Minimizing the deleterious effects of endophytes in plant shoot tip cryopreservation

Abstract

Plant cryopreservation technologies are used within gene banks for the long‐term preservation of vegetatively propagated collections. Surface‐sterilized plant tissues grown in the field, greenhouse/screenhouse, growth chamber, or in vitro are the source of shoot tips subjected to vitrification‐based cryopreservation methods. Here, we describe the methods used to minimize microbial contamination during the tissue culture initiation process. We also discuss the occurrence and possible elimination of endophytes after extended in vitro culture and during recovery after liquid nitrogen exposure. We describe two case studies in which bacterial endophytes were observed in Citrus gene bank accessions during recovery after cryopreservation. These were identified using the MinION Oxford Nanopore system and Kirby–Bauer disc diffusion assays to examine the bacterial responses to antibiotic exposure. The methods used in this case study could be applied to identify endophytes to better target antimicrobial treatments of plant tissue collections.

PLANT–ENDOPHYTE RELATIONSHIPS

Endophytes, microorganisms living within plant tissues that are not eradicated after surface sterilization, can interfere with or inhibit the initiation and growth of in vitro cultures or the recovery of shoot tips after cryopreservation. Genome‐sequencing technologies have revealed that a diverse community of endophytes resides within plants, more than were previously identified through traditional culture techniques (Thomas et al., 2008; Finkel et al., 2017). The types and quantity of endophytes vary throughout plant tissues, with roots in soil often being heavily colonized and young actively growing shoots possessing fewer endophytes (Wang et al., 2016; Gómez‐Lama Cabanás et al., 2021). In many cases, endophytes are believed to have positive effects on the plant, including assisting with nutrient acquisition, nitrogen fixation and phosphate solubilization, phytohormone and siderophore modulation and production, protection against abiotic stresses, and phytopathogen control (Dubois et al., 2004; Hardoim et al., 2008, 2015; Reinhold‐Hurek and Hurek, 2011; Liaqat and Eltem, 2016; Rajamanickam et al., 2018; Proboningrum et al., 2019; Acuña‐Rodríguez et al., 2020; Kanani et al., 2020; Jiang et al., 2021). Some studies have shown that introducing endophytic bacteria is beneficial for both plant field performance and growth in in vitro culture systems (Murthy et al., 1999; Khan and Doty, 2009; Pohjanen et al., 2014; Quambusch et al., 2016; Patle et al., 2018), indicating a synergistic effect for these plant–endophyte interactions. However, endophytes can also have negative effects on plant growth, particularly for in vitro–grown plants (Thomas, 2004a, 2004b). Endophytes may become apparent immediately upon introduction into culture. Alternatively, endophytes may emerge from the in vitro plants and appear as bacterial contamination, which then can inhibit proliferation efforts (Reed et al., 1997). Within this paper, we limit ourselves to describing the strategies used to overcome the undesirable effects of endophytes on in vitro plant culture introduction, extended culture durations, and shoot tip cryopreservation and recovery. We also present a case study in which bacterial contaminants are isolated from cryopreserved citrus (Citrus L.) shoot tips and subsequently sequenced for identification.

The plant–endophyte relationship can be problematic for plants that are propagated in tissue culture (Thomas, 2010). In vitro plant establishment is necessary for a number of applications, including maintaining and propagating clean stock materials for nurseries and industrial/bioreactor production. In addition, in vitro cultures are used in gene banks and to safeguard exceptional plant species that cannot currently be conserved using conventional seed storage methods (Pence, 2011). For in vitro culture, the plants are surface‐sterilized to remove any external bacteria and fungi before being grown in sterile sucrose‐rich culture conditions. For some plants, it is relatively straightforward to establish cultures that do not show evidence of microbial “contamination” in culture conditions; however, such contamination can be difficult to observe. The cultures must be observed at eye level, from above, and from underneath; furthermore, it may be helpful to have a dark background. Contaminants may appear as a faint cloudy growth at the base of plants or along the roots. Contamination could also present as halos, hazy streaks, or dry grainy appearances on the surface of the medium, or it could emit odors (Thomas, 2004a). For some crops, particularly those of wild or tropical origin, it may be difficult to establish clean cultures, despite extensive surface sterilization (Pence, 2005; Thomas and Kumari, 2010; Herman, 2017). In addition to negatively affecting the growth of the plant in culture, the presence of bacterial contamination can prevent the exchange of plant genetic resources across international borders due to phytosanitary restrictions. It may not be possible to eliminate all endophytes from plants; however, it is often possible to eliminate the ones that prevent successful in vitro plant establishment and growth.

