Supplemental Materials

Optical fiber light source directs neurite growth

Forrest Jesse^1*, Zhenjiang Miao¹, Li Zhao², Yao Chen², Yuan Yuan Lv²

¹Human Computer Interface Lab, Beijing Jiaotong University, Beijing, 100044, China

²Department of Physiology, Beijing Sports University, Beijing, 100084, China

*jesse@jesse.org

1. Fiber Tapers

The simplest implementation of the fiber taper is a relatively wide taper that is stiff enough to not bend under its own weight or be pushed by the liquid it is immersed in. The tapers we found most easy to work with were similar in width to a neurite. Without adding lenses or shaping the end of the taper, our wide (micrometer) taper will cause a wide beam, which in our experiments often encompassing the entire growth area of the neurite. Unless energy is closely regulated according to the emission tip’s distance from target, a wide beam can cause neurites to quickly (within 5 minutes) be thickened by too much radiation, causing growth to stop, perhaps permanently. Narrow beams can be accomplished by creating a narrow taper (which we created and used but found difficult to move), or by possibly shaping the end surface of the taper into a lens or adding other optics. Our experiment dealt with a simple fiber taper only and did not add optics to its end or shape the end other than as flat and as half-sphere.

A variety of fiber tapers with similar but varying optical characteristics were created. Best results were achieved with a fiber which tapered to 1 micrometer through a length of 1 cm, ending in a smooth, regular and un-chipped tip. 1 cm long uncurved tapers were found to be the best compromise between optical precision and fiber manipulability as light loss is minimized and short tapers bend less at the air-liquid interface when being partially immersed in the cell culture solution.

1.1 Taper Characteristics in Detail

Fiber tapers were analyzed under a microscope[j]. Tapers of two shapes were selected: a “narrow”class of tapers with a tip width of 0.1μm (as narrow as the cell neurites) and a “micrometer”class of tapers, with a tip width of 1μm (1x to 4x the width of the neurites). Viewing through a video microscope[k] allowed eye safe visual characterization of the emitted beam by projecting it through a cloudy solution. Narrow tapers were found to have lower divergence of 15 to 30 degrees because the emitting end was flatter. The narrow type also had higher light loss which could be linked to bending (up to 45 degrees) of the fiber at the air - liquid interface where the fiber entered the liquid in the sample dish. Micrometer tapers had a projected beam with a 45 to 120 degree divergence and less light loss, the fiber was unaffected by bending at the air-liquid interface. Beam divergence in culture liquid was low (less than 5 degrees) for fiber tips that were flat, and high (15 to 120 degrees) for rounded fiber tips. Micrometer tapers were then created with a divergence near 90 degrees.

We did not re-coat the exposed fiber taper area with a new reflection layer, which, considering that the refractive index of water¹⁰ (about 1.3) and that of fused silica¹¹ (about 1.5) are so close, care must be taken to keep the fiber straight as small bends will leak light. For this reason we chose to use thicker fiber tapers which were less prone to bending when unsupported.

We tried fiber tapers with many diversion angles between 90^o and 10^o, but the fiber tapers we used in this experiment had a diversion measured to be 90^o or 1/4^th of a sphere.

1.2 Measuring beam characteristics under the microscope

The refractive index of air⁹ is about 1 and the index of water¹⁰ is about 1.3, so beam divergence from the fiber tip measured in air is higher than in the culture liquid (which is mostly water) The refractive index of the silica fiber¹¹ is about 1.5. We measured beam divergence in culture solution by directing an LED laser emitting 808nm light into the fiber. The fiber was immersed in a cloudy culture solution. In a light-sealed laboratory operators wearing laser safety glasses powered on the laser and set an emitted power of 500mW. The beam from the fiber taper was projected in the cloudy solution and imaged by the video microscope which is sensitive to infrared light. The beam profile was measured from the microscope image to determine divergence (figure 5).

figure 1

The dark black line surrounding the white area represents edge detection of the beam, which can be compared to a 90^o angle line superimposed on the image, showing near 90^o divergence. Image contrast increased, cropped.

Fiber characteristics such as bends and imperfections in the fiber taper influence light loss through the fiber. We primarily used the wider and higher divergence micrometer class of tapers as this type had the least light loss as measured through the microscope. The rapid drop-off of energy with distance makes the ideal region of influence a narrow band between 1 and 5μm from the end of a taper with 90 degree divergence.

There is a change in the direction of extension of the growth edge toward the taper tip within 2 or 3 minutes when the taper is emitting about 135mW with a distance of between 1 and 5μm from the neurite growth edge.

The energy of the beam can be approximately modeled as the following relationship:

where divergence div is 360^o/90^o = 4/1 = 4; r is distance to a 1 μm² section of the target neurite. The model is diagramed in figure 4, showing v¹,v³,v⁵, as increments of distance from the source of radiation at origin o⁰. Radiation is about 161mW into the fiber and light loss is about 16%, so the imaginary sphere surface at v¹ has 135mW over a section with an area of about 3μm².

1.3 Beam directability and usability.

The narrow fiber tapers produced the least directable beam as a great percentage of light was lost before emitting through the taper, the fiber was narrow enough to be highly influenced by the air - liquid interface at the top of the sample dish, and was influenced by fluid currents in the dish and was subject to drag induced bending when moved unless supported externally. The large taper produced a well defined beam with higher divergence, and was easy to precisely position.

1.4 Fiber Taper Construction

A 2m multimode optical fiber[i] was cut in half, and the outside cover layers were stripped away with scissors. A 10cm section of the interior fiber with the inner glass, reflection layer, and two outer polymer protective layers was trimmed bare. The two outer protective layers were cut through with scissors to the reflection layer. The protective layer was carefully pulled to leave a gap of bare glass fiber between two protected sections. The fiber's two protected sections were gripped with two hemostat clamps and the bare section was heated over an alcohol flame while pulling the fiber firmly to stretch and taper the fiber, and finally sever the two protected sections. Varying heat, stretching speed, and tension allowed the production of tapers down to 0.1μm (and less) wide and with tips varying from flat (when the fiber cools more quickly) to semi-spherical (when the fiber cools more slowly).

2. Light power

The laser used in this expirament[l] is a 500mW 808nm diode laser, calibrated in 2011-5 and 2012-7, the laser diode is directed into the fiber through a two element lens. The fiber coupling is a FC connector for 150μm core (multimode) fiber.

We calibrated the light output from the end of the fiber taper using a 1 cm² silicon photodiode with a plastic attenuation filter covering the diode so that the diode has a linear[16] light-to-current response from 3mW to 150mW. The current generated from the photodiode receiving light from another new, recently power-calibrated known-power laser source (measured after the fiber) at 135mW was measured and then adjusted our expirament laser to reach 135mW after the fiber tip.

We built a calibration curve linking the control values for the laser power to fiber tip output power values. We then chose 135 mW as a permanent output value. All power variation was then managed as a distance value of fiber tip to neurite surface and laser output power was held constant. The laser was given a 20 minute warm up period.

Light loss at entry into the fiber and attenuation⁸ in the fiber is about 16%, 161mW output from the laser is about 135mW exiting from the fiber taper.

figure 3

Calibration profile (2011-5) for the laser used in the expirament[l], We calibrated an after-the fiber power for 135mW using the current from a photodiode and cross comparing with another known-power laser (again after the fiber).

This is a simple laser power meter. The particle content of the emitted light can be measured with a photocell, where each photon interacting with the cell will release a photoelectron. Minus losses of photons due to their not reaching or not interacting with the photodiode, we have electrons that are measured as a current from the cell.

Photon flux can be measured with a photodiode if adjustments are made for the particularities of the diode and its environment. 1mW = about 3.9 x 10¹⁵ photons/second (and photelectrons/second from the photodiode).

The photdiode will have a current response to light, which is linear in a range that varies with each diode[16]. Our diode’s linear response band is from 3 mW to 150mW after adding an attenuation filter. The response plateaued after about 170mW. Response varies with temperature, but this effect is far less than 1mW error per minute, our measurement was averaged over 1 minute.

The light output drops off quickly with the distance between the detector and the fiber tip, because the divergence of the light emitted from the fiber tip is about 90 degrees, so we used a large photodiode and placed the fiber tip close to it.

2. Cell Cultures

Two types of cells were used. Rat cortex neurons[a] were extracted from rat pups within 24 hours of birth, and PC12 cells[b] were cultured from frozen stored cells.

