Plant materials.

Protocorms of D. wardianum were obtained by seed germination (45 d). The seed capsules were harvested before dehiscence from a wild population (Tengchong County, Yunnan Province, China) approximately 7 months after natural pollination and sterilized (Miguel and Leonhardt 2011). The capsules were brushed carefully with detergent and rinsed under tap water; then, they were surface-sterilized by stirring in sodium hypochlorite solution (active chlorine 2%; RHAWN-R049945) containing 0.01% Tween-20 for 15 min. The sterilized capsules were washed three times with sterile distilled water. A photo of the D. wardianum flower is provided (Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Culture media.

The maintenance medium was half-strength Murashige and Skoog (MS; Solarbio-M8520) supplemented with 0.5 mg⋅L⁻¹ 3-indolebutyric acid, 2.0 mg⋅L⁻¹ 6-benzylaminopurine, 0.5 mg⋅L⁻¹ 1-naphthylacetic acid, 1 g⋅L⁻¹ activated carbon, 2% sucrose (weight/volume), and 7.0 g⋅L⁻¹ agar. The rooting medium was half-strength MS (Solarbio-M8525) modified with 1.5 mg⋅L⁻¹ 3-indolebutyric acid, 0.2 mg⋅L⁻¹ zeatin, 0.1 mg⋅L⁻¹ 1-naphthylacetic acid, 1 g⋅L⁻¹ pepton, and 2.5% sucrose (weight/volume). The pH of media was adjusted to 5.8 with 1 N of hydrochloric acid or sodium hydroxide (ALADIN-MFCD00132264) and autoclaved at 121 °C for 20 min before use.

In vitro culture conditions.

The culture vessels were continuously shaken on a gyratory shaker set at 50 rpm and constantly illuminated by cool white florescence (30–40 μmol·m⁻²·s⁻¹; 16 h/8 h light/dark photoperiod). Using the mixed culture method, the vessels were placed in total darkness after they were transplanted in media with colchicine and placed under the same conditions as those of the soaking method after colchicine treatment. To ensure the proliferation of culture, protocorms were transplanted monthly, and the culture room temperature was maintained between 23 to 27 °C.

Protocorms treated by pulsing colchicine solution.

Colchicine was made by dissolving 500 mg in 5 mL of deionized water and then diluting to a final stock volume of 50 mL. This solution was vacuum-filtered, sterilized (0.22-μM pores), and added to previously autoclaved liquid half-strength MS medium to adjust the final colchicine concentrations of 25, 75, 125, 250, and 500 μM (weight/volume). A total of 20 mL of each solution was dispensed into 50-mL treatment vessels. Modified MS media with no colchicine were used as the control. The protocorms were transferred from their mother vessels to their treatment vessels; then, they were withdrawn from treatment after 6 h, 12 h, and 24 h. Protocorms from each treatment were rinsed three times in sterile deionized water and spread on solidified orchid maintenance media (25 mL). A completely randomized 3 × 6 factorial design was created with 120 protocorms per treatment. There were three duration treatments (hours) and six treatment levels (control + colchicine concentration), with each treatment replicated three times. After the colchicine treatments, observations were conducted every 3 d. The surviving protocorms were subcultured monthly and maintained in a tissue culture chamber under the conditions described.

Protocorms treated by adding colchicine in solid medium.

Colchicine solutions were added to autoclaved half-strength MS medium before solidification. The final concentrations were 25, 75, 125, and 250 μM (weight/volume) colchicine. A volume of 20 mL of this solution was dispensed into 50-mL treatment vessels. All the steps were aseptically performed. For 30 d, protocorms from their mother vessels were kept at 25 ± 2 °C in darkness on this colchicine-fortified growth medium. The protocorms that survived were transferred to the solidified maintenance medium without colchicine for a 16-h/8-h light/dark photoperiod at a light intensity of 30 to 40 μmol⋅m⁻²⋅s⁻¹ provided by white florescent lamps. During this experiment, one treatment duration and five treatment levels were applied, with each treatment replicated three times. After 1 month of the initial culture and 2 months of subculture with half-strength MS, protocorm survival and mutational seedlings were identified and calculated from 8:00 to 10:00 AM.

Morphological measurements.

The 60-d variations and control seedlings were compared to determine morphological traits, including height (cm) and diameter of stem (cm). Leaves were evaluated to determine the leaf length (cm), width (cm), thickness, and index (length/width).

Guard cell characteristics.

