Sperm Migration
Sperm Dynamics in Tubular Confinement Veronika Magdanz,* Britta Koch, Samuel Sanchez, and Oliver G. Schmidt* Recent studies suggest that the in vivo microenvironment plays a decisive role for the behavior of motile cells.[1–3] It has been an on-going challenge to mimic environments of motile cells in order to investigate the interaction of moving cells and their surroundings. Especially, microenvironments of tubular shape that can be found widely in nature such as in sperm migration,[4] artery networks, lymphatic vascular systems,[5] or during the formation of biofilms by bacterial motility,[6,7] have been challenging to reproduce in vitro. Until now, the journey of spermatozoa to the fertilization site in the body is still not well understood and thus subject to many studies that try to elucidate how exactly sperm cells reach the egg cell and explain reasons for infertility. It was for example reported that special passageways exist for sperm cells that guide them through the cervix. The cervical canal is about 3 mm long and sperms were found to travel at speeds of about 35 µm*s−1 through these roughly 10 µm wide microchannels.[8,9] Denissenko et al. mention that the cervical crypts and folds of the ampullary fallopian tubes are comparable to channels that have a width in the order of 100 µm.[4] One of the required tools for elucidating the journey of sperm cells to the oocyte and understanding fertilization would be a device that mimics the microenvironment in which the sperm cells swim, since little is known about how sperm cells actually travel in vivo. In vitro experiments often lack comparability to in vivo conditions because the microenvironment is imitated insufficiently. Recent approaches to studying sperm cell behaviour in these microenvironments were presented using microfluidic channels with rectangular cross sections made
V. Magdanz, B. Koch, Prof. O. G. Schmidt Institute for Integrative Nanosciences Leibniz Institute for Solid State and Materials Research Helmholtzstr. 20 01069, Dresden, Germany E-mail: [email protected]; [email protected] Dr. S. Sanchez Max Planck Institute for Intelligent Systems Heisenbergstr. 3 70569, Stuttgart, Germany Prof. O. G. Schmidt Material Systems for Nanoelectronics Chemnitz University of Technology Reichenhainer Str. 70, Chemnitz 09107, Germany Prof. O. G. Schmidt Center for Advancing Electronics Dresden Dresden University of Technology 01187, Dresden, Germany DOI: 10.1002/smll.201401881 small 2014, DOI: 10.1002/smll.201401881
from PDMS.[9–13] They address the question why sperm cells move towards walls and suggest that, besides sperm cell motility, chemotaxis, thermotaxis, rheotaxis and the interaction with surfaces (defined as thigmotaxis) might also play a significant role in in vivo sperm guidance. Thigmotaxis describes the tendency of motile cells to remain close to walls and respond to the mechanical stimulus of physical contact with surfaces. Denissenko et al.[4] found that sperm cells rarely swim in the center of microchannels, but rather in the channel corners. Additionally, sperm cell migration depended strongly on the channel geometry. The reason for sperm cells swimming against walls is explained by the geometry of the head and flagella movement. The conical envelope of the flagella wave is larger compared to the amplitude of head oscillations which leads to a propulsion direction inclined towards the wall. Another recent approach tried to recreate the fluid flow and the surface topography during sperm migration in the reproductive tract by the use of microfluidic PDMS channels with microgrooves.[12] This study suggests that microgrooves assist sperm cells in staying on a surface and swimming against flow. It also demonstrated that spermatozoa swim faster in 20 µm wide grooves than in 10 µm wide grooves. However, it is unlikely that the microgrooves in the reproductive system have rectangular cross-sections. Sperm cells seem to sense the complex 3D environment in vivo and use them as guidance assistance towards the fertilization site. Studying sperm movement through microchannels may reveal undiscovered swimming parameters underlying successful tract migration.[4] There is high interest in improving the selection methods for spermatozoa for in vitro fertilization (IVF). State-of-the-art methods to select the most fertile spermatozoa for IVF often rely on the swim-up method. This method is cheap but the selection relies solely on the ability of spermatozoa to swim upwards in a vial with medium. New methods with improved selection quality are demonstrated for instance using sperm chemotaxis,[14] sedimentation,[15] immobilizing motile sperm on fibronectin islands[16] or racing cells through microfluidic channels.[9] All these methods are interesting selection processes promising to improve IVF procedures. However, the knowledge about how spermatozoa migrate through tubular channels might help to improve existing selection methods of spermatozoa for IVF. In this article, we present tubular microenvironments on chip as a tool for studying sperm cell motion near cylindrical surfaces consisting of transparent rolled-up SiO/SiO2 microtubes with diameters from 5–45 µm and lengths up to 600 µm. Transparent rolled-up microtubes have recently been
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1
communications
V. Magdanz et al.
shown by our group to serve as tools in biological studies to investigate cell behavior in confinement during mitosis[17] and DNA damage.[18] Moreover, rolled-up nanotechnology offers mass fabrication of microtubes of various materials that can be useful, e.g., in nano- and microrobotics,[19,20] sensing[21] and energy storage applications.[22,23] In contrast to existing microchannels, these microtubes contain a rounded inner shape without corners. These transparent microtubes are used in this study to analyze the motion of sperm cells inside the tubular confinements in comparison to cells swimming on planar surface. To eliminate the influence of surface topography in our experiments, we choose glass microtubes for the spacial confinement studies. The SiO/SiO2 microtubes have a surface roughness comparable to two-dimensional glass surfaces (e.g., microscope slides commonly used for analysis of sperm motion) which rules out any additional influence. Silicon oxide microtubes have a fairly smooth surface in the lower nanometer range (atomic force microscopy measurements of the SiO/SiO2 microtube surface confirmed a roughness of less than 2 nanometers). This is the first report of directly relating sperm cell velocities, directionality and linearity inside tubular confinement to the tube diameter. Additionally, it is shown how the directionality of spermatozoa movement is improved and cell trajectories can be controlled by using transparent microtubes as tubular channels for motile cells. The observation of increased cell velocity of sperm cells inside tubular channels in a certain range of tube diameter is a hint that tubular channels might improve sperm cell migration. The transparent microtubes are fabricated with diameters ranging from 10–45 µm using strain-induced roll-up of bilayers of silicon oxide/dioxide nanomembranes that were deposited on 18 x 18 mm glass substrates patterned with photoresist.[24] When etching the sacrificial photoresist by dissolution in acetone, the silicon oxide layers roll up into microtubes with several windings and stay attached on the chip. The final diameter of rolled up microtubes is predetermined by the adjustment of deposition parameters such as layer thickness and deposition rate.[25] The result is an array of fixed rolled up microtubes, as displayed in Figure 1a, that can be used to observe moving patterns of spermatozoa on-chip. For the observation of moving bovine spermatozoa through microtubes with different diameters on chip, the microtube array is immersed in the sperm cell solution which contains thawed bull semen (200 µL) suspended in 3 mL SP-TALP (modified Tyrode’s Albumin Lactate Pyruvate Medium) with a final cell concentration of 108 cells mL−1. The spermatozoa are observed under an inverted microscope and their trajectories are recorded at 20 frames per second. The image analysis software ImageJ is used to analyze the cell trajectories. An optical image series is displayed in Figure 1b and shows a bovine spermatozoon entering and moving through a 100 µm long silicon oxide microtube. Figure 1c and 1d show trajectories of the same 3 bovine spermatozoa (blue, red and black lines) swimming through a tubular channel in (c) compared to cells swimming on the flat surface in (d). As expected, it can be seen that the cells swimming through the 35 µm microtube display a much straighter path than the same cells swimming on a two-dimensional surface. It can also be seen
2 www.small-journal.com
Figure 1. Microtubes as channels for guiding motile cells. a) Schematic of microtube array with spermatozoa swimming on chip. b) Image series of a bovine spermatozoon moving through a silicon oxide microtube. Red arrows point at sperm head. Scale bar: 20 µm. c) Trajectories of 3 sperm swimming inside 35 µm wide microtubes and d) trajectories of the same 3 cells on flat surface. Each colored line is a 2.4 s long trajectory. The coordinate (0,0) marks the starting point of all the tracks. See also supporting video S1 which tracks the swimming path of a bovine spermatozoon on flat surface (green line) and inside a tubular 35 µm wide channel (blue line) over 2.4 s.