2.1 Rat cortex neurons:

2.1.1 Culture liquids:

Hepes buffered saline (“HBS”) containing 10 mM HEPES[c], 140 mM NaCl, 5 mM KCl, 2mM CaCl，1mM MgCl, 10 mM glucose[d], pH 7.3, was always filter-sterilized before use. Trypsin solution (bovine pancreas) [c] was dissolved in HBS to 1.25mg/ml plating and feeding media comprised of DMEM supplemented with high glucose[e], 10% fetal bovine serum (“FBS”)[f], penicillin (100 U/mL) and streptomycin (100 μg/mL)[e]. Glutamine[g] was added to the media on the day of use by diluting frozen stocks to a final concentration of 2 mM.

2.1.2 Culture dishes:

Plastic culture dishes[h] were coated with L-polyornithine[c] prior before use. After 2 hours of incubation, dishes were washed twice with DMEM before use.

2.1.3 Process:

Cortical tissue was isolated under sterile conditions, rinsed in HBS, the meninges were removed and cut into 0.5mm pieces. Tissues were placed over ice while being operated on.

The minced tissue (approximately 0.2 mL) was then treated with 4ml 0.125% pancreas solution for 15 min at 37°C with gentle inversion every 5 minutes. Tissue was then triturated with a 1.5ml pipet 15-20 times. A few drops of FBS were then added to stop digestion. Cells were centrifuged at a rate of 1000 rpm for 10 minutes before the pancreas solution was removed and replaced with 6 mL of pre-warmed plating medium. Cells were again dispersed and then counted and adjusted to a maximum of 3.5×105 cells/mL. 2 mL of this cell solution was added to each 35-mm L-polyornithine-coated dish.

Cells were then observed through a video microscope (Olympus, 10x ocular, 40x objective) and found to be round and floating in solution, with diameters of 2 to5μm. Cells were cultured in an incubator at 37°C in 5% CO² for 1 hour and then observed. Cells that were round and plated to the bottom of the sample dish were accepted for use and samples that did not appear healthy were not used.

2.2 PC12 cells

Feeding media comprised of DMEM supplemented with high glucose[e], 10% horse serum and 5% fetal calf serum [f], penicillin (100 U/mL) and streptomycin (100 μg/mL)[e].

Dish media for PC12 cells were same[h] as for rat cortical cells.

2.3 Further processes for both types of cells:

After 6 hours of incubation cell diameters were 5 to 10μm, and cells were observed to be growing neurites. After 12 hours of incubation, neurites had extended about 3 cell body lengths (15 to 30μm) away from the cell. Over a hundred samples were prepared, with about 10% of our sample dishes left unused because the cells did not appear healthy (healthy as indicated by cells plating to the bottom of the dish after incubation).

The experiment was conducted in sample dishes that were removed from incubation (37 degrees C, 5% CO2) and placed under a microscope in open air at about 27 degrees C. Sample dishes were plastic, coated with L-polyornithine which allows the cells to plate onto the surface, and becomes a uniform traction surface for neurites to physically interact with. Cells were examined to find growing neurites, cells were examined at intervals of 30 seconds to find neurites which had moving growth areas.

Additional references

1. Anders, J. J., Borke, R. C., Woolery, S. K. and van de Merwe, W. P. (1993), Low power laser irradiation alters the rate of regeneration of the rat facial nerve. Lasers in Surgery and Medicine, 13: 72–82. doi: 10.1002/lsm.1900130113

2. Higuchi, A., Kitamura, H., Shishimine, K., Konishi, S., Yoon, B.O., and Hara, M. Visible light is able to regulate neurite outgrowth. J Biomater Sci Polym Ed14, 1377, 2003.

3. Higuchi, A., Watanabe, T., Matsubara, Y., Matsuoka, Y., and Hayashi, S. Regulation of neurite outgrowth by intermittent irradiation of visible light. J Phys Chem B 109, 11033, 2005.

4. Higuchi, A., Watanabe, T., Noguchi, Y., Chang, Y., Chen, W.Y., and Matsuoka, Y. Visible light regulates neurite out-growth of nerve cells. Cytotechnology 54, 181, 2007.