After subculturing twice, great differences in morphological characteristics were observed in the surviving seedlings from the treated individuals. To determine the length and width of the stoma, the fully expanded leaves were used. The leaves were affixed to the transparent tape and then hand-scraped with a razor blade, leaving only the abaxial region of the epidermis. After washing, the epidermal segments were examined under a light microscope (Nikon CI-L; Nikon, Tokyo, Japan). The microscope displayed the images on a computer display monitor via the microscope’s built-in camera (Nikon Digital Sight DS-Fi1c, Japan). The length and width of the stoma were measured using the image analyzing software. Ten to 15 views per leaf sample were randomly selected to measure the stomata. They were identified as polyploids if their stomatal lengths were ≥1.25 × the control (Russell 2004).

Chromosome counting.

Actively growing root tips approximately 2 to 3 mm in length were collected under aseptic conditions at 8:00 to 10:00 AM and pretreated with 2 mM 8-hydroquinoline for 2 h at 15 °C. Then, samples were washed twice with distilled water and transferred to fixative solution (1 part acetic acid to 3 parts 95% ethanol, volume/volume) for 24 h at 4 °C. The fixed samples were hydrolyzed in 1 N of hydrochloric acid for 8 min. Then, they were rinsed with distilled water and stained by carbol fuchsin solution at room temperature for approximately 1 h. The stained root tips were excised, squashed onto slides in 55% acetic acid, and observed under a light microscope.

Statistical analysis.

To study the differences between the induced polyploids and control plants (with respect to the stomatal length and width, leaf length and width, and the diameter of stem), a one-way analysis of variance and multiple range tests (Student’s t test) were performed using the statistical software SPSS version 19.0 (IBM Corporation, Armonk, NY, USA).

Survival rate in the initial culture and subculture.

The number of surviving protocorms was recorded to evaluate the side effects of colchicine in different treatments. The survival rates differed with varying colchicine treatments (Table 1). In 1-month old culture, the differentiation of protocorms was lower, but the death rate was higher.

Table 1. Effects of colchicine treatment on the survival of protocorms of D. wardianum in the initial culture and subculture after treatment (soaked at colchicine solution).

View Fullscreen

The maximum survival rate was observed with 25 μM of colchicine with an immersion duration of 6 h (87.5%), whereas 250 μM with an immersion time of 24 h (53.33%) resulted in the shortest survival rates. In contrast, the control treatment with sterilized water showed a survival rate more than 90% with different durations (Table 1). Both the concentration and exposure time are essential to survival, but interaction between them was evident (Dhooghe et al. 2011). Survival rates were significantly lower for all treated plants regardless of the treatment duration and colchicine dose. With the increased exposure time, the survival rate was lower under the same dose of colchicine. The survival rate decreased with the increased concentration and constant exposure time.

After two periods of subculturing comprising 30 d, surviving protocorms began differentiating into seedlings with two or three leaves. Better survival rates were observed for the subcultures (62%–96%) than for the initial culture (53.33%–87.5%), which indicated that the toxicity of colchicine slowed with subculturing.

However, the result of cultivating in solid medium with colchicine was completely opposite (Table 2), and the survival rate was significantly lower than that of the initial culture, ranging from 44% to 88%. Similarly, in subculture, the survival rate during both experiments decreased with the colchicine concentration and exposure time. Compared with the pulsing treatment, protocorms treated by solid medium with colchicine had similar survival. With the increase in the colchicine concentration, the survival rate of the protocorms in the initial culture was lower, ranging from 60.83% to 91.67% (Table 2); these rates were higher than those of the colchicine solution treatment (53.33%–87.5%) in the initial culture (Table 1).

Table 2. Effects of colchicine treatment on the survival of protocorms of D. wardianum in the initial culture and subculture after treatment (added colchicine in medium).

View Fullscreen

Analysis of putative polyploids.

The potential polyploids of surviving plants were identified by morphological characteristics and stomatal features. Stomatal lengths of variant plants, evaluated as 1.25 × the controls (stomatal lengths ≥ control lengths), were identified as putative polyploids (Russell 2004).

The 250 μM and 500 μM colchicine solutions applied for 12 h resulted in higher numbers of variant plants (13 and 11 putative polyploids, respectively). The number of variant plants increased (from 1 to 9) with the increasing concentration of colchicine during the 6-h treatment. However, with 12 h of exposure, the number of variant plants began to decrease, and the concentration more than 250 μM resulted in the highest number of variant plants. In addition, with the 25, 75, and 125 μM concentrations, the efficiency was positively correlated with the exposure time. These results indicated that the concentration and exposure time interacted and are essential to produce polyploids. As is shown in Table 2, similar results were observed for protocorms cultivated in solid medium with 25, 75, 125, and 250 μM of colchicine for 30 d. This method could result in more putative polyploids than exposure in colchicine solution. The treatment with 75 μM of colchicine was the most effective (17 putative polyploids). However, the survival rate of this treatment was lower and the seedling growth was delayed. Similarly, when attempting to induce polyploidy in banana, negative effects on seedlings development were detected in medium with high colchicine concentrations (Van Duren et al. 1996).

