Application of femtosecond laser microsurgery in assisted reproductive technologies for preimplantation embryo tagging

: Femtosecond laser pulses were applied for precise alphanumeric code engraving on the zona pellucida (ZP) of mouse zygotes for individual embryo marking and their identification. The optimal range of laser pulse energies required for safe ZP microsurgery has been determined. ZP was marked with codes in three different planes to simplify the process of embryo identification. No decrease in developmental rates and no morphological changes of embryos post laser-assisted engraving have been observed. ZP thickness of embryos post laser-assisted code engraving has been shown to differ significantly from that of control group embryos at the hatching stage. Due to moderate ZP thinning as compared to its initial width at 0.5 dpc (days post coitum), readability of the code degrades slightly and it still remains recognizable even at hatching stage. Our results demonstrate that application of femtosecond laser radiation could be an effective approach for noninvasive direct embryo tagging, enabling embryo identification for the whole period of preimplantation development.


Introduction
Lasers have become an efficient tool in assisted reproductive technologies (ART) [1][2][3]. Various laser sources are extensively employed for oocytes as well as for spermatozoa treatment and manipulations. Thus, for example, possibility of using a non-contact infrared diode laser (wavelength λ = 1.48 μm) for spermatozoa immobilization and permeabilization of the sperm membrane has been demonstrated in [4]. Low-power He-Ne laser has been used in in vitro fertilization for immature oocytes treatment and improving the system of in vitro embryo production [5,6]. Although first studies regarding spermatozoa movement stimulation have been conducted in 1980s, they are still underway. Earlier studies of Sato et al [7] and Lenzi [8] have demonstrated the stimulating effect of red (λ = 647 nm) and infrared (parameters of laser light were not clarified) laser light on sperm motility. Today, main attempts are made to develop methodology (usually based on optical tweezers) for safe and exact measurement of stimulating effects of laser light [9][10][11]. Optical tweezers have been successfully applied not only for sperm motility measurements, but also for sperm trapping and insertion into the perivitelline space of oocytes for in vitro fertilization [12], and for noncontact removal of polar bodies for their genetic analysis [13,14] (the so-called embryo biopsy). Openings in the outer shell surrounding oocytes and embryos required for sperm insertion during in vitro fertilization [12,15] or cell extraction during embryo biopsy can also be created by means of lasers [13].
Nowadays, infrared diode lasers (λ = 1.48 um) with milli-to microsecond pulse durations are the most popular lasers applied in the field of assisted reproduction for microdissection. Such systems are widely used for opening the ZP in assisted hatching [16][17][18][19][20]. Efficacy of various types of laser assisted hatching (LAH), for example, partial, quarter and total LAH has been analyzed [21]. The main danger of pointing a laser at an embryo is thermal damage. Infrared diode lasers seem to be an effective and safe tool, nevertheless strong recommendations regarding optimum regimes of embryo exposure should be taken into account to minimize possible laser-related thermal risks [22][23][24]. According to this, application area of infrared diode lasers is commonly limited to ZP dissection and spermatozoa immobilizing prior to use.
Recently, new approaches to assisted reproduction problems based on a novel, more delicate and effective laser systems generating laser pulses with shorter durations have been proposed. Femtosecond lasers have proven to be an excellent tool for noninvasive and precise microsurgery at cellular and even subcellular levels, for micromanipulation and optical modification of living biological objects. Femtosecond lasers have been successfully applied for fully noncontact optical microinjection and trapping of developing embryos [25], for oocyte enucleation by automated ablation of entire metaphase plates in porcine oocytes [26], and for blastomere fusion [27,28]. Efficacy of femtosecond laser use for laser-assisted hatching [29], noncontact polar body [30], and trophectoderm biopsy [31] (by simultaneous use of femtosecond laser and optical tweezers) has been previously shown by our group. We also used the unique ability of femtosecond lasers to perform precise microdissections to develop a novel technique for individual labeling of preimplantation embryos [32]. The technique is based on femtosecond laser microsurgery of ZP and "engraving" small (~5 µm in width and ~20 µm in length) alphanumeric codes in the depth of ZP. This technique may be useful in assisted reproductive technologies for preventing medical accidents relating to mixups. Although such errors are rare, several cases of mix-up in IVF laboratories have been reported [33][34][35][36] and various strategies, safety polices, and devices aimed at eliminating the risk of mistakes during the entire ART procedure are still being developed. By using femtosecond laser pulses relatively fast, precise, and delicate microsurgery can be performed with a minimal risk of thermal damage. The process of laser code engraving is performed in a contactless mode under sterile conditions and can be fully automated in the future. Moreover, only the ZP is subjected to laser microsurgery while leaving the embryo cells intact.
In our previous study [32] we performed one-plane code engraving (the code on ZP was created in a single, usually equatorial plane) as well as three-plane code engraving on 0.5 dpc (days post coitum) mouse embryos. No detrimental effects of laser-assisted code engraving on embryo developmental and hatching rates as well as on trophectoderm-to-inner cell mass ratio as compared to control group embryos have been observed. We have demonstrated that code created on the ZP could be clearly visualized at least until 3.5 dpc that allowed successful utilization of such technique for embryo identification from the day 0.5 to day 3 when embryo transfer could be done. However, the questions about code visualization at later stages of preimplantation embryo development and embryo identification during the entire preimplantation period were left unanswered. Our current study aims to answer these questions and provide interesting observations regarding features of ZP subjected to laser engraving and embryo hatching. Our observations allow us to suppose that the technique proposed may be used not only for embryo labelling but also for stimulating embryo hatching to start at prescribed location.

