Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jul 8;34(7):1173-1185.
doi: 10.1093/humrep/dez056.

Rapid Sperm Capture: High-Throughput Flagellar Waveform Analysis

Affiliations
Free PMC article

Rapid Sperm Capture: High-Throughput Flagellar Waveform Analysis

M T Gallagher et al. Hum Reprod. .
Free PMC article

Abstract

Study question: Can flagellar analyses be scaled up to provide automated tracking of motile sperm, and does knowledge of the flagellar waveform provide new insight not provided by routine head tracking?

Summary answer: High-throughput flagellar waveform tracking and analysis enable measurement of experimentally intractable quantities such as energy dissipation, disturbance of the surrounding medium and viscous stresses, which are not possible by tracking the sperm head alone.

What is known already: The clinical gold standard for sperm motility analysis comprises a manual analysis by a trained professional, with existing automated sperm diagnostics [computer-aided sperm analysis (CASA)] relying on tracking the sperm head and extrapolating measures. It is not currently possible with either of these approaches to track the sperm flagellar waveform for large numbers of cells in order to unlock the potential wealth of information enclosed within.

Study design, size, duration: The software tool in this manuscript has been developed to enable high-throughput, repeatable, accurate and verifiable analysis of the sperm flagellar beat.

Participants/materials, setting, methods: Using the software tool [Flagellar Analysis and Sperm Tracking (FAST)] described in this manuscript, we have analysed 176 experimental microscopy videos and have tracked the head and flagellum of 205 progressive cells in diluted semen (DSM), 119 progressive cells in a high-viscosity medium (HVM) and 42 stuck cells in a low-viscosity medium. Unscreened donors were recruited at Birmingham Women's and Children's NHS Foundation Trust after giving informed consent.

Main results and the role of chance: We describe fully automated tracking and analysis of flagellar movement for large cell numbers. The analysis is demonstrated on freely motile cells in low- and high-viscosity fluids and validated on published data of tethered cells undergoing pharmacological hyperactivation. Direct analysis of the flagellar beat reveals that the CASA measure 'beat cross frequency' does not measure beat frequency; attempting to fit a straight line between the two measures gives ${\mathrm{R}}^2$ values of 0.042 and 0.00054 for cells in DSM and HVM, respectively. A new measurement, track centroid speed, is validated as an accurate differentiator of progressive motility. Coupled with fluid mechanics codes, waveform data enable extraction of experimentally intractable quantities such as energy dissipation, disturbance of the surrounding medium and viscous stresses. We provide a powerful and accessible research tool, enabling connection of the mechanical activity of the sperm to its motility and effect on its environment.

Large scale data: The FAST software package and all documentation can be downloaded from www.flagellarCapture.com.

Limitations, reasons for caution: The FAST software package has only been tested for use with negative phase contrast microscopy. Other imaging modalities, with bright cells on a dark background, have not been tested but may work. FAST is not designed to analyse raw semen; it is specifically for precise analysis of flagellar kinematics, as that is the promising area for computer use. Flagellar capture will always require that cells are at a dilution where their paths do not frequently cross.

Wider implications of the findings: Combining tracked flagella with mathematical modelling has the potential to reveal new mechanistic insight. By providing the capability as a free-to-use software package, we hope that this ability to accurately quantify the flagellar waveform in large populations of motile cells will enable an abundant array of diagnostic, toxicological and therapeutic possibilities, as well as creating new opportunities for assessing and treating male subfertility.

Study funding/competing interest(s): M.T.G., G.C., J.C.K-B. and D.J.S. gratefully acknowledge funding from the Engineering and Physical Sciences Research Council, Healthcare Technologies Challenge Award (Rapid Sperm Capture EP/N021096/1). J.C.K-B. is funded by a National Institute of Health Research (NIHR) and Health Education England, Senior Clinical Lectureship Grant: The role of the human sperm in healthy live birth (NIHRDH-HCS SCL-2014-05-001). This article presents independent research funded in part by the NIHR and Health Education England. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. The data for experimental set (2) were funded through a Wellcome Trust-University of Birmingham Value in People Fellowship Bridging Award (E.H.O.).The authors declare no competing interests.

Keywords: computer-aided sperm analysis; flagellar tracking; fluid dynamics; high-throughput analysis; image analysis; mathematical modelling; sperm kinematics.