Morphological analysis.

The development and differentiation of protocorms were suppressed by exposure to colchicine, and they grew slowly in the initial culture. Consequently, primordial shoots and leaves were visible after two subcultures. To screen the putative polyploids, the morphological characteristics of the leaf were observed. Compared with the control, the treated plantlets presented abnormal morphological characteristics, such as short stems, short internodes, hard leaves, and fewer roots (Fig. 2A and B). The chimera protocorms could simultaneously differentiate into diploid plants and variant plants (Fig. 2D). However, the chimera tissue of plantlets could fight for space in different locations of the plant; therefore, many leaves were clustered among the neighbor internodes or regenerated shoots sprouted at the leaf axils (Fig. 2C).

View Figure

• Download as Powerpoint
• Download Figure

As shown in Table 3, the leaf width of variants was nearly twice that of the controls. A comparison of five morphological features (leaf length, leaf width, leaf thickness, leaf index, and stem diameter) of treated plantlets after two periods of subculture and controls revealed that the average leaf width, leaf thickness, and stem diameter of the variants were significantly higher than those of the controls (Table 3). In contrast, the leaf length was not significantly different (i.e., only slightly longer in controls than in variants).

Table 3. Comparison of the morphology indexes of normal and putative polyploid plants.

View Fullscreen

Stomata and guard cell characteristics.

Stomata is the most common characteristic used to evaluate the effects of the ploidy level. In fact, diploids usually have smaller stomata and lower density than polyploids. As is shown in Fig. 3, the sizes of guard cells and epidermal cells in tetraploids were obviously larger than those of diploids. Furthermore, the stomatal density of tetraploids was high, and the stomatal frequency of tetraploids was 31.25% less than that of diploids (Table 4). Regarding the stomatal width and length, the tetraploids were significantly larger than diploids, with average increments of 12.88 and 14.64 μm, respectively. The stomatal area of tetraploids was slightly more than twice as large as that of the diploids. This wide variation between polyploids and diploids is a useful indication of ploidy of D. wardianum.

View Figure

• Download as Powerpoint
• Download Figure

Table 4. Comparison of stomata densities and guard cell sizes of D. wardianum diploid plants and polyploid plants.

View Fullscreen

Chromosome counting.

The ploidy assessment performed using stomata characteristics is convenient and rapid, and it allows breeders to prescreen putative polyploids at an early stage. However, this method is easily interfered with by mixoploids or chimeras (Kaensaksiri et al. 2011). To corroborate the putative polyploids measured using this method, a chromosome counting assay was performed for further confirmation. The results of chromosome counting of the root tip were partly in line with the stomata characteristics.

The chromosome number of the control plants was 2n = 2X = 38 (Fig. 4A), whereas that of the tetraploid plants was 2n = 4x = 76 (Fig. 4B). However, counting the chromosome number in metaphase seemed to have limitations because chromosomes of D. wardianum are microchromosomes and can easily overlap, thus making counting difficult. Therefore, it is extremely difficult to determine the effective chromosome number of the plants. Other methods are needed to further verify the ploidy. Finally, we obtained 11 tetraploid plantlets from the pulsing colchicine treatment and seven from the solid medium with colchicine.

View Figure

• Download as Powerpoint
• Download Figure

Survival rate.

Mitotic chromosome doubling of plant tissues (such as protocorms, protocorm-like bodies, shoot tips, seedlings, or other parts of the meristem) has been intensively studied by many researchers and has expanded the use of germplasm resources. However, colchicine can cause side effects because of its high toxicity, which blocks mitosis and modifies the differentiation process (Forkosh et al. 2020; Pintos et al. 2007; Potenza and Tellez-Iñón 2015). In addition, the sensitivity of this antimitotic agent differs in various species and in different parts of tissue. Thus, the concentration and exposure time are important when applying this method. An effective and nonlethal protocol should be created before large-scale application. Regardless of the treatment, correlations between survival and the concentration of colchicine or exposure duration are negatively proportional (Tables 1 and 2). That is, higher concentrations or longer exposure times are more detrimental to plants growth and survival. Similar results were reported for other species, such as Watsonia lepida (Ascough et al. 2008), Miscanthus (Głowacka et al. 2010), and Eriobotrya japonica (Blasco et al. 2015).