Experimental design
The femtosecond laser-based system for embryo microsurgery shown in Fig. 1 is based on our previously reported setup [32]. A femtosecond ytterbium 1028 nm wavelength laser 1 (TETA, Avesta LLC) that operates at 280 fs with repetition rate of 2.5 kHz was used. After power attenuation (beam attenuator 2 consists of half-wave plate and prism polarizer), the beam was sent through a second-harmonic generator 3. We employed second-harmonic radiation (514 nm wavelength) to perform microdissections on ZP in the form of arbitrary alphanumeric magnification through an e (Olympus) an (Olympus). T experimental a speed of ~1 videos were t (NF514-17, T radiation.

Embryo
Basic princip details in [32] by simply dra live image of commands to radiation "tur and post over on its ZP is sh Figs. 3D-F). As far as marking the e searching and As soon as c embryos (typi Then the proc dish was shak the procedure

Analysis
We have prev also clearly re with codes (3 Ref [32].  Fig. 4(B) was performed) ual to 7.08 ± 0. for engraving a e considered to P thickness betwee stage. ryo identificat n Fig. 5. At fir o ZP at 0.5 dp hat the code "V eas newly crea demonstrates cond plane) is of control gro C). All of the t (81.8% and 77 ngly) at the tim ol group (≈80. laser engravin rom the day 0. e readability e to ZP thinnin ment of 20 em compared with e embryos was found that the t m that of untre measured at 5 same procedur he experimenta e difference in onstrated in Fig  .30  al stage using ted as "1" in F s repeated for t Fig. 5(A)) is sli d as "2") is in f yo hatching. T ess, it is slight in control developed plane and 5-5.0 dpc) han in the wn to be ansfer can e at late r to verify ane laserup (a total measured d to lasergroups B value and rol group 84 ± 0.60 or control ckness of 5 μm for hen laserental and In case of een them, ed even at p the code Fig. 5(A)) two other ightly out focus and The code tly out of The code "VIVO" indicated as "2" is engraved in the second plane. (C) ZP after embryo hatching; the code "2" is still readable.
The second peculiar fact is that code marking in multiple planes did not decrease embryo hatching rate compared to control groups. Moreover, code engraving procedure similarly to assisted hatching can facilitate hatching to start right at the location of laser treatment. To demonstrate this, an angle between the code "1" and code "2" shown in Fig. 5(B) in the ZP of embryo at 0.5 dpc was measured. This angle equal to 107° is also shown in Fig. 5(C). One can see that hatching occurred in the rear part of ZP exactly where the code "1" had been engraved. This was not a single case, but determining the exact rate of embryo hatching through the code in the ZP was beyond the framework of current research; detailed studying of this phenomenon is a matter of further investigations.