There are many examples where “axenic” plant tissue cultures show evidence of latent bacterial contamination after an extended culture duration, despite the use of best practices for culture maintenance. In these cases, the endophytic bacteria gradually appear around the base of plants and on the culture medium, often directly associated with plant tissues submerged in the medium, and inhibit or prevent subsequent plant growth. These bacteria may not have exhibited negative effects while inside the plant tissue as endophytes, but could pose significant problems when they emerge from the plant. This emergence may be a result of a change in plant health or quality that may be difficult to detect visually. The presence of bacterial contamination impedes the intended use of the plant cultures, particularly because they may not grow as vigorously as is necessary for culture propagation and multiplication.

In vitro plants are often the source material for shoot tip cryopreservation (long‐term storage of plant shoot tip tissues in liquid nitrogen), which is one method used for the preservation of vegetatively propagated plant cultivars in gene banks (Bettoni et al., 2021). The shoot tip excision, preculture, cryoprotectant exposure, desiccation, and freezing stresses introduced during the cryopreservation process may result in microbial contamination, which is subsequently observed during shoot tip recovery (Senula et al., 2018). Plant shoot tips often fail to recover and grow when they are cryopreserved in a contaminated state (Keller et al., 2011).

ESTABLISHING AND MAINTAINING “CLEAN” IN VITRO CULTURES

In vitro culture establishment

Explants must be sampled from plants grown in the field, greenhouse, screenhouse, or growth chamber, surface‐sterilized to remove external contaminants, and then placed into a sterile growth environment for the successful establishment of an in vitro culture. The tissue collection method (particularly from the wild) affects the success of the culture establishment (Pence, 2005). Often, multiple replicate plants are sampled and placed into individual culture vessels, such as test tubes, with the goal of avoiding possible cross‐contamination between explants and at least one plant being successfully introduced without visual evidence of contamination (Thomas and Kumari, 2010).

When the first attempt at plant introduction is not successful, remediation efforts begin. There are multiple options, including (1) alternative surface‐sterilization methods, (2) resampling from the source plant, (3) using young, actively growing tissue for explants, (4) growing the source plant in a protected (non‐field) environment, (5) screening for the presence of contaminants in surface‐sterilized explants, and (6) adjusting culture conditions (e.g., pH, carbohydrate source, osmotic conditions, and temperature). In addition, antimicrobial and antibiotic treatments may be necessary.

Standard and more intensive surface‐sterilization methods

Surface‐sterilization methods are effective at reducing or eliminating surface contaminants, but they usually do not affect endogenous microbes because the sterilization methods do not penetrate deeply into the plant tissue. The agents used for the surface sterilization of plant tissues include sodium hypochlorite, calcium hypochlorite, ethanol or isopropanol, hydrogen peroxide, chlorine gas, sodium dichloroisocyanurate, isothiazolone‐based biocide, antibiotics, and mercuric chloride (Enjalric et al., 1988; Parkinson et al., 1996; Singh et al., 2011). Some of these compounds, such as chlorine gas, sodium dichloroisocyanurate, and mercuric chloride, are highly toxic and must be handled and disposed of with care. More commonly used surface‐sterilization protocols usually include a treatment with 70% ethanol or isopropanol and/or exposure to 5–20% sodium hypochlorite with a surfactant such as Tween 20 (Croda International, Snaith, United Kingdom). The concentration of a disinfecting agent, or a combination of multiple agents, and the amount of time the plant tissue is exposed to it varies greatly with tissue type and source, plant species, microbial load, and so on. Because the toxicity of sterilizing agents varies with species and tissue types, some plant extracts may benefit from using a lower concentration of disinfecting agent for a longer time, while others may benefit from using a higher concentration for a shorter time.