A comparison of the lethality of the initial culture with subculture showed that holding the colchicine treatment resulted in higher mortality of the former than for the latter (Tables 1 and 2). However, in pulsing solution, the mortality of the subculture was lower than that of the initial culture, ranging from 62% to 96% (Tables 1 and 2). The difference might be attributable to the long duration, thus leading to more chemical penetration into protocorms. Contrary to this, the control (noncolchicine-treated protocorms) showed a survival rate of 100%. Interestingly, the survival rates of the control with immersion treatment were less than 100% (i.e., inverse to the exposure time). This phenomenon indicated that the immersion time of water might have few effects on survival.

Effects of the colchicine concentration and treatment duration on polyploidy.

The success of polyploidization depends on the type of explant and permeability of the tissue and affinity of chemicals to the meristem (Allum et al. 2007; Dhooghe et al. 2011). Based on previous research, the best explants of orchid for polyploidy induction are the undifferentiated embryonic cells, such as protocorms or protocorm-like bodies (Griesbach 1981; Miguel and Leonhardt 2011; Sarathum et al. 2010). In addition to the effect of the explant, the genotype is a key factor (Khosravi et al. 2008). However, this experiment was difficult because D. wardianum has 38 small chromosomes (Felix and Guerra 2010), thus increasing the challenge of polyploidy induction and confirmation.

To develop an effective protocol for polyploidization, both solid medium with colchicine and liquid medium with colchicine were applied. The results indicated that the liquid medium with colchicine treatment was more effective than the solid medium treatment. Lethality of the solid medium was lower in the initial culture than in subculture (Tables 1 and 2); however, this trend was reversed in liquid treatment (Tables 1 and 2). This totally different response may be attributed to the long duration of solid medium treatment, thereby leading to excessive absorption of chemicals by protocorms.

The liquid medium with colchicine treatment had high induction rates, and 250 μM for 12 h was the best protocol during this study (Table 1). The identified tetraploids recovered normal growth in subsequent culture. A similar result was found for Cattleya (Orchidaceae); the optimal concentration of the colchicine solution ranged from 125 to 250 μM, but the durations might be different for each species (de Mello e Silva et al. 2000; Huy et al. 2019; Lone et al. 2010; Wu et al. 2020). Although the solid medium with colchicine treatment showed higher lethality, it is possible to obtain tetraploids in lower concentrations, thus reducing the amount of expensive and toxic colchicine necessary. Further experiments are needed to optimize the solid medium with colchicine protocol, especially its duration.

During the soaking experiment, we found that the concentration and exposure time had obvious interactions. The effect of these on chromosome doubling was determined by examining the morphological parameters, size and density of stomata, and chromosome counting. The variant plants increased with the increasing time (6 h, 12 h, 24 h) and concentrations (from 25 to 125 μM). For the 250 and 500 μM treatments, with an application time of 12 h, the maximum number of polyploids was observed, but the survival rate was slightly lower than that associated with 25, 75, and 125 μM. The best concentration of solid medium was 75 μM colchicine for in vitro culture for 30 d, and the variant rate was higher than that of the solution with colchicine, but the survival rate was slightly lower (Tables 1 and 2). Thus, it can be inferred that high colchicine doses and long durations more effectively produce variant plants within a certain scope. Solid medium with colchicine can be more advantageous for generating stable polyploids because the lethality is controlled. Thus, the reasonable duration of the solid medium with colchicine deserves further research.

Differences between the three methods of identification.

The morphology of subculture plantlets of polyploids and controls showed greatly significant differences in the widths of leaves and the diameters of stems (Table 3). The thickness and leaf index of the polyploids were significantly thicker and lower than those of the control plants. During the study of Phalaenopsis, differences in leaf and stem characteristics of diploids and tetraploids were genotype-specific (Chen et al. 2009). Thus, an analysis of morphology is essential to save space, time, and labor during early investigations.

Stomatal size and density were regarded as useful indicators of the discrimination of ploidy levels of many species (Aryavand et al. 2003; Gu et al. 2005; Miguel and Leonhardt 2011; Tavan et al. 2015). Furthermore, many studies determined ploidy levels based on stomatal characteristics to discriminate plants with different ploidy (Blasco et al. 2015; Chen et al. 2009; de Mello e Silva et al. 2000; Miguel and Leonhardt 2011). A previous study claimed that stomatal lengths more than 1.25-times that of the controls indicated polyploids (Russell 2004). Among the parameters of stomatal characteristics, stomatal density may be the most reliable when performing rapid prescreening of polyploids and might avoid the influence of external factors (Chen et al. 2009; Chung et al. 2014; de Mello e Silva et al. 2000).

