Establishment and Optimization of Molecular Cytogenetic Techniques (45S Rdna-FISH, GISH And Fiber-FISH) In Kiwifruit (Actinidia L.)

Background: Kiwifruit has long been regarded as ‘the king of fruits’ for its nutritional importance. However, the molecular cytogenetics of kiwifruit has long been hampered because of the large number of basic chromosome (x=29), the inherent small size and highly similar morphology of metaphase chromosomes. Fluorescence in situ hybridization (FISH) is an indispensable molecular cytogenetic technique widely used in many plant species. Herein, the effects of post-hybridization washing temperature on FISH, blocking DNA concentration on genomic in situ hybridization (GISH), extraction method on nuclei isolation and the incubation time on the DNA ber quality in kiwifruit were evaluated. Results: The post-hybridization washing in 2×SSC solution for 3×5 min at 37 ˚C ensured high stringency and distinct specic FISH signals in kiwifruit somatic chromosomes. The use of 50× blocking DNA provided an ecient and reliable means of discriminating between chromosomes derived from in the hybrids of A. chinensis var. chinensis (2n=2x=58) × A. eriantha Benth (2n=2x=58), and inferring the participation of parental genitors. The chopping method established in the present study were found to be very suitable for preparation of leaf nuclei in kiwifruit. A high-quality linear DNA ber was achieved by an incubation of 20 min. The physical size of 45S rDNA signals was approximately 35-40 μmm revealed by the highly reproducible ber-FISH procedures established and optimized in this study. Conclusions: The molecular cytogenetic techniques (45S rDNA-FISH, GISH, and high-resolution ber-FISH) for kiwifruit was for the rst time established and optimized in the present study, which is the foundation for the future genomic and evolutionary studies. until Root apices were in 4.0% (w/v) cellulase and 0.4% (w/v) pectolyase min hypotonic The root meristems were squashed drop prepared on a clean and dried on a ame. Slides featuring good-quality metaphase chromosomes were kept at until further application of molecular cytogenetic techniques.

representation of molecular biology and cytogenetics, and involves the number, structure, function and behavior of mitotic and meiotic chromosomes, chromosome recombination and transmission, and the physical organization of certain DNA sequences [9]. The advent of uorescence in situ hybridization (FISH) almost 40 years ago [10] marked the beginning of a new era for molecular cytogenetics [11], and has become an indispensable technique for chromosome identi cation [12,13] and genome sequencing in plant species [14].
The advancements in availability of genomic resources and degree of resolution, such as genomic in situ hybridization (GISH) and ber-FISH have widely widened the scope of FISH applications [13], which now range from karyotype characterization to integration of genetic linkage maps with chromosomal maps [12]. Using the total genomic DNA as probe and blocking DNA, GISH is a modi cation of FISH, and it enables to distinguish chromosomes from different genomes in an intact cell [13,15]. As a straightforward technique, GISH has been widely applied to the study of chromosomal evaluation, cytogenetical classi cation, genomic constitution, polyploidy con rmation, hybrid veri cation, introgression breeding in horticultural crops [15]. Fiber-FISH allows high-resolution mapping of the repetitive DNA sequences, large and complex genomic loci, and the cloned and organelle DNA molecules on DNA and chromatin bers [11,16]. The application of FISH on extended DNA and chromatin bers allows the physical mapping of individual genes or other small DNA molecules at a resolution of 1-400 kb [17], with 1 kb corresponding to ca. 340 nm on a completely relaxed DNA double helix [18]. The advantage of high resolution of ber-FISH has thus attracted considerable interest in molecular cytogenetics of different groups of species [11][12][13].
In the present study, a molecular cytogenetic study was conducted to kiwifruit by means of applying FISH physical mapping of 45S ribosomal DNA (rDNA) sites. In addition, genomic identi cation of the kiwifruit hybrids was performed by GISH. The high-resolution ber-FISH technique for kiwifruit was also developed and optimized. The establishment and optimization of 45S rDNA-FISH, GISH and ber-FISH techniques in kiwifruit herein would make the detailed FISH-based karyotypes of Actinidia possible and serve as an important molecular cytogenetic basis for future genomic and evolutionary studies.