Resample from the original plant or a different individual

If the first round of plant introduction does not yield clean in vitro plants, it is often easiest to attempt to reintroduce materials into culture. Sometimes a different plant (of the same genotype) can be successfully introduced, and altered cultural conditions (e.g., a shorter time between harvest and use, sunny dry field conditions) can make a significant improvement in culture establishment. Reed (1999) attempted reintroductions into tissue culture before attempting culture clean‐up efforts.

Introduce from young actively growing tissue

Selecting younger, actively growing tissue may result in fewer issues with endophytes compared with the use of older tissues, which likely have a higher microbial load. Sampling explants from aboveground plant structures, such as shoot tips and nodal sections, is usually more effective than sampling from belowground structures, such as bulbs or rhizomes. The time of year that explants are taken can also have an impact. Hohtola (1988) had better success in establishing Pinus sylvestris L. shoot tip explants using younger tissue in the spring and summer seasons than during other times of the year, when significantly more contamination was observed.

Explants from etiolated shoots often possess significantly fewer internal and surface contaminants, improving the number of clean cultures after establishment (George, 1993). Murasaki and Tsurushima (1988) discovered that Cyclamen L. petioles were free of endogenous contaminants after etiolation. Similarly, Cooper (1987) found that avocado (Persea americana Mill.) had decreased contamination levels when etiolated shoots were introduced into tissue culture.

Grow the source plant in greenhouse conditions to improve cleanliness

Growing plants in a greenhouse or growth chamber using clean cultural practices can often yield explants with a lower microbial load than those grown in the field. Usually, explants collected from plants in the greenhouse do not require as long of a disinfection period as those from the field. Decreasing humidity and temperature in a controlled growth environment can also be beneficial, in addition to using drip irrigation to avoid excess water on and around the plants (Knauss and Knauss, 1980). Debergh and Maene (1981) found that growing source plants in a greenhouse for several months at fairly low humidity (70%) and minimizing watering led to significantly more healthy, non‐contaminated cultures than when plants were not grown in the same conditions.

There are also instances of endophytic contaminants being eliminated from in vitro plants by establishing and growing rooted plants in the greenhouse with or without antimicrobial additives, and then reintroducing explants into tissue culture. Izarra et al. (2020) were successful in eliminating endophytic contaminants from in vitro sweet potato (Ipomoea batatas (L.) Lam.) accessions by growing the plants in greenhouse conditions for four months in soil supplemented with benomyl and pentachlorobenzene and watering weekly with 2 mL L⁻¹ Dimanin (Bayer HealthCare, Whippany, New Jersey, USA). The plants were then reintroduced in vitro and tested negative for bacteria, as determined using PCR. In another case, contaminated cultures of Populus L. were rooted and acclimated to a greenhouse environment (Garten and Moses, 1985). New explants were taken from the soft new growth and reestablished into tissue culture, yielding clean cultures. Enjalric et al. (1988) treated mother rubber plants (Hevea brasiliensis (Willd. ex A. Juss.) Müll. Arg.) with antimicrobial agents prior to their reintroduction into tissue culture, while Tanner et al. (2020, 2021) minimized the contamination of introduced plant materials using the antimicrobial agent 8‐hydroxyquinoline citrate.

Initial screening on a bacterial medium to select for clean cultures

Some tissue culture laboratories have adopted routine microbial screening procedures as part of the plant introduction process. This may be labor intensive, but it can prevent contamination issues in the short and long term. In these cases, plant materials introduced into culture are surface‐sterilized, and then tissue‐explant pieces are split into propagatable portions and placed on solid plant‐growth medium, while corresponding pieces (often internode sections) are exposed to liquid microbial growth media such as nutrient broth or potato dextrose broth in individual test tubes. The liquid cultures with microbial growth media are grown at 25°C to test for microbial growth by observing their turbidity; only the plant cultures corresponding to the clear liquid cultures (and presumably without microbial contamination) are kept (Reed et al., 1997; Reed, 1999; Thomas, 2004a; Volk et al., 2021). Similar methods have been used for Dieffenbachia Schott (Knauss, 1976) and Chrysanthemum L. (Panicker et al., 2007; Thomas et al., 2009).