that the tracks inside the microtubes span a longer path (in average 60% longer) than on the flat surface which indicates a higher velocity of cells inside microtubes. When a spermatozoon is swimming freely or on a planar surface, it can move in a wavy, circular or even corkscrew-like motion. In order to quantify the change in motion of spermatozoa through tubular microchannels, we first analyze the velocity and linearity as parameters to assess the cell trajectories. A common method for the evaluation of velocity of spermatozoa is the determination of curvilinear velocity (VCL) and straight line velocity (VSL) from tracked cell paths.[10,26] VCL describes the actual velocity along the path (point-topoint) whereas the VSL refers to the total straight-line distance between the starting and ending points of the tracked sperm trajectory.[9] In order to normalize the velocity of each cell to its velocity before it enters the microtube, the relative average straight line velocity was calculated as ratio of the VSL of a spermatozoon swimming inside the microtube and outside the microtube and displayed as percentage values: VSLrel =
VSLin × 100 VSLout
(1)
In the same way, the relative average curvilinear velocity is calculated as VCLrel =
VCLin × 100 VCLout
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
(2)
small 2014, DOI: 10.1002/smll.201401881
Sperm Dynamics in Tubular Confinement
In general, VSL is smaller than VCL for spermatozoa since the cell travels a longer track on its curvilinear path than on its total straight line path. The relative velocities VSLrel and VCLrel are equal as long as there is no difference between the velocity outside and inside the microtube. If the microtube creates a straightening of the cell motion, VSLrel will be larger than VCLrel. Furthermore, the linearity is defined as LIN =
VSL × 100 VCL
(3)
and provides information about the straightness of the path. As the value LIN = 1 is being approached, the high linearity illustrates a straighter swimming path. As shown in Figure 2a, for a bovine spermatozoon to be able to pass through a microtube a minimum diameter of 5 µm is required. The average diameter of a bovine sperm head was measured to be 4.83 ± 0.59 µm (N = 50); therefore, this is the limiting factor for the cells in order to enter a microtube. An increasing tube diameter allows the cells to swim faster through the tube on-chip which causes an increase of VSLrel and VCLrel. At a diameter of 22 µm, the speed does not increase much with increasing tube diameter. This can be explained with the amplitude of the bovine sperm flagella, which is 3 – 8 µm in each direction as previously reported[26] and confirmed by our experiments (data not shown). This means that a microtube of 22 µm diameter and larger allows a nearly undisturbed flagella movement. At diameters larger than 30 µm, the straight line velocity of the sperm cell inside the tube is in some cases even higher than outside the tube (values higher than 100%). The tube walls have a guiding effect on the cell motion and turn its rotational or helical movement into a more directed and straight motion. This effect decreases at smaller tube diameters due to the increasing number of collisions of the sperm head and flagella on the rigid tube walls which leads to a slow-down of the cell. In addition, this “straightening” effect can be quantified by the difference between VSLrel and VCLrel as illustrated in Figure 2a by the marked area between the VCLrel and VSLrel curve. The motion-smoothing effect occurs in all microtubes with diameters smaller than 45 µm. The smoothing effect is especially significant at tube diameters of 20–40 µm. In this range, the flagella motion is unaffected by the tube diameter and the cell swims in straighter paths compared to the motion on flat surfaces. Additionally, the velocity decrease is only minimal (ca. 80% of the curvilinear velocity outside the tube, see Figure 2a). The straightening effect disappears at larger tube diameters because the microtubes are too wide to have influence on the cell motion and therefore, the curvilinear and straight line velocity of spermatozoa are the same inside and outside the microtube. For that reason, VCLrel and VSLrel overlap at about 100% at tube diameters of 45 µm. That means that the curvilinear velocity of cells does not change when they enter microtubes with a diameter 40 µm and larger, compared to their swimming speed on two-dimensional surfaces. small 2014, DOI: 10.1002/smll.201401881
Figure 2. a) Relative average straight line velocity VSLrel and curvilinear velocity VCLrel of bovine sperm cells swimming through microtubes with different tube diameters. The filled area between the two curves marks the range of microtube diameter in which the straightening effect occurs. b) Linearity LIN of spermatozoa as function of microtube diameter. The linearity is calculated as the ratio of straight line velocity and curvilinear velocity. The red dashed line represents the average linearity of spermatozoa swimming on planar surface. X error bars are range of tube diameters, Y error bars are standard error of the mean.
It was calculated from tracked cell paths that the linearity of swimming patterns through the microtubes decreases with increasing tube diameter, as shown in Figure 2b. At a diameter of about 40 µm, the linearity inside tubes corresponds to the average linearity observed for cells swimming on planar surfaces. Hence, the influence of the tubular confinement on the linearity is lost at tube diameters larger than 40 µm; in smaller microtubes an improved straightness of path is observed. This confirms the observation that the tubular confinement has a guiding effect on the spermatozoa mainly in tubular channels of dimension 20–45 µm in diameter. In the next step, the directionality, as defined by Paxton[27] was determined from the recorded cell trajectories. Directionality is defined as cos (θ), where θ is the angle between sperm
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
3
communications
V. Magdanz et al.
Figure 3. Directionality cos (θ) of cells moving through tubes of 12, 17, 20 and 30 µm tube diameter over time. The green background marks the point in time when the cells leave the tube which is characterized by the sharp drop in all 4 data sets. Directionality is defined as cos (θ), where θ is the angle between sperm head axis and directionality vector. The inset shows two rolled up silicon oxide microtubes with diameter 25 µm (top) and 35 µm (bottom). Scale bar: 50 µm.
head axis and directionality vector. The graph in Figure 3 shows the directionality of sperm cell movement over time when swimming through different microtubes. It illustrates how different diameters of microtubes (12 – 30 µm) determine the directionality. An average directionality of 0.96 ± 0.09 inside microtubes was derived. This value of directionality is much closer to 1 (meaning straighter cell motion) than that of a bull spermatozoon swimming on a planar surface (average directionality outside tubes = 0.80 ± 0.15). At 2.4 s in the graph in Figure 3, when the cell leaves the tube, the directionality abruptly drops. Additionally, the directionality inside microtubes increases with decreasing tube diameter due to the channeling effect of the surrounding walls that straighten the path of the spermatozoa. With increasing tube diameter from 17 to 30 µm, the average directionality slightly decreases from 0.99 to 0.96. The increasing tube diameter allows the cells to freely follow their naturally bending motion and hence decreases their directionality. The inset of Figure 3 shows two 250 µm long transparent microtubes with 25 µm diameter (top) and 35 µm diameter (bottom), respectively. The on-chip guidance of sperm cells inside microtubes offers an improvement in directionality and linearity as well as an increased velocity in microtubes with diameters of 20–45 µm. However, in microtubes narrower than 20 µm the cell speed is reduced due to the collisions of sperm head and flagellum with the tubular walls. The microtubes can serve as a tool for predetermining cell swimming paths for applications in lab-on-a-chip devices. We demonstrated that transparent silicon oxide microtubes are a very useful system to study motion of motile cells in cylindrical confinement. The microtubes exhibit excellent optical properties for visualizing and tracking of single cells. From trajectory measurements we can conclude that the motion of motile cells inside a tube on-chip shows a higher
4 www.small-journal.com
directionality and linearity than the motion of cells swimming unconfined on a glass surface. This confirms observations made by Tung et al.[12] that sperm cells in microgrooves have an improved directional persistence. The cell velocity also directly correlates with the tube diameter. At a microtube diameter of 20 µm and larger, the cells reach >100% of their initial straight line velocity outside the tube. The guidance of sperm cells inside microtubes offers a great way to direct the motion of sperm cells on-chip. It is an alternative to chemotactically or thermotactically control spermatozoa movement which has been a very challenging goal because the chemotactic behavior of sperm cells is poorly understood. An interesting observation during this study is that straight line velocities of sperm cells inside microtubes of diameters 20–45 µm can be higher than the velocity of cells swimming on a two-dimensional surface. The tube walls have a “straightening” effect on the trajectories of the motile cells. This is an indication that the sperm cell motion in confinement, as it can be found in vivo in cervical crypts or folds of ampullary fallopian tubes, significantly differs from the motion on 2D flat surfaces. The tubular microchannels could serve as a tool to more realistically investigate or even mimic this motion. There are no existing studies on the influence of surface roughness on sperm motion, to the best of our knowledge. Therefore, it would be an interesting topic for future studies and one could imagine creating tubular structures with different inner roughness grades and studying the influence of this surface effect on sperm motion. The application of these transparent microtubes can be further extended by integrating sensors for on-line cell analysis which is of high interest for the selection and separation of the most fertile spermatozoa for in vitro fertilization procedures. Up to our knowledge, it is not proven that the motility of cells is evidence alone for the most fertile sperm cells.[28] Therefore, a device with tubular channels that is able to integrate the selection of progressively moving cells with further analysis of metabolic features that hint to high fertilization capability would be promoting a high-quality sperm selection method. This approach confirms the assumption that microenvironment is important for sperm migration and provides information for further chip development that mimics natural environment and possibly provides a sperm selection mechanism similar to how it occurs in vivo. Sperm migration in vivo is a very complex process influenced by chemical and mechanical cues, morphology of the surroundings, sperm membrane modifications, pH, temperature, fluid flow and possibly many other factors.[29–33] Further research is required to define the role of each influencing factor and to reveal the mechanisms by which sperm travel through the female reproductive tract and perform successful fertilization.
Experimental Section Fabrication of Microtubes: The fabrication of the microtube samples followed established procedures that are in detail described elsewhere.[17,18,34] Briefly, rectangular photoresist patterns (100 µm in width, 100–500 µm in length) were photolithographically structured on 18 × 18 mm glass cover slides. In an electron beam evaporation system first a 5 nm thin layer of
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2014, DOI: 10.1002/smll.201401881
Sperm Dynamics in Tubular Confinement
silicon monoxide (SiO) and then a 60–100 nm thick layer of silicon dioxide (SiO2) were deposited on the photoresist patterns with a glancing angle of 30°. The layer thickness ratio of SiO to SiO2 later on defined the diameter of the microtubes. Because of the shadowing effect during the angular deposition, an uncovered region behind the rectangular structures was maintained. Through this window the photoresist sacrificial layer was in the next step dissolved in DMSO. The strain-engineered SiO/SiO2 bilayer film self-assembled into micron-sized tubes upon release from the photoresist structures. In order to stabilize the microtubes, the sample surfaces were covered with 18 nm of aluminium oxide in a subsequent atomic layer deposition step. Spermatozoa Treatment and Analysis: Cryopreserved bovine semen straws were thawed for 10 minutes at 37 °C. The 200 µL of bovine semen were diluted in 2 mL SP-TALP (modified Tyrode’s Albumin Lactate Pyruvate Medium) and incubated for additional 10 minutes. The array with SiO/SiO2 microtubes was immersed in the sperm cell solution. For the analysis of spermatozoa movement an inverted Zeiss microscope (10× and 20× objectives) was used with an attached Miro eX2 high speed camera from Vision Research. Videos were recorded at 20 frames per second and the software Phantom MulitCam was used for the image acquisition. The velocity measurements and tracking was done by the use of the ImageJ plugin Manual Tracking.
Supporting Information Supporting Information is available from the Wiley Online Library or from the author.