Results And Discussion
Effect of post-hybridization washing temperature on uorescence in situ hybridization in kiwifruit After the probe hybridized to the chromosomal DNA, post-hybridization washing is an essential step of FISH protocol to remove the excess unbound or loosely bound probes and separate the non-speci c hybrid signals [25]. Post-hybridization washes associated parameters including temperature, salt and detergent solution concentration can be manipulated to remove non-speci c interactions [25]. In this study, we tested three different temperatures (35, 37, and 42 ˚C) of post-hybridization washes in a 2×SSC solution (Fig. 1). The salt solution concentration was empirically selected [22].
Stringent post-hybridization washing in 2×SSC solution for 3×5 min at 37 ˚C ensured the removal of unbound FISH signals and speci city of the detected 45S rDNA signals ( Fig. 1a-c). Under low posthybridization washing temperature (35 ˚C), some non-speci c FISH signals, excessive background auto uorescence, and some unscorable green signals diminished the visibility of true FISH signals of 45S rDNA ( Fig. 1d-f). There existed some generalized background on slides but not at chromosome regions ( Fig. 1d-f). The hybridized probes appeared faint and some of them were even washed away under high post-hybridization washing temperature (42 ˚C) ( Fig. 1g-h).
Optimization of post-hybridization washing temperature is important for achieving optimal hybridization conditions, though several factors affected the e ciency and quality of hybridization [26]. It is recognized that low temperature can lead to inadequate stringency of post-hybridization washing conditions, while high temperature lead to excessive stringency of posti-hybridization washing conditions [27]. Thus, the possible reasons for the non-speci c FISH signals ( Fig. 1d-f) and the loss of signals ( Fig. 1g-h) were assumed to be the insu cient and excessive stringency of post-hybridization conditions, respectively. By evaluating the effects of different post-hybridization washing temperatures on FISH results as presented in Fig. 1, we demonstrated that the post-hybridization washing in 2×SSC solution for 3×5 min at 37 ˚C ensured high stringency and distinct speci c FISH signals in kiwifruit somatic chromosomes.
Effect of blocking DNA concentration on genomic in situ hybridization in kiwifruit GISH is a variation of FISH and has been widely used to distinguish parental chromosomes or chromosome segments [13,15]. The utilization of fragmented genomic DNA as probe and non-target genome as blocking DNA in GISH differentiate it from FISH analysis [15]. Higher plant genomes are composed of high proportion of repetitive DNA families and considered highly conserved in plants [28]. Therefore, genome discrimination by GISH often meets considerable complications resulting from the existence of the highly conserved repetitive DNA sequences emerged during the long-term evolution events. GISH works primarily on hybridization of these repetitive DNA sequences [29]. In this case a blocking DNA serving as a DNA competitor to avoid the staining of both genomes by the probe DNA is important in hybrids in particular derived from closely related parental genotypes [29].
In the current study, seven different concentrations of blocking DNA were compared to determine what the ratio of probe/blocking DNA are su cient to inhibit the chromosome labeling of both parental genomes together. The blocking DNA in a concentration 50× higher than that of the labeled maternal genomic DNA probe (2 ng/µl) allows e cient hybridization and discrimination of the parental chromosome sets (Fig. 2a). The use of lower concentrations (less than 50× than that of the labeled paternal DNA probe) of the blocking DNA did not exhibit good results and not enable unambiguous differentiation of the two parental chromosome sets (Fig. 2b). High concentration (over 50 times) of blocking DNA, however, did not generate hybridization signals, probably because of the high similarities between repetitive DNA sequences that are common in the parental genomes [29,30]. The absence of blocking DNA resulted in substantial hybridization sites on whole chromosome sets (Fig. 2c), revealing the need of using blocking DNA. Hence, the use of 50× blocking DNA is an option to GISH protocol for kiwifruit.