Altering growth conditions

Modifying the culture parameters (e.g., pH, carbohydrate source, osmoticum, or temperature), so that conditions are less favorable for microbial growth and more favorable for plant growth, can also be effective at reducing endogenous contamination (Leifert et al., 1994). Changing the pH of the medium to a more acidic environment may suppress the growth of undesirable microbes that interfere with plant propagation and allow for more normal tissue growth. Cantabella et al. (2021) found that adjusting the pH of a liquid culture medium to 3 was beneficial in significantly reducing endophyte levels after six days of culture without affecting the number or fresh weight of shoots, in comparison with those grown at pH 5.7, at which the endophyte levels did not significantly change over the same period. Four different endophytes were controlled in this manner, indicating that a more acidic pH may control a variety of organisms.

In vitro culture maintenance

Successfully established in vitro cultures sometimes show evidence of contamination after extended periods of tissue culture (Leifert et al., 1994; Thomas, 2004a, 2004b; Thomas and Prakash, 2004; Thomas et al., 2009; Abreu‐Tarazi et al., 2010). It is hypothesized that this is the result of a changing environment, such as the nutritional status in planta, that makes the culture medium more favorable than the plant tissue for bacterial growth (Hallmann et al., 1997).

In our laboratory, we have observed latent bacterial infections in grape (Vitis L.) and kiwifruit (Actinidia Lindl.) shoot cultures that became apparent after many transfers following culture establishment. Cultures of Vitis romanetii Rom. Caill. ‘C‐166‐025’ (DVIT 3272) were established from field‐grown plants and appeared to be free of bacteria initially, but after approximately a year, or six to eight transfer cycles, a bacterial haze started to appear near the base of the plants, around the stems and roots, and exuded from lower leaves when in contact with the medium or sides of the culture vessel (Figure 1). In addition, cultures of Actinidia chinensis Planch. ‘P.IC16’ (PI 667932) were established from field‐grown plants and appeared to be free of bacteria until after a similar length of time and number of transfer cycles, when bacterial growth began to appear. In this case, the bacterial growth was noticed only in older cultures (8–12 weeks after the previous transfer) and grew from the leaves at the base of the plants and onto the surface of the medium. In both cases, the latent bacteria did not seem to significantly impact the propagation and growth of the shoot cultures.

Figure 1.

Vitis in vitro cultures established from field plants. (A) Vitis ficifolia (DVIT 1160.02) exhibiting latent bacterial growth two months after in vitro initiation. An off‐white bacterial colony has formed around the explant on the surface of medium. (B) Vitis species (DVIT 1445) culture that does not exhibit signs of contamination two months after introduction. No apparent bacterial or fungal presence can be detected visually. (C) Vitis species (DVIT 1445) exhibiting latent bacterial growth two months after in vitro initiation. Yellow bacterial colonies have formed around the base of explant, on the surface of the medium, and on the sides of the culture vessel. (D) Vitis romanetii (DVIT 3272) exhibiting latent bacterial growth after one year in culture. Note the white bacteria around the base of stem and exuding from the leaves onto the sides of the culture vessel (circled in red). Scale bar = 1 cm.

If possible, it may be easiest to re‐establish in vitro cultures when latent bacterial contamination occurs. When that approach is not feasible, in vitro cultures can be treated with antibiotics or grown on a medium that contains antimicrobial compounds.

Shoot tip cryopreservation and recovery

Traditionally, plant shoot tip cryopreservation methods use 1–2‐mm shoot tips derived from plants grown in vitro. There are examples of shoot tips derived from apparently clean in vitro plants that are overgrown with bacteria as they recover from liquid nitrogen exposure (Senula et al., 2018). This may be the result of latent bacteria emerging after the shoot tip has undergone the stresses of the cryopreservation procedure, preventing the shoot tip from recovering after the cryopreservation process (Keller et al., 2011). When this happens, it may be necessary to re‐introduce the plants into a tissue culture or to treat the recovering shoot tips with antimicrobial agents (Edesi et al., 2017).