Discussion and conclusions
Although events of eggs, sperm or oocytes switching occur very rare at fertility clinics, sample mix-up has been reported in the literature several times. According to the latest research, 90.4% of the respondents (patients undergoing IVF treatment in a single private infertility center in Europe) expressed significant concerns relating to biological sample mixup [36]. To eliminate the risk of any mix-up, strong recommendations and protocols by leading ART-related organizations (ESHRE and HFEA (Europe), FLASEF (Latin America) have been developed [38,39]. According to their guidelines, accurate labelling of all labware for correct patient identification and "double-witnessing" procedure are mandatory. Recently, novel electronic witness systems have been developed. Systems based on Radio Frequency Identification technology [40,41], barcode labels [42], and even direct embryo tagging system based on silicon barcode injection into zygotes/embryos [43] have been proposed. However, nearly all of these approaches have some limitations. Thus, for example, additional equipment is required (such as label printer or code reader) when using safety systems based on QR (quick response) code generation and recognition [44]; volatile organic components in the printing and adhesive materials should be thoroughly selected so as not to be toxic to embryo development [44]. Moreover, possible effects of polysilicon barcodes proposed in [43] on fetal growth and development should be studied in future. Thus, optimisation of existing methods aimed at preventing biological sample mix-up and development of new alternative devices and techniques are still required.
In this paper the possibility of direct embryo tagging with femtosecond laser microsurgery as well as possibility of embryo identification during the whole preimplantation period have been demonstrated. Due to highly localized effect during the action of fs-laser pulses, which fades away for out-of-focus cellular structures, relatively low pulse energy, and ultrahigh intensity, fs-lasers could be used for precise and delicate microsurgery of ZP with minimal risk of thermal damage to the adjacent embryo cells. The advantages of femtosecond laserassisted microsurgery over milli-/microsecond or even nano-/picosecond duration pulses for minimizing collateral damage have been discussed in [45][46][47]. We have performed laserassisted engraving of alphanumeric codes on mouse embryo's ZP in three different planes in order to simplify the process of code searching and embryo identification. The codes engraved have been proven to be readable even after embryo hatching. No morphological changes of embryos subjected to three-plane laser-assisted engraving as compared to control group embryos have been observed. We have demonstrated that the average ZP thickness of embryos in experimental group was larger than ZP thickness of control group embryos. In spite of this fact, the blastocysts broke out of the shell successfully and hatching rates in the experimental and control groups embryos were nearly the same. A possible explanation for this is that formation of cuts on the ZP during laser-assisted code engraving leads to a weakening of the ZP stiffness and its easier rupture with no need for significant thinning.
In this study femtosecond laser pulses have been successfully applied for ZP microsurgery of mouse embryos. The thickness of the ZP at the time of code engraving (0.5 dpc) was measured to be ~7 µm. It was enough for creating high-quality, clearly readable codes. ZP of human embryos is usually wider than that of mouse embryos. Data regarding typical thickness of ZP of mouse and human embryos at various days of in vitro culture are summarized in Table 1. As can be seen, the thickness of the human embryo ZP usually lies within the range of 14 -18 µm. Taking into account relatively higher ratio of ZP width to the size of focused laser beam in human embryo as compared to the mouse one, we suppose that the technique of femtosecond laser-assisted code engraving will be easier to implement for human embryo labelling as compared to mouse one. In conclusion, we have demonstrated that femtosecond lasers could be employed as precise and effective tools for embryo microsurgery. Potential applications for laser-assisted code engraving technique would not be limited to safety systems aimed at preventing embryo mix-ups during the IVF treatment. The technique may be also useful in the field of developmental biology for studying the peculiarities of embryo development during their culture in groups.

Disclosures
The authors declare that there are no conflicts of interest related to this article.