During our study, stomatal density and size were noticeably 1.25-times more than those of the controls (Table 4); therefore, they could be regarded as rapid, efficient, and economical tools for distinguishing ploidy of D. wardianum. Similar results have been reported (Miguel and Leonhardt 2011). Moreover, there have been reports of the significantly different stomatal characteristics of diploids and tetraploids in other orchids (Chen et al. 2009; Chung et al. 2014), in line with our findings. However, stomatal size was easily influenced by the external environment, leaf age and position, and endopolyploid cells in the leaf tissue (Blasco et al. 2015; Chen et al. 2009; Sakhanokho et al. 2009). Thus, we suggest that the stomatal density may be a better parameter for prescreening the polyploids, but stomatal sizes can be used to assist with identification.

However, ploidy determined by morphological and stomatal characteristics has limitations because mixoploids and periclinal chimeras were not easy to verify. Chromosome counting is the most intuitive and effective method of confirming ploidy levels (Dhooghe et al. 2011; Winarto et al. 2010; Zlesak et al. 2005), even though it is rather difficult and time-consuming for a series of treatments involving fixation, enzymatic hydrolysis, staining, and microscopy of the root tip or shoot cells (Dhooghe et al. 2011; Portela de Carvalho et al. 2005), and especially for small chromosomes and roots of Dendrobium (Liao et al. 2012; Wilfret and Kamemoto 1971). However, we used chromosome counting during our investigation because of its effectiveness.

All Dendrobium species analyzed were previously determined to have 2n = 2N = 38 chromosomes, except for D. formosum, with 2n = 2N = 40 chromosomes (Jones et al. 1998; Tanaka and Kamemoto 1984). However, the chromosome numbers of D. wardiadium might be questionable. According to another study (Sau 1983), D. wardiadium showed 2n = 40 chromosomes, which was in contrast to n = 38 chromosomes reported by another study (Felix and Guerra 2010). Differences in the chromosome number may be correlated with their geographic distributions and ecological adaptations, or with atypical individuals such as aneuploid (Sau 1983; Vallejo‐Marín and Hiscock 2016; Wilfret and Kamemoto 1971). During our study, we showed that diploids had 2n = 2x = 38 chromosomes (Fig. 4A), in agreement with another study (Felix and Guerra 2010). According to these results, 2n = 4x = 76 chromosomes are representative of tetraploids. However, a large number of chromosomes in tetraploids can easily concentrate together, thus hindering the precise determination of the chromosome number. Therefore, combined with morphology and stomatal lengths, chromosomes numbers more than 70 have been recorded as tetraploids (Fig. 4B). The use of an alternative method helped to discard any potential mixoploids and periclinal chimeras. Flow cytometry is a precise and efficient technique for analyzing plant ploidy (Dhooghe et al. 2011; Ma et al. 2015). Further research using flow cytometry may provide more evidence to corroborate the ploidy of D. wardiadium.

Polyploidy has an essential role in improving special and favorable characteristics of orchids, such as enlarged flower sizes or extended blooming times (de Medeiros Vichiato et al. 2014; Tang and Chen 2007; Zeng et al. 2020). In the present study, an effective protocol for in vitro polyploidization was usually limited to the concentration and duration of exposure. For polyploid induction of Dendrobium, we suggest that protocorms should be used as explants with colchicine concentrations of 75 to 125 μM; however, the duration time can be adjusted according to the treatment condition (solution with colchicine or solid medium with colchicine). The optimal treatment for inducing polyploidy in D. wardiadium could be 250 μM colchicine solution for 12 h; during this study, it resulted in the highest percentage of tetraploids and lower lethality. The solid medium with colchicine is a protocol worth considering because more stable polyploids are induced, even though its survival rate is slightly lower than that associated with colchicine immersion. The most successful treatment protocol was solid culture medium supplemented with 125 μM colchicine for 30 d.

Introducing favorable genes from tetraploid species to another fine diploid species is a difficult and challenging task because crosses between the two groups can easily produce sterile triploid plants. However, increasing the ploidy level of diploids or triploid may overcome this problem. Polyploids that we induced may be an excellent germplasm resource for further breeding, and we expect that the ongoing selection process will cultivate morphologically diverse and highly adaptable cultivars of D. wardianum. Our protocol contributes to the protection and utilization of this endangered species and alleviates the demand for this natural resource because of its medicinal and ornamental use. Based on the results, further studies should be performed to achieve the desired objective.