Herein, we described an optimal blocking DNA concentration for obtaining reliable and informative GISH results, though the paternal genitors are closely related with morphologically similar and relatively small chromosome sets. In the present study, the GISH technique was rst applied to Actinidia L. Our results show that GISH will probably provide an e cient and reliable means of discriminating between chromosomes derived from in the hybrids of A. chinensis var. chinensis (2n = 2x = 58) × A. eriantha Benth (2n = 2x = 58), and inferring the participation of parental genitors in the karyotypic constitution of in terspeci c hybrids in the future.
An optimal ber-FISH in physical mapping of 45S rDNA of Actinidia L.
FISH performed onto extend DNA bers released from nuclei has provided a valuable high-resolution tool for physical mapping of up to a few kilobases [31][32][33]. For example, [33] and [23] reported the extended DNA bers of Arabidopsis thaliana as 3.27 and 2.87 kb/µm, respectively. Nuclei extraction is one of the most critical aspects of a Fiber-FISH protocol, and it cannot be overstressed that nely extended DNA ber preparations begin with good nuclei extraction [24]. In the present study, two different methods for isolation of nuclei by directly grinding and chopping fresh leaves in isolation buffer were compared to prepare extended DNA bers (Fig. 3). Fewer nuclei were destroyed using chopping with a blade (Fig. 3a) than grinding in liquid nitrogen (Fig. 3b). Most of the nuclei extracted by the chopping method remained intact (Fig. 3a), while there were more debris presented in the extraction by grinding in liquid nitrogen (Fig. 3b). The concentration of nuclei harvested by the chopping method established in the present study reached approximately 5×10 6 to 5×10 7 nuclei/ml (Fig. 3c). The chopping method established in the present study were found to be very suitable for preparation of leaf nuclei in kiwifruit.
A key element in ensuring ber-FISH data reproducibility is to obtain well-separated/stretched intact DNA bers [24]. One critical parameter to achieve the best possible stretching/spreading high-quality DNA bers is the incubation time of nuclei suspension on the glass slide as well as the lysis time [34,35]. To examine the effects of incubation time on the DNA ber quality, we tested different incubation times of 10, 15, 20, 25, and 30 min after adding the lysis buffer (Fig. 4).
From our experiment it is evident that 10 min of the incubation time after adding the lysis buffer proved to be insu cient to obtain extended DNA bers. Increasing the incubation time to 15 min resulted in, to a certain extent, extended DNA bers, but some of the nuclei have not been fully lysed (Fig. 4b). As illustrated in Fig. 4c, over 25 min of the incubation time proved to be excessive to obtain extended DNA ber because the DNA molecules formed a rosary-, like chain of DNA structures with low density. The possible reason for the poor-quality DNA ber was assumed to be DNA degradation. A high-quality linear DNA molecule was achieved by an incubation of 20 min, and the DNA molecule obtained was straightened and uniformly stretched as presented in Fig. 4d.
In addition, FISH analysis for 45S rDNA on extended DNA bers in kiwifruit was for the rst time performed in this study (Fig. 5). By applying the highly reproducible ber-FISH procedures in this study, it was estimated the physical size of 45S rDNA signals was approximately 35-40 µmm (Fig. 5). As shown in Fig. 5, the 45S rDNA signals appeared as typical beads-on-string pattern of green spots. With the aid of such high-resolution molecular cytogenetic technique established and optimized in the present study, it is possible to accurately chromosome identi cation, elucidation of evolutionary relationships and delineation of possible chromosomal variations of Actinidia L.
To our knowledge, Actinidia species have large number of basic chromosomes, relatively small chromosomes with highly similar morphology [7,8]; as a result, unequivocal discrimination of individual chromosomes based on traditional cytogenetic methods is rather challenging. Herein, the establishment and optimization of modern molecular cytogenetic techniques (45S rDNA-FISH, GISH, and high-resolution ber-FISH) will probably provide the necessary means, coupled with other molecular and bioinformatics approaches such as using repetitive DNA sequences as probes, to accurate chromosome identi cation (currently underway in our laboratory). The outcome presented here are the foundation for the continued fascinating research in the near future.