Figures

Figure 1
Figure 1
Tracking of a human sperm from experimental data set (1) in the DSM medium. Panels (a)–(d) show an overlay of the tracked flagellum over experimental frames at four points in a beat cycle, with a 5 μm white scale bar. Panel (e) shows the sperm head track in magenta with associated flagellum plotted 0.014 s apart. Panel (f) plots the flagellar beat over a single experimental frame with the colour of each flagellum representing time from dark blue to yellow. Panel (g) plots the tangent angle along the flagellum in the cell frame for 0.5 s. Panel (h) shows the power exerted by the flagellum on the fluid distal to the point in arclength chosen, with the power exerted by the full flagellum shown in panel (i).
Figure 2
Figure 2
Tracking of a human sperm from experimental data set (1) in the HVM. Panels (a)–(d) show an overlay of the tracked flagellum over experimental frames at four points in a beat cycle, with a 5 μm white scale bar. Panel (e) shows the sperm head track in magenta with associated flagellum plotted 0.007 s apart. Panel (f) plots the flagellar beat over a single experimental frame with the colour of each flagellum representing time from dark blue to yellow. Panel (g) plots the tangent angle along the flagellum in the cell frame for 0.35 s. Panel (h) shows the power exerted by the flagellum on the fluid distal to the point in arclength chosen, with the power exerted by the full flagellum shown in panel (i).
Figure 3
Figure 3
Tracking of stuck sperm from experimental data set (2) enabling long-time analysis of cells. Panels (a)–(d) show the flagellar beat, tangent angle, power exerted by the flagellum distal to a point in arclength and total power exerted by the flagellum, respectively. Panels (e)–(h) show the same plots for a hyperactivated cell after stimulation with 4AP.
Figure 4
Figure 4
Scatter plot matrix. Scatter plot matrix showing relationships between arc-wavelength λ, flagellar beat frequency f, power generated by the first 30 μm of flagellum P30 (measured in watts and plotted on a log scale), the curvilinear velocity of the head (VCL), the average path velocity of the head (VAP) and the beat cross frequency of the head (BCF). Axes persist from left to right and top to bottom except on the leading diagonal where frequencies are shown in green. In each plot, sperm swimming through HVM are shown as magenta crosses and sperm swimming through DSM are shown as green dots.
Figure 5
Figure 5
Simulated velocity fields with NEAREST for the tracked sperm in Figs 1 and 2. The FAST tracked flagellum for each sperm has been paired with an idealized head and simulated in a three-dimensional environment. Panels (a)–(d) show the flow fields in DSM at times corresponding to Fig. 1a–d, while (e)–(h) are in HVM, corresponding to Fig. 2a–d. In each figure the colour depicts the fluid velocity magnitude, with the sperm cell overlain in magenta and streamlines shown as white dots.
Figure 6
Figure 6
ROC curves for characterizing sperm as progressive compared to the gold standard manual classification by a trained analyst. In panel (a) the ROC curve using TCS is plotted, with the more standard use of VSL and VAP shown for comparison in panels (b) and (c). The green lines highlight the sensitivity and specificity when using the 5 μm/s WHO categorization (WHO, 1999).

Comment in

Similar articles

See all similar articles

Cited by 1 article

References

    1. Alasmari W, Costello S, Correia J, Oxenham S, Morris J, Fernandes L, Ramalho-Santos J, Kirkman-Brown J, Michelangeli F, Publicover S et al. Ca2+ signals generated by CatSper and Ca2+ stores regulate different behaviors in human sperm. J Biol Chem 2013;288:6248–6258. - PMC - PubMed
    1. Amann R, Waberski D. Computer-assisted sperm analysis (CASA): capabilities and potential developments. Theriogenology 2014;81:5–17. - PubMed
    1. Friedrich B, Riedel-Kruse I, Howard J, Jülicher F. High-precision tracking of sperm swimming fine structure provides strong test of resistive force theory. J Exp Biol 2010;213:1226–1234. - PubMed
    1. Gaffney E, Gadêlha H, Smith D, Blake J, Kirkman-Brown J. Mammalian sperm motility: observation and theory. Annu Rev Fluid Mech 2011;43:501–528.
    1. Gallagher M, Choudhuri D, Smith D. Sharp quadrature error bounds for the nearest-neighbor discretization of the regularized stokeslet boundary integral equation. SIAM J Sci Comput 2019;41:B139–B152.

Publication types

Feedback