Acknowledgements We thank Dr. R. Träger for design of schematics in Figure 1a and the Table of Content graphic. We thank Masterrind GmbH in Meißen for donating frozen bull semen and Dr. B. Eichler for AFM measurements. The results leading to this work have been financially supported by the Volkswagen Foundation (86 362), the German Science Foundation (DFG) through the priority program SPP1726 “Microswimmers: From Single Particle Motion to Collective Behaviour” and the European Research Council (ERC) for Starting Grant “Lab-in-a-tube and Nanorobotics biosensors; LT-NRBS” [no. 311529].
[1] W. J. Polacheck, I. K. Zervantonakis, R. D. Kamm, Cell. Mol. Life Sci. 2013, 70, 8, 1335–1356. [2] P. Chiarugi, Cell Commun. Signal. 2010, 8, 25. [3] J. Brabek, C. T. Mierke, D. Roesel, P. Vesely, B. Fabry, Cell Commun. Signal. 2010, 8, 22. [4] P. Denissenko, V. Kantsler, D. J. Smith, J. Kirkman-Brown, J. Proc. Nat. Ac. Soc. 2012, 109, 21, 8007–8010.
small 2014, DOI: 10.1002/smll.201401881
[5] K. H. K. Wong, J. G. Truslow, A. H. Khankhel, K. L. S. Chan, J. Tien, J. Biomed. Mater. Res. 101A 2013, 8, 2181–219. [6] Y. M. Romanova, T. A. Smirnova, A. L. Andreev, T. S. Il’Ina, L. V. Didenko, A. L. Gintsburg, Biology 2006, 75, 4, 481–485. [7] L. Hall-Stoodley, W. Costerton, P. Stoodely, Nature Rev. Microbiol. 2004, 2, 95–108. [8] S. S. Suarez, Exp. Mech. 2010, 50, 1267–1274. [9] S. Tasoglu, H. Safaee, X. Zhang, J. L. Kingsley, P. N. Catalano, U. A. Gurkan, A. Nureddin, E. Kayaalp, R. M. Anchan, R. L. Maas, E. Tuezel , U. Demirci, Small 2013, 9, 20. [10] L. J. Kricka, O. Nosaki, S. Heyner, W. T. Garside, P. Wilding, Clin. Chem. 1993, 39, 9, 1944–1947. [11] L. Xie, R. Ma, C. Han, K. Su, Q. Zhang, T. Qiu, L. Wang , G. Huang, J. Qiao, J. Wang, J. Cheng, Clin. Chem. 2010, 56, 8. [12] C.-K. Tung, F. Ardon, A. G. Fiore, S. S. Suarez, M. Wu, Lab Chip 2014, 14, 1348. [13] Y.-A. Chen, K.-C. Chen, V. F. S. Tsai, Z.-W. Huang, J.-T. Hsieh, A. M. Wo, Clin. Chem. 2013, 59, 3. [14] Y.-J. Ko, J.-H. Maeng, B.-C. Lee, S. Lee, S. Hwang, Y. Ahn, Analyt. Sci 2012, 28, 27. [15] C.-Y. Chen, T.-C. Chiang, C.-M. Lin, S.-S. Lin, D.-S. Jong, V. F.-S. Tsai, J.-T. Hsieh, A. M. Wo, Analyst 2013, 138, 4967–4974. [16] J.-P. Frimat, M. Bronkhorst, B. de Wagenaar, J. G. Bomer, F. van der Heijden, A. van den Berg, L. I. Segerink, Lab Chip 2014, 00, 1, 1–3. [17] W. Xi, C. K. Schmidt, S. Sanchez, D. H. Gracias, R. E. Carazo-Salas, S. P. Jackson, O. G. Schmidt, Nano Lett. 2014, DOI: 10.1021/ nl4042565. [18] B. Koch, S. Sanchez, C. K. Schmidt, A. Swiersy, S. P. Jackson, O. G. Schmidt, Adv. Healthcare Mater. 2014, DOI: 10.1002/ adhm.201300678. [19] Y. F. Mei, A. A. Solovev, S. Sanchez, O. G. Schmidt, Chem. Soc. Rev. 2011, 40, 2109. [20] V. Magdanz, S. Sanchez, O. G. Schmidt, Adv. Mater. 2013, 25, 6581. [21] S. M. Harazim, V. A. Bolaños Quiñones, S. Kiravittaya, S. Sanchez, O. G. Schmidt, Lab Chip 2012, 12, 2649. [22] J. Deng, C. Yan, L. Yang, S. Baunack, S. Oswald, H. Wendrock, Y. F. Mei, O. G. Schmidt, ACS Nano 2013, 7, 8, 6948. [23] J. Deng, H. Ji, C. Yan, J. Zhang, W. Si, S. Baunack, S. Oswald, Y. F. Mei, O. G. Schmidt, Angew. Chem. Int. Edit. 2013, 52, 2326. [24] G. Huang, Y. F. Mei, D. J. Thurmer, A. Coric, O. G. Schmidt, Lab Chip 2009, 9, 263–268. [25] B. Robaire, B. T. Hinton, The Epididymis:from molecules to clinical practice, Orgebin-Crist, Springer, NY, USA 2002. [26] R.Rikmenspoel, J. Exp. Biol. 1984, 108, 205–230. [27] W. F. Paxton, K. C. Kistler, C. C. Olmeda, A. Sen, S. K. St Angelo, Y. Cao, T. E. Mallouk, P. E. Lammert, V. H. Crespi, J. Am. Chem. Soc. 2004, 126, 41, 13424–31. [28] J. E. Ellington, D. P. Evenson, R. W. Wright Jr., A. E. Jones, C. S. Schneide, G. A. Hiss, R. S. Brísbois, Fertil Steril 1999, 71, 5, 924–929. [29] M. Eisenbach, L. C. Giojalas, Nat Rev Mol. Cell. Bio. 2006, 7, 276–285. [30] J. Cosson, P. Huitorel, C. Gagnon, Cell Motil. Cytoskel. 2003, 54, 56–63. [31] M. A. Scott, Anim. Reprod. Sci. 2000, 60, 61, 337–348. [32] K. Miki, D. E. Clapham, Curr. Biol. 2013, 23, 6, 443–452. [33] A. Contri, A. Gloria, D. Robbe, C. Valorz, L. Wegher, A. Carluccio, Anim. Reprod. Sci. 2013, 136, 4, 252–259. [34] S. M. Harazim, W. Xi, C. K. Schmidt, S. Sanchez, O. G. Schmidt, J. Mater. Chem. 2012, 22, 7, 2878–2884. Received: June 27, 2014 Revised: September 15, 2014 Published online:
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
5
Sperm Dynamics in Tubular Confinement Veronika Magdanz,* Britta Koch, Samuel Sanchez, and Oliver G. Schmidt* Recent studies suggest that the in vivo microenvironment plays a decisive role for the behavior of motile cells.[1–3] It has been an on-going challenge to mimic environments of motile cells in order to investigate the interaction of moving cells and their surroundings. Especially, microenvironments of tubular shape that can be found widely in nature such as in sperm migration,[4] artery networks, lymphatic vascular systems,[5] or during the formation of biofilms by bacterial motility,[6,7] have been challenging to reproduce in vitro. Until now, the journey of spermatozoa to the fertilization site in the body is still not well understood and thus subject to many studies that try to elucidate how exactly sperm cells reach the egg cell and explain reasons for infertility. It was for example reported that special passageways exist for sperm cells that guide them through the cervix. The cervical canal is about 3 mm long and sperms were found to travel at speeds of about 35 µm*s−1 through these roughly 10 µm wide microchannels.[8,9] Denissenko et al. mention that the cervical crypts and folds of the ampullary fallopian tubes are comparable to channels that have a width in the order of 100 µm.[4] One of the required tools for elucidating the journey of sperm cells to the oocyte and understanding fertilization would be a device that mimics the microenvironment in which the sperm cells swim, since little is known about how sperm cells actually travel in vivo. In vitro experiments often lack comparability to in vivo conditions because the microenvironment is imitated insufficiently. Recent approaches to studying sperm cell behaviour in these microenvironments were presented using microfluidic channels with rectangular cross sections made
V. Magdanz, B. Koch, Prof. O. G. Schmidt Institute for Integrative Nanosciences Leibniz Institute for Solid State and Materials Research Helmholtzstr. 20 01069, Dresden, Germany E-mail: [email protected]; [email protected] Dr. S. Sanchez Max Planck Institute for Intelligent Systems Heisenbergstr. 3 70569, Stuttgart, Germany Prof. O. G. Schmidt Material Systems for Nanoelectronics Chemnitz University of Technology Reichenhainer Str. 70, Chemnitz 09107, Germany Prof. O. G. Schmidt Center for Advancing Electronics Dresden Dresden University of Technology 01187, Dresden, Germany DOI: 10.1002/smll.201401881 small 2014, DOI: 10.1002/smll.201401881