Use of antimicrobial compounds

Media containing antimicrobial compounds may suppress or eliminate microbial growth. Plant Preservative Mixture (PPM) (Plant Cell Technology, Washington, D.C., USA) is a mixture of two isothiazolones, methylchloroisothiazolinone and methylisothiazolinone, that are described as biocides that control microbes by penetrating their cell walls and inhibiting several enzymes involved with the citric acid cycle and electron transport chain (Niedz, 1998). In our laboratory, we have used PPM to control, and in some cases apparently eliminate, microbial growth in plant tissue cultures. In one example, shoot cultures of breadfruit (Artocarpus altilis (Parkinson) Fosberg) were contaminated with bacteria that grew at the base of the plants of older cultures (10–12 weeks). The bacteria were isolated, sequenced, and determined to be Propionibacterium acnes (data not shown). As an alternative to using antibiotics, breadfruit shoot growth medium was supplemented with 5 mL L⁻¹ of PPM prior to autoclaving, and single apical shoots (2–3 cm) were inserted into the solidified medium. Two replicates for each of 40 plants were cultured for two transfer cycles of 12 weeks each on the PPM‐supplemented medium, after which the amount of PPM in the medium was reduced to 1 mL L⁻¹ and the plants were cultured for another two transfer cycles of 12 weeks each. The plants were then transferred to a medium without any PPM and cultured for four transfer cycles of 12 weeks each, after which time no bacterial growth was detected. These results suggest that the bacterium was either suppressed to very low levels or eliminated; however, additional testing is needed to confirm either scenario. No toxicity to the breadfruit plants from the PPM‐supplemented medium was observed, but its effects vary with different plant species. We have also used PPM in our laboratory to control bacterial contaminants in explants used for cryopreservation (Bettoni et al., 2019). In this case, using 15 mL L⁻¹ PPM in both the pretreatment (14 days) and the preculture (three days) media was effective at controlling endophytic bacteria in growth chamber–sourced Vitis explants and shoot tips that otherwise could cause disruptions during the cryopreservation process.

Although PPM can be effective in controlling, and in some cases eliminating, endogenous microbial contaminants, it also has been shown to have a negative impact on plant growth and regeneration. In Petunia Juss., the number of explants producing shoots and the number of shoots and buds produced decreased significantly with increasing concentrations of PPM over 2 mL L⁻¹ (Compton and Koch, 2001). Orlikowska et al. (2015) reported that PPM was toxic to chrysanthemum ‘Ludo’ explants, and decreased the shoot or root lengths and number of axillary or adventitious shoots produced by blackberry (Rubus fruticosis L.), raspberry (Rubus idaeus L.), strawberry (Fragaria ×ananassa Duchesne), and Anthurium Schott at concentrations ranging from 3 to 10 mL L^–1. Furthermore, PPM also caused blackberry, raspberry, and strawberry plants to produce a black exudate that leached into the medium (Orlikowska et al., 2015).

There are a limited number of reports of other antimicrobial compounds used in plant tissue culture. Kathon LXE (Rohm and Haas, Philadelphia, Pennsylvania, USA) is a product containing 5‐chloro‐2‐methyl‐4‐isothiazolin‐3‐one and 2‐methyl‐4‐isothiazolin‐3‐one that has been used for antimicrobial control in kiwifruit in vitro cultures (Debenham and Pathirana, 2021). Vitrofural, produced by the Center of Biologically Active Chemicals (Villa Clara, Cuba), has been investigated for use in the control of bacterial contamination in bamboo (Dendrocalamus asper (Schult.) Backer) tissue cultures (Ornellas et al., 2017), as well as in chrysanthemum, Anthurium, Hosta Jacq., raspberry, blackberry, and strawberry in vitro cultures (Orlikowska et al., 2015), with some success. Essential oils with antimicrobial properties have also been explored for use in controlling contamination in plant cultures (Hamdeni et al., 2022). Rosemary (Salvia rosmarinus Schleid.) and thyