48:1819–27. [PubMed] [Google Scholar]
  21. Huang P, Stern MJ. FGF signaling in flies and worms: more and more relevant to vertebrate biology. Cytokine Growth Factor Rev. 2005;16:151–8. doi: 10.1016/j.cytogfr.2005.03.002. [DOI] [PubMed] [Google Scholar]
  22. Huh D, Kim HJ, Fraser JP, Shea DE, Khan M, Bahinski A, et al. Microfabrication of human organs-on-chips. Nat Protoc. 2013;8:2135–57. doi: 10.1038/nprot.2013.137. [DOI] [PubMed] [Google Scholar]
  23. Hulkower KI, Herber RL. Cell migration and invasion assays as tools for drug discovery. Pharmaceutics. 2011;3:107–24. doi: 10.3390/pharmaceutics3010107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kim MS, Bhang SH, Yang HS, Rim NG, Jun I, Kim SI, et al. Development of functional fibrous matrices for the controlled release of basic fibroblast growth factor to improve therapeutic angiogenesis. Tissue Eng Part A. 2010;16:2999–3010. doi: 10.1089/ten.TEA.2009.0828. [DOI] [PubMed] [Google Scholar]
  25. Kim SH, Lee GH, Park JY, Lee SH. Microplatforms for gradient field generation of various properties and biological applications. J Lab Autom. 2015;20:82–95. doi: 10.1177/2211068214562247. [DOI] [PubMed] [Google Scholar]
  26. Kim WS, Tarbell JM. Macromolecular transport through the deformable porous media of an artery wall. J Biomech Eng. 1994;116:156–63. doi: 10.1115/1.2895714. [DOI] [PubMed] [Google Scholar]
  27. Kong Q, Able RA, Jr, Dudu V, Vazquez M. A microfluidic device to establish concentration gradients using reagent density differences. J Biomech Eng. 2010;132:121012. doi: 10.1115/1.4002797. [DOI] [PubMed] [Google Scholar]
  28. Kong Q, Majeska RJ, Vazquez M. Migration of connective tissue-derived cells is mediated by ultra-low concentration gradient fields of EGF. Exp Cell Res. 2011;317:1491–502. doi: 10.1016/j.yexcr.2011.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kopf MH. Collective cell migration induced by mechanical stress and substrate adhesiveness. Phys Rev E Stat Nonlinear Soft Matter Phys. 2015;91:012712. doi: 10.1103/PhysRevE.91.012712. [DOI] [PubMed] [Google Scholar]
  30. Kornberg TB, Guha A. Understanding morphogen gradients: a problem of dispersion and containment. Curr Opin Genet Dev. 2007;17:264–71. doi: 10.1016/j.gde.2007.05.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kramer N, Walzl A, Unger C, Rosner M, Krupitza G, Hengstschlager M, et al. In vitro cell migration and invasion assays. Mutat Res. 2013;752:10–24. doi: 10.1016/j.mrrev.2012.08.001. [DOI] [PubMed] [Google Scholar]
  32. Kumar A, Gupta T, Berzsenyi S, Giangrande A. N-cadherin negatively regulates collective Drosophila glial migration through actin cytoskeleton remodeling. J Cell Sci. 2015;128:900–12. doi: 10.1242/jcs.157974. [DOI] [PubMed] [Google Scholar]
  33. Lemke G. Glial control of neuronal development. Annu Rev Neurosci. 2001;24:87–105. doi: 10.1146/annurev.neuro.24.1.87. [DOI] [PubMed] [Google Scholar]
  34. Levario TJ, Zhan M, Lim B, Shvartsman SY, Lu H. Microfluidic trap array for massively parallel imaging of Drosophila embryos. Nat Protoc. 2013;8:721–36. doi: 10.1038/nprot.2013.034. [DOI] [PubMed] [Google Scholar]
  35. Li S, Sukeena JM, Simmons AB, Hansen EJ, Nuhn RE, Samuels IS, et al. DSCAM promotes refinement in the mouse retina through cell death and restriction of exploring dendrites. J Neurosci: Off J Soc Neurosci. 2015;35:5640–54. doi: 10.1523/JNEUROSCI.2202-14.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Mishra S, Thakur A, Redenti S, Vazquez M. A model microfluidics-based system for the human and mouse retina. Biomed Microdevices. 2015;17:107. doi: 10.1007/s10544-015-0002-6. [DOI] [PubMed] [Google Scholar]
  37. Okada Y. Transcriptional regulation of the endothelial cell-specific receptor, Robo4. Yakugaku Zasshi: J Pharm Soc Japan. 2012;132:1045–9. doi: 10.1248/yakushi.132.1045. [DOI] [PubMed] [Google Scholar]
  38. Parker RJ, Auld VJ. Roles of glia in the Drosophila nervous system. Semin Cell Dev Biol. 2006;17:66–77. doi: 10.1016/j.semcdb.2005.11.012. [DOI] [PubMed] [Google Scholar]
  39. Pocha SM, Montell DJ. Cellular and molecular mechanisms of single and collective cell migrations in Drosophila: themes and variations. Annu Rev Genet. 2014;48:295–318. doi: 10.1146/annurev-genet-120213-092218. [DOI] [PubMed] [Google Scholar]
  40. Rangarajan R, Gong Q, Gaul U. Migration and function of glia in the developing Drosophila eye. Development. 1999;126:3285–92. doi: 10.1242/dev.126.15.3285. [DOI] [PubMed] [Google Scholar]
  41. Riahi R, Sun J, Wang S, Long M, Zhang DD, Wong PK. Notch1-Dll4 signalling and mechanical force regulate leader cell formation during collective cell migration. Nat Commun. 2015;6:6556. doi: 10.1038/ncomms7556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rico-Varela J, Singh T, McCutcheon S, Vazquez M. EGF as a new therapeutic target for medulloblastoma metastasis. Cell Mol Bioeng. 2015:1–13. doi: 10.1007/s12195-015-0395-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ricoult SG, Kennedy TE, Juncker D. Substrate-bound protein gradients to study haptotaxis. Front Bioeng Biotechnol. 2015;3:40. doi: 10.3389/fbioe.2015.00040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sackmann EK, Fulton AL, Beebe DJ. The present and future role of microfluidics in biomedical research. Nature. 2014;507:181–9. doi: 10.1038/nature13118. [DOI] [PubMed] [Google Scholar]
  45. Shamloula HK, Mbogho MP, Pimentel AC, Chrzanowska-Lightowlers ZM, Hyatt V, Okano H, et al. rugose (rg), a Drosophila A kinase anchor protein, is required for retinal pattern formation and interacts genetically with multiple signaling pathways. Genetics. 2002;161:693–710. doi: 10.1093/genetics/161.2.693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Silies M, Yuva-Aydemir Y, Franzdottir SR, Klambt C. The eye imaginal disc as a model to study the coordination of neuronal and glial development. Fly. 2010;4:71–9. doi: 10.4161/fly.4.1.11312. [DOI] [PubMed] [Google Scholar]
  47. Singh T, Kothapalli C, Varma D, Nicoll SB, Vazquez M. Carboxymethylcellulose hydrogels support central nervous system-derived tumor-cell chemotactic migration: comparison with conventional extracellular matrix macromolecules. J Biomater Appl. 2014;29:433–41. doi: 10.1177/0885328214532969. [DOI] [PubMed] [Google Scholar]
  48. Stonko DP, Manning L, Starz-Gaiano M, Peercy BE. A mathematical model of collective cell migration in a three-dimensional, heterogeneous environment. PLOS ONE. 2015;10:e0122799. doi: 10.1371/journal.pone.0122799. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tabata H. Diverse subtypes of astrocytes and their development during corticogenesis. Front Neurosci. 2015;9:114. doi: 10.3389/fnins.2015.00114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Taylor-Weiner H, Schwarzbauer JE, Engler AJ. Defined extracellular matrix components are necessary for definitive endoderm induction. Stem Cells. 2013 doi: 10.1002/stem.1453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Tulin S, Stathopoulos A. Analysis of Thisbe and Pyramus functional domains reveals evidence for cleavage of Drosophila FGFs. BMC Dev Biol. 2010a;10:83. doi: 10.1186/1471-213X-10-83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Tulin S, Stathopoulos A. Extending the family table: insights from beyond vertebrates into the regulation of embryonic development by FGFs. Birth Defects Res C Embryo Today. 2010b;90:214–27. doi: 10.1002/bdrc.20182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Unachukwu UJ, Sauane M, Vazquez M, Redenti S. Microfluidic generated EGF-gradients induce chemokinesis of transplantable retinal progenitor cells via the JAK/STAT and PI3kinase signaling pathways. PLoS One. 2013;8:e83906. doi: 10.1371/journal.pone.0083906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yu SR, Burkhardt M, Nowak M, Ries J, Petrasek Z, Scholpp S, et al. Fgf8 morphogen gradient forms by a source-sink mechanism with freely diffusing molecules. Nature. 2009;461:533–6. doi: 10.1038/nature08391. [DOI] [PubMed] [Google Scholar]
  55. Yuan F, Chien S, Weinbaum S. A new view of convective-diffusive transport processes in the arterial intima. J Biomech Eng. 1991;113:314–29. doi: 10.1115/1.2894890. [DOI] [PubMed] [Google Scholar]

RESOURCES