from PDMS.[9–13] They address the question why sperm cells move towards walls and suggest that, besides sperm cell motility, chemotaxis, thermotaxis, rheotaxis and the interaction with surfaces (defined as thigmotaxis) might also play a significant role in in vivo sperm guidance. Thigmotaxis describes the tendency of motile cells to remain close to walls and respond to the mechanical stimulus of physical contact with surfaces. Denissenko et al.[4] found that sperm cells rarely swim in the center of microchannels, but rather in the channel corners. Additionally, sperm cell migration depended strongly on the channel geometry. The reason for sperm cells swimming against walls is explained by the geometry of the head and flagella movement. The conical envelope of the flagella wave is larger compared to the amplitude of head oscillations which leads to a propulsion direction inclined towards the wall. Another recent approach tried to recreate the fluid flow and the surface topography during sperm migration in the reproductive tract by the use of microfluidic PDMS channels with microgrooves.[12] This study suggests that microgrooves assist sperm cells in staying on a surface and swimming against flow. It also demonstrated that spermatozoa swim faster in 20 µm wide grooves than in 10 µm wide grooves. However, it is unlikely that the microgrooves in the reproductive system have rectangular cross-sections. Sperm cells seem to sense the complex 3D environment in vivo and use them as guidance assistance towards the fertilization site. Studying sperm movement through microchannels may reveal undiscovered swimming parameters underlying successful tract migration.[4] There is high interest in improving the selection methods for spermatozoa for in vitro fertilization (IVF). State-of-the-art methods to select the most fertile spermatozoa for IVF often rely on the swim-up method. This method is cheap but the selection relies solely on the ability of spermatozoa to swim upwards in a vial with medium. New methods with improved selection quality are demonstrated for instance using sperm chemotaxis,[14] sedimentation,[15] immobilizing motile sperm on fibronectin islands[16] or racing cells through microfluidic channels.[9] All these methods are interesting selection processes promising to improve IVF procedures. However, the knowledge about how spermatozoa migrate through tubular channels might help to improve existing selection methods of spermatozoa for IVF. In this article, we present tubular microenvironments on chip as a tool for studying sperm cell motion near cylindrical surfaces consisting of transparent rolled-up SiO/SiO2 microtubes with diameters from 5–45 µm and lengths up to 600 µm. Transparent rolled-up microtubes have recently been
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1
communications
V. Magdanz et al.
shown by our group to serve as tools in biological studies to investigate cell behavior in confinement during mitosis[17] and DNA damage.[18] Moreover, rolled-up nanotechnology offers mass fabrication of microtubes of various materials that can be useful, e.g., in nano- and microrobotics,[19,20] sensing[21] and energy storage applications.[22,23] In contrast to existing microchannels, these microtubes contain a rounded inner shape without corners. These transparent microtubes are used in this study to analyze the motion of sperm cells inside the tubular confinements in comparison to cells swimming on planar surface. To eliminate the influence of surface topography in our experiments, we choose glass microtubes for the spacial confinement studies. The SiO/SiO2 microtubes have a surface roughness comparable to two-dimensional glass surfaces (e.g., microscope slides commonly used for analysis of sperm motion) which rules out any additional influence. Silicon oxide microtubes have a fairly smooth surface in the lower nanometer range (atomic force microscopy measurements of the SiO/SiO2 microtube surface confirmed a roughness of less than 2 nanometers). This is the first report of directly relating sperm cell velocities, directionality and linearity inside tubular confinement to the tube diameter. Additionally, it is shown how the directionality of spermatozoa movement is improved and cell trajectories can be controlled by using transparent microtubes as tubular channels for motile cells. The observation of increased cell velocity of sperm cells inside tubular channels in a certain range of tube diameter is a hint that tubular channels might improve sperm cell migration. The transparent microtubes are fabricated with diameters ranging from 10–45 µm using strain-induced roll-up of bilayers of silicon oxide/dioxide nanomembranes that were deposited on 18 x 18 mm glass substrates patterned with photoresist.[24] When etching the sacrificial photoresist by dissolution in acetone, the silicon oxide layers roll up into microtubes with several windings and stay attached on the chip. The final diameter of rolled up microtubes is predetermined by the adjustment of deposition parameters such as layer thickness and deposition rate.[25] The result is an array of fixed rolled up microtubes, as displayed in Figure 1a, that can be used to observe moving patterns of spermatozoa on-chip. For the observation of moving bovine spermatozoa through microtubes with different diameters on chip, the microtube array is immersed in the sperm cell solution which contains thawed bull semen (200 µL) suspended in 3 mL SP-TALP (modified Tyrode’s Albumin Lactate Pyruvate Medium) with a final cell concentration of 108 cells mL−1. The spermatozoa are observed under an inverted microscope and their trajectories are recorded at 20 frames per second. The image analysis software ImageJ is used to analyze the cell trajectories. An optical image series is displayed in Figure 1b and shows a bovine spermatozoon entering and moving through a 100 µm long silicon oxide microtube. Figure 1c and 1d show trajectories of the same 3 bovine spermatozoa (blue, red and black lines) swimming through a tubular channel in (c) compared to cells swimming on the flat surface in (d). As expected, it can be seen that the cells swimming through the 35 µm microtube display a much straighter path than the same cells swimming on a two-dimensional surface. It can also be seen
2 www.small-journal.com
Figure 1. Microtubes as channels for guiding motile cells. a) Schematic of microtube array with spermatozoa swimming on chip. b) Image series of a bovine spermatozoon moving through a silicon oxide microtube. Red arrows point at sperm head. Scale bar: 20 µm. c) Trajectories of 3 sperm swimming inside 35 µm wide microtubes and d) trajectories of the same 3 cells on flat surface. Each colored line is a 2.4 s long trajectory. The coordinate (0,0) marks the starting point of all the tracks. See also supporting video S1 which tracks the swimming path of a bovine spermatozoon on flat surface (green line) and inside a tubular 35 µm wide channel (blue line) over 2.4 s.
that the tracks inside the microtubes span a longer path (in average 60% longer) than on the flat surface which indicates a higher velocity of cells inside microtubes. When a spermatozoon is swimming freely or on a planar surface, it can move in a wavy, circular or even corkscrew-like motion. In order to quantify the change in motion of spermatozoa through tubular microchannels, we first analyze the velocity and linearity as parameters to assess the cell trajectories. A common method for the evaluation of velocity of spermatozoa is the determination of curvilinear velocity (VCL) and straight line velocity (VSL) from tracked cell paths.[10,26] VCL describes the actual velocity along the path (point-topoint) whereas the VSL refers to the total straight-line distance between the starting and ending points of the tracked sperm trajectory.[9] In order to normalize the velocity of each cell to its velocity before it enters the microtube, the relative average straight line velocity was calculated as ratio of the VSL of a spermatozoon swimming inside the microtube and outside the microtube and displayed as percentage values: VSLrel =
VSLin × 100 VSLout
(1)
In the same way, the relative average curvilinear velocity is calculated as VCLrel =
VCLin × 100 VCLout
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
(2)
small 2014, DOI: 10.1002/smll.201401881
Sperm Dynamics in Tubular Confinement
In general, VSL is smaller than VCL for spermatozoa since the cell travels a longer track on its curvilinear path than on its total straight line path. The relative velocities VSLrel and VCLrel are equal as long as there is no difference between the velocity outside and inside the microtube. If the microtube creates a straightening of the cell motion, VSLrel will be larger than VCLrel. Furthermore, the linearity is defined as LIN =
VSL × 100 VCL
(3)
and provides information about the straightness of the path. As the value LIN = 1 is being approached, the high linearity illustrates a straighter swimming path. As shown in Figure 2a, for a bovine spermatozoon to be able to pass through a microtube a minimum diameter of 5 µm is required. The average diameter of a bovine sperm head was measured to be 4.83 ± 0.59 µm (N = 50); therefore, this is the limiting factor for the cells in order to enter a microtube. An increasing tube diameter allows the cells to swim faster through the tube on-chip which causes an increase of VSLrel and VCLrel. At a diameter of 22 µm, the speed does not increase much with increasing tube diameter. This can be explained with the amplitude of the bovine sperm flagella, which is 3 – 8 µm in each direction as previously reported[26] and confirmed by our experiments (data not shown). This means that a microtube of 22 µm diameter and larger allows a nearly undisturbed flagella movement. At diameters larger than 30 µm, the straight line velocity of the sperm cell inside the tube is in some cases even higher than outside the tube (values higher than 100%). The tube walls have a guiding effect on the cell motion and turn its rotational or helical movement into a more directed and straight motion. This effect decreases at smaller tube diameters due to the increasing number of collisions of the sperm head and flagella on the rigid tube walls which leads to a slow-down of the cell. In addition, this “straightening” effect can be quantified by the difference between VSLrel and VCLrel as illustrated in Figure 2a by the marked area between the VCLrel and VSLrel curve. The motion-smoothing effect occurs in all microtubes with diameters smaller than 45 µm. The smoothing effect is especially significant at tube diameters of 20–40 µm. In this range, the flagella motion is unaffected by the tube diameter and the cell swims in straighter paths compared to the motion on flat surfaces. Additionally, the velocity decrease is only minimal (ca. 80% of the curvilinear velocity outside the tube, see Figure 2a). The straightening effect disappears at larger tube diameters because the microtubes are too wide to have influence on the cell motion and therefore, the curvilinear and straight line velocity of spermatozoa are the same inside and outside the microtube. For that reason, VCLrel and VSLrel overlap at about 100% at tube diameters of 45 µm. That means that the curvilinear velocity of cells does not change when they enter microtubes with a diameter 40 µm and larger, compared to their swimming speed on two-dimensional surfaces. small 2014, DOI: 10.1002/smll.201401881
Figure 2. a) Relative average straight line velocity VSLrel and curvilinear velocity VCLrel of bovine sperm cells swimming through microtubes with different tube diameters. The filled area between the two curves marks the range of microtube diameter in which the straightening effect occurs. b) Linearity LIN of spermatozoa as function of microtube diameter. The linearity is calculated as the ratio of straight line velocity and curvilinear velocity. The red dashed line represents the average linearity of spermatozoa swimming on planar surface. X error bars are range of tube diameters, Y error bars are standard error of the mean.