5. Ehrlicher, A., Betz, T., Stuhrmann, B., Koch, D., Milner, V., Raizen, M.G., and Kas, J. Guiding neuronal growth with light. Proc Natl Acad Sci USA 99, 16024, 2002.

6. Wollman, Y., Rochkind, S., and Simantov , R. Low power laser irradiation enhances migration and neurite sprouting of cultured rat embryonal brain cells. Neurol Res 18, 467,1996.

7. Daniel Koch, Timo Betz, Allen Ehrlicher, Michael Gogler, Bjorn Stuhrmann, Josef Kas,

Optical control of neuronal growth. Proceedings of SPIE. Vol. SPIE-5514, pp. 428-436. 2004

8. Fiber attenuation data 2.75dB/Km attenuation (at 800nm) - (data from corning.com)

9. Philip E. Ciddor, Refractive index of air: new equations for the visible and near infrared. Applied Optics, Vol. 35, Issue 9, pp. 1566-1573 (1996)

10. P. Scheibener and J. Straub, J.M.H. Levelt Sengers and J.S. Gallagher, Refractive Index of Water and Steam as Function of Wavelength, Temperature and Density.

J. Phys. Chem. Ref. Data 19, 677-715 (1990)

11. Rei Kitamura, Laurent Pilon, and Miroslaw Jonasz, Optical constants of silica glass from extreme ultraviolet to far infrared at near room temperature.

Applied Optics, Vol. 46, Issue 33, pp. 8118-8133 (2007)

12. Harald Ries, Akiba Segal, and Jacob Karni, Extracting concentrated guided light. Appl. Opt. 36, 2869-2874 (1997)

13. Catherine E. Graves, Ryan G. McAllister, William J. Rosoff, Jeffrey S. Urbach, Optical neuronal guidance in three-dimensional matrices.

Journal of Neuroscience Methods, Volume 179, Issue 2, 15 May 2009, Pages 278–283

14. Allen Ehrlicher, Timo Betz, Björn Stuhrmann, Michael Gögler, Daniel Koch, Kristian Franze, Yunbi Lu, Josef Käs, Optical Neuronal Guidance. Methods in Cell Biology, Volume 83, 2007, Pages 495–520

15. Valeri Vasioukhin, Christoph Bauer, Mei Yin, Elaine Fuchs, Directed actin polymerization is the driving force for epithelial cell-cell adhesion. Cell, Volume 100, Issue 2, 21 January 2000, Pages 209–219

16. Toomas Ku¨ barsepp, Atte Haapalinna, Petri Ka¨ rha¨, Erkki Ikonen, Nonlinearity measurements of silicon photodetectors. Applied Optics, 1 May 1998.

Materials

[a] China Medical Science Institute for Experimental Animal Research - 中国医学科学院医学实验动物研究所

[b] PC12 cells from 2 sources at Peking Union Medical Institute Cell-Bank - 北京协和细胞库; and Wang Fei at Peking Union Medical Institute - 北京协和医学院王斐

[c] Sigma-Aldrich Company

[d] Analytical pure, sourced in China

[e] Thermo Fisher Scientific/Hyclone

[f] Hangzhou Sijiqing Biological Engineering Materials Co. 杭州四季青公司

[g] Invitrogen/Gibco

[h] Thermo Fisher Scientific/Nunc

[i] Beijing Zhongguancun Zhongfa Electronics Supply Market, 3160

[j] Narishige MP-830 ocular 15x, objective 35x

[k] Olympus 10x, 40x

[l] Beijing Viasho Technology Co., Ltd.

Notes

Calculations for the energy of the radiation at various distances r (in μm) from the fiber taper tip. The preceeding vⁿ corresponds to locations in figure 6.

v¹ r = 1; area ≈ 12μm²/4 = 3μm²; 135mw / 3 μm² = 45mW /μm²

v³ r = 3; area ≈ 113μm²/4 = 28.25μm²; 135mw / 28.25μm² = 4.8mW /μm²

v⁵ r = 5; area ≈ 314μm²/4 =78.5μm²; 135mw / 78.5μm² = 1.72mW /μm²

v⁷ r = 7; area ≈ 615μm²/4 =153.8μm²; 135mw / 153.8μm² = 0.88mW /μm²