Conclusions
The molecular cytogenetics of Actinidia species have long been hampered because of the large number of basic chromosome (x = 29), the inherent small size and highly similar morphology of metaphase chromosomes. In the present study, the effect of post-hybridization washing temperature on FISH in kiwifruit was evaluated. The post-hybridization washing in 2×SSC solution for 3×5 min at 37 ˚C ensured high stringency and distinct speci c FISH signals in kiwifruit somatic chromosomes. GISH technique was rst applied to Actinidia and the use of 50× blocking DNA provided an e cient and reliable means of discriminating between chromosomes derived from in the hybrids of A. chinensis var. chinensis (2n = 2x = 58) × A. eriantha Benth (2n = 2x = 58), and inferring the participation of parental genitors. The chopping method established in the present study were found to be very suitable for preparation of leaf nuclei in kiwifruit. A high-quality linear DNA ber was achieved by an incubation of 20 min. The physical size of 45S rDNA signals was approximately 35-40 µmm revealed by the highly reproducible ber-FISH procedures established and optimized in this study. In conclusion, the molecular cytogenetic techniques (45S rDNA-FISH, GISH, and high-resolution ber-FISH) for kiwifruit established and optimized here are the foundation for the future genomic and evolutionary studies.

Materials And Methods
Plant materials and genomic DNA extraction Seeds of A. chinensis cv. Hongyang were germinated at room temperature on a moist lter paper in petri dishes. Root tips and fresh young leaves of the seedling of A. chinensis cv. Hongyang were used as source materials for the mitotic chromosome preparation of FISH and DNA ber prepariotn of ber-FISH, respectively. Root tips for the chromosome preparation of GISH were obtained from germinated seeds derived from the cross-hybridization between A. chinensis var. chinensis (2n = 2x = 58) × A. eriantha Benth (2n = 2x = 58). Parental genomic DNA was extracted from fresh young leaves using a DNeasy® Plant Mini Kit (Qiagen Inc., Valencia, CA, USA) following the manufacturer's instructions. DNA quality (A 260/280 and A 260/230 ratios) and concentration were assessed with a NanoDrop spectrophotometry (ThermoScienti c, Waltham, MA, USA). Sterile ultrapure water was ued to prepare solutions in this study.

Chromosome Preparation
Mitotic chromosomes were obtained following our previously reported protocol [22] with minor modi cations. Brie y, the actively growing root tips of approximately 0.5cm in length were pretreated with 2 mM 8-hydroxyquinoline (Sigma; dissolving 0.029 g 8-hydroxyquinoline in 100 ml water) for 2.5 h in the dark at room temperature (RM), then xed in a freshly prepared Carnoy's solution, composed of 75% ethanol and 25% glacial acetic acid, for a minimum of 3.0 h at RM and stored at -20 ˚C until further use. Root apices were softened in an enzyme solution consisting of 4.0% (w/v) cellulase and 0.4% (w/v) pectolyase Y-23 for 100 min at 37 ˚C, followed by a 20 min hypotonic treatment. The root meristems were squashed in a drop of freshly prepared Carnoy's solution on a clean pre-chilled slide and dried on a ame. Slides featuring good-quality metaphase chromosomes were kept at -20 ˚C until further application of molecular cytogenetic techniques.

Probe Labeling
Dr. Ruiyang Chen (Nankai University, Tianjin, China) is gratefully acknowledged for providing plasmid containing 45S rDNA. Genomic probes were prepared by sonication to 100-500 bp fragments and DNA size was checked using 1% agarose gel. Fragmented DNA (100 ng/µl) and plasmid harboring the 45S rDNA (100 ng/µl) were labeled with digoxigenin-16-dUTP by random-primed labeling method using DIG-High Prime DNA Labeling and Detection Starter Kit II (Roche, Mannheim, Germany) following the instruction manual supplied by the manufacturer. Blocking DNA was prepared by autoclaving the total genomic DNA for 5 min which fragmented it into approximately 200 bp.

Fluorescence in situ hybridization
Slide with cytological preparations for 45S rDNA-FISH and GISH was rstly dried at 60 ˚C in an oven for at a minimum of 3.0 h, treated with 40 µl of RNase solution (100 µg/ml in 2×SSC) and incubated in a humidi ed chamber (37 ˚C) for 1.0 h, followed by three times wash in 2×SSC at RT for 5min each. The slide was then treated with 0.01% (w/v) pepsin (Sigma) for 10 min at 37 ˚C, and washed twice in 1×PBS at RT for 5 min each. Following this, the slide was immersed in 1% formaldehyde (Sigma) at RT for 5 min, and then rinsed three times in 2×SSC at RT for 3 min each. The wash steps above and below were performed with a shaker platform set at 150 rpm.
The hybridization mix of FISH (20 µl for one slide), consisting of 10 µl deionized formamide (Sigma), 4 µl 50% dextran sulfate (Sigma), 2 µl 20×SSC (pH 7.0), 1 µl 10mg/ml sperm ssDNA, 2 ng/µl probe DNA, and ddH 2 O, were thoroughly mixed, denatured at 100 ˚C for 10 min in a thermocycler (Eppendorf Mastercycler) and immediately cooled it on ice for at least 10 min. The chromosomal DNA was denatured by putting slides in 70% (v/v) formamide (in 2×SSC) solution at 72 ˚C for 2 min. The slides were then dehydrated in a series of 70%, 95% and 100% ethanol at -20 ˚C for 5min in each solution, followed by air dry. Probe hybridization, signal detection and chromosome photo-documentation were in the same manner as previously reported [22]. Post-hybridization washes are essential to a FISH protocol to remove the non-speci c hybrid signals. Three temperatures (i. e. 35, 37, and 42 ˚C) of post-hybridization washes were compared.

Interphase Nuclei Extraction
Two different leaf nuclei extraction methods by grinding and chopping followed the methods of [23] and [24], respectively, with some modi cations. Two grams of fresh leaf tissue were pooled together and ground to a ne powder in liquid nitrogen. Then, 80 mg powder were transferred to a clean 2-ml tube with 2 ml of precooled nuclei isolation buffer containing 10 mM MgSO 4 , 5 mM KCl, 0.5 mM HEPES, 1 mg/ml DTT, and 0.25% (v/v) Trition X-100. The samples were then gently shaken on ice for a homogenized stage. The resulting suspension were ltered sequentially via 60 and 30-µm mesh nylon membranes, while on an ice-cold metal block. The ltrate was centrifuged at 2000×g for 2 min at 4 ˚C. The supernatant was removed and the nuclei pellet was suspended in two different volumes (30 and 60 µl) of nuclei store solution containing 10 mM MgSO 4 , 5 mM KCl, 0.5 mM HEPES, and 1 mg/ml DTT. The concentration of nuclei was determined by staining nuclei with 2-(4-amidinophenyl)-1H-indole-6carboxamidine (DAPI) (1 µg/ml in 1×PBS).
Instead of grinding leaves, fresh young leaf samples (80 mg) were nely chopped with a clean razor blade in a total of 2 ml cold nuclei isolation buffer in the second methods. The following procedure was in the same manner as described above. The nuclei in storage solution (10 µl) was mixed in an equal volume (10 µl) of DAPI solution, and analyzed using an Olympus BX61 epi uorescence microscope (Tokyo, Japan) eq uipped with a Sensys CCD camera (Qimaging Retiga™ SRV Fast 1394, Vancouver, BC, Canada).

Dna Fiber Extension And Fiber-sh
Extension of DNA bers followed the method of [23] with some modi cations. The nuclei suspension (1µl) was deposited in a line across one end of poly-L-lysine glass slide (Sigma) and left to air dry for approximately 5 min. Then, 30 µl of lysis buffer (pH 7.0) containing 0.5% (w/v) SDS, 5 mM EDTA, and 100 m M Tris was added on the top of the nuclei and incubated at room temperature for 10, 15, 20, 25, and 30 min. DNA bers were then extended by dragging the suspension of lysed nuclei down the slide slowly and smoothly with the edge of an 18 mm×18 mm coverslip. The spreading by gravity were not chosen because it can lead to a nonuniform spreading of biers that result in frequent crossing.
Slide was air-dried for approximately 15 min to a 'sticky' point, neither wet nor overdried. Slide was x in Carnoy's solution for 2 min, dried at 60 ˚C for 45 min and then can be used immediately for FISH or stored in 4 ˚C for 5-7 days. DNA ber quality was checked by staining with 5 µg/ml DAPI at RT for 20 min and brie y rinsing twice in distilled water, followed by air dry. After staining, the effect of incubation time on the DNA ber quality was observed under epi uorescence microscope. The ber-FISH procedures are the same as those in regular FISH protocol above.
Declarations performed the GISH experiment. ZH and KL participated in and performed the ber-FISH experiment. HD wrote and revised the manuscript. All authors read and approved the nal manuscript.

Funding
This work was nancially supported by the Fundamental Research Funds for the Central Universities (XDJK2018B038 and XDJK2020C076).

Availability of data and materials
All data generated or analysed during this study are included in this published article.
Ethics approval and consent to participate Not applicable Consent for publication Not applicable