It was calculated from tracked cell paths that the linearity of swimming patterns through the microtubes decreases with increasing tube diameter, as shown in Figure 2b. At a diameter of about 40 µm, the linearity inside tubes corresponds to the average linearity observed for cells swimming on planar surfaces. Hence, the influence of the tubular confinement on the linearity is lost at tube diameters larger than 40 µm; in smaller microtubes an improved straightness of path is observed. This confirms the observation that the tubular confinement has a guiding effect on the spermatozoa mainly in tubular channels of dimension 20–45 µm in diameter. In the next step, the directionality, as defined by Paxton[27] was determined from the recorded cell trajectories. Directionality is defined as cos (θ), where θ is the angle between sperm
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
3
communications
V. Magdanz et al.
Figure 3. Directionality cos (θ) of cells moving through tubes of 12, 17, 20 and 30 µm tube diameter over time. The green background marks the point in time when the cells leave the tube which is characterized by the sharp drop in all 4 data sets. Directionality is defined as cos (θ), where θ is the angle between sperm head axis and directionality vector. The inset shows two rolled up silicon oxide microtubes with diameter 25 µm (top) and 35 µm (bottom). Scale bar: 50 µm.
head axis and directionality vector. The graph in Figure 3 shows the directionality of sperm cell movement over time when swimming through different microtubes. It illustrates how different diameters of microtubes (12 – 30 µm) determine the directionality. An average directionality of 0.96 ± 0.09 inside microtubes was derived. This value of directionality is much closer to 1 (meaning straighter cell motion) than that of a bull spermatozoon swimming on a planar surface (average directionality outside tubes = 0.80 ± 0.15). At 2.4 s in the graph in Figure 3, when the cell leaves the tube, the directionality abruptly drops. Additionally, the directionality inside microtubes increases with decreasing tube diameter due to the channeling effect of the surrounding walls that straighten the path of the spermatozoa. With increasing tube diameter from 17 to 30 µm, the average directionality slightly decreases from 0.99 to 0.96. The increasing tube diameter allows the cells to freely follow their naturally bending motion and hence decreases their directionality. The inset of Figure 3 shows two 250 µm long transparent microtubes with 25 µm diameter (top) and 35 µm diameter (bottom), respectively. The on-chip guidance of sperm cells inside microtubes offers an improvement in directionality and linearity as well as an increased velocity in microtubes with diameters of 20–45 µm. However, in microtubes narrower than 20 µm the cell speed is reduced due to the collisions of sperm head and flagellum with the tubular walls. The microtubes can serve as a tool for predetermining cell swimming paths for applications in lab-on-a-chip devices. We demonstrated that transparent silicon oxide microtubes are a very useful system to study motion of motile cells in cylindrical confinement. The microtubes exhibit excellent optical properties for visualizing and tracking of single cells. From trajectory measurements we can conclude that the motion of motile cells inside a tube on-chip shows a higher
4 www.small-journal.com
directionality and linearity than the motion of cells swimming unconfined on a glass surface. This confirms observations made by Tung et al.[12] that sperm cells in microgrooves have an improved directional persistence. The cell velocity also directly correlates with the tube diameter. At a microtube diameter of 20 µm and larger, the cells reach >100% of their initial straight line velocity outside the tube. The guidance of sperm cells inside microtubes offers a great way to direct the motion of sperm cells on-chip. It is an alternative to chemotactically or thermotactically control spermatozoa movement which has been a very challenging goal because the chemotactic behavior of sperm cells is poorly understood. An interesting observation during this study is that straight line velocities of sperm cells inside microtubes of diameters 20–45 µm can be higher than the velocity of cells swimming on a two-dimensional surface. The tube walls have a “straightening” effect on the trajectories of the motile cells. This is an indication that the sperm cell motion in confinement, as it can be found in vivo in cervical crypts or folds of ampullary fallopian tubes, significantly differs from the motion on 2D flat surfaces. The tubular microchannels could serve as a tool to more realistically investigate or even mimic this motion. There are no existing studies on the influence of surface roughness on sperm motion, to the best of our knowledge. Therefore, it would be an interesting topic for future studies and one could imagine creating tubular structures with different inner roughness grades and studying the influence of this surface effect on sperm motion. The application of these transparent microtubes can be further extended by integrating sensors for on-line cell analysis which is of high interest for the selection and separation of the most fertile spermatozoa for in vitro fertilization procedures. Up to our knowledge, it is not proven that the motility of cells is evidence alone for the most fertile sperm cells.[28] Therefore, a device with tubular channels that is able to integrate the selection of progressively moving cells with further analysis of metabolic features that hint to high fertilization capability would be promoting a high-quality sperm selection method. This approach confirms the assumption that microenvironment is important for sperm migration and provides information for further chip development that mimics natural environment and possibly provides a sperm selection mechanism similar to how it occurs in vivo. Sperm migration in vivo is a very complex process influenced by chemical and mechanical cues, morphology of the surroundings, sperm membrane modifications, pH, temperature, fluid flow and possibly many other factors.[29–33] Further research is required to define the role of each influencing factor and to reveal the mechanisms by which sperm travel through the female reproductive tract and perform successful fertilization.
Experimental Section Fabrication of Microtubes: The fabrication of the microtube samples followed established procedures that are in detail described elsewhere.[17,18,34] Briefly, rectangular photoresist patterns (100 µm in width, 100–500 µm in length) were photolithographically structured on 18 × 18 mm glass cover slides. In an electron beam evaporation system first a 5 nm thin layer of
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2014, DOI: 10.1002/smll.201401881
Sperm Dynamics in Tubular Confinement
silicon monoxide (SiO) and then a 60–100 nm thick layer of silicon dioxide (SiO2) were deposited on the photoresist patterns with a glancing angle of 30°. The layer thickness ratio of SiO to SiO2 later on defined the diameter of the microtubes. Because of the shadowing effect during the angular deposition, an uncovered region behind the rectangular structures was maintained. Through this window the photoresist sacrificial layer was in the next step dissolved in DMSO. The strain-engineered SiO/SiO2 bilayer film self-assembled into micron-sized tubes upon release from the photoresist structures. In order to stabilize the microtubes, the sample surfaces were covered with 18 nm of aluminium oxide in a subsequent atomic layer deposition step. Spermatozoa Treatment and Analysis: Cryopreserved bovine semen straws were thawed for 10 minutes at 37 °C. The 200 µL of bovine semen were diluted in 2 mL SP-TALP (modified Tyrode’s Albumin Lactate Pyruvate Medium) and incubated for additional 10 minutes. The array with SiO/SiO2 microtubes was immersed in the sperm cell solution. For the analysis of spermatozoa movement an inverted Zeiss microscope (10× and 20× objectives) was used with an attached Miro eX2 high speed camera from Vision Research. Videos were recorded at 20 frames per second and the software Phantom MulitCam was used for the image acquisition. The velocity measurements and tracking was done by the use of the ImageJ plugin Manual Tracking.
Supporting Information Supporting Information is available from the Wiley Online Library or from the author.
Acknowledgements We thank Dr. R. Träger for design of schematics in Figure 1a and the Table of Content graphic. We thank Masterrind GmbH in Meißen for donating frozen bull semen and Dr. B. Eichler for AFM measurements. The results leading to this work have been financially supported by the Volkswagen Foundation (86 362), the German Science Foundation (DFG) through the priority program SPP1726 “Microswimmers: From Single Particle Motion to Collective Behaviour” and the European Research Council (ERC) for Starting Grant “Lab-in-a-tube and Nanorobotics biosensors; LT-NRBS” [no. 311529].
[1] W. J. Polacheck, I. K. Zervantonakis, R. D. Kamm, Cell. Mol. Life Sci. 2013, 70, 8, 1335–1356. [2] P. Chiarugi, Cell Commun. Signal. 2010, 8, 25. [3] J. Brabek, C. T. Mierke, D. Roesel, P. Vesely, B. Fabry, Cell Commun. Signal. 2010, 8, 22. [4] P. Denissenko, V. Kantsler, D. J. Smith, J. Kirkman-Brown, J. Proc. Nat. Ac. Soc. 2012, 109, 21, 8007–8010.
small 2014, DOI: 10.1002/smll.201401881
[5] K. H. K. Wong, J. G. Truslow, A. H. Khankhel, K. L. S. Chan, J. Tien, J. Biomed. Mater. Res. 101A 2013, 8, 2181–219. [6] Y. M. Romanova, T. A. Smirnova, A. L. Andreev, T. S. Il’Ina, L. V. Didenko, A. L. Gintsburg, Biology 2006, 75, 4, 481–485. [7] L. Hall-Stoodley, W. Costerton, P. Stoodely, Nature Rev. Microbiol. 2004, 2, 95–108. [8] S. S. Suarez, Exp. Mech. 2010, 50, 1267–1274. [9] S. Tasoglu, H. Safaee, X. Zhang, J. L. Kingsley, P. N. Catalano, U. A. Gurkan, A. Nureddin, E. Kayaalp, R. M. Anchan, R. L. Maas, E. Tuezel , U. Demirci, Small 2013, 9, 20. [10] L. J. Kricka, O. Nosaki, S. Heyner, W. T. Garside, P. Wilding, Clin. Chem. 1993, 39, 9, 1944–1947. [11] L. Xie, R. Ma, C. Han, K. Su, Q. Zhang, T. Qiu, L. Wang , G. Huang, J. Qiao, J. Wang, J. Cheng, Clin. Chem. 2010, 56, 8. [12] C.-K. Tung, F. Ardon, A. G. Fiore, S. S. Suarez, M. Wu, Lab Chip 2014, 14, 1348. [13] Y.-A. Chen, K.-C. Chen, V. F. S. Tsai, Z.-W. Huang, J.-T. Hsieh, A. M. Wo, Clin. Chem. 2013, 59, 3. [14] Y.-J. Ko, J.-H. Maeng, B.-C. Lee, S. Lee, S. Hwang, Y. Ahn, Analyt. Sci 2012, 28, 27. [15] C.-Y. Chen, T.-C. Chiang, C.-M. Lin, S.-S. Lin, D.-S. Jong, V. F.-S. Tsai, J.-T. Hsieh, A. M. Wo, Analyst 2013, 138, 4967–4974. [16] J.-P. Frimat, M. Bronkhorst, B. de Wagenaar, J. G. Bomer, F. van der Heijden, A. van den Berg, L. I. Segerink, Lab Chip 2014, 00, 1, 1–3. [17] W. Xi, C. K. Schmidt, S. Sanchez, D. H. Gracias, R. E. Carazo-Salas, S. P. Jackson, O. G. Schmidt, Nano Lett. 2014, DOI: 10.1021/ nl4042565. [18] B. Koch, S. Sanchez, C. K. Schmidt, A. Swiersy, S. P. Jackson, O. G. Schmidt, Adv. Healthcare Mater. 2014, DOI: 10.1002/ adhm.201300678. [19] Y. F. Mei, A. A. Solovev, S. Sanchez, O. G. Schmidt, Chem. Soc. Rev. 2011, 40, 2109. [20] V. Magdanz, S. Sanchez, O. G. Schmidt, Adv. Mater. 2013, 25, 6581. [21] S. M. Harazim, V. A. Bolaños Quiñones, S. Kiravittaya, S. Sanchez, O. G. Schmidt, Lab Chip 2012, 12, 2649. [22] J. Deng, C. Yan, L. Yang, S. Baunack, S. Oswald, H. Wendrock, Y. F. Mei, O. G. Schmidt, ACS Nano 2013, 7, 8, 6948. [23] J. Deng, H. Ji, C. Yan, J. Zhang, W. Si, S. Baunack, S. Oswald, Y. F. Mei, O. G. Schmidt, Angew. Chem. Int. Edit. 2013, 52, 2326. [24] G. Huang, Y. F. Mei, D. J. Thurmer, A. Coric, O. G. Schmidt, Lab Chip 2009, 9, 263–268. [25] B. Robaire, B. T. Hinton, The Epididymis:from molecules to clinical practice, Orgebin-Crist, Springer, NY, USA 2002. [26] R.Rikmenspoel, J. Exp. Biol. 1984, 108, 205–230. [27] W. F. Paxton, K. C. Kistler, C. C. Olmeda, A. Sen, S. K. St Angelo, Y. Cao, T. E. Mallouk, P. E. Lammert, V. H. Crespi, J. Am. Chem. Soc. 2004, 126, 41, 13424–31. [28] J. E. Ellington, D. P. Evenson, R. W. Wright Jr., A. E. Jones, C. S. Schneide, G. A. Hiss, R. S. Brísbois, Fertil Steril 1999, 71, 5, 924–929. [29] M. Eisenbach, L. C. Giojalas, Nat Rev Mol. Cell. Bio. 2006, 7, 276–285. [30] J. Cosson, P. Huitorel, C. Gagnon, Cell Motil. Cytoskel. 2003, 54, 56–63. [31] M. A. Scott, Anim. Reprod. Sci. 2000, 60, 61, 337–348. [32] K. Miki, D. E. Clapham, Curr. Biol. 2013, 23, 6, 443–452. [33] A. Contri, A. Gloria, D. Robbe, C. Valorz, L. Wegher, A. Carluccio, Anim. Reprod. Sci. 2013, 136, 4, 252–259. [34] S. M. Harazim, W. Xi, C. K. Schmidt, S. Sanchez, O. G. Schmidt, J. Mater. Chem. 2012, 22, 7, 2878–2884. Received: June 27, 2014 Revised: September 15, 2014 Published online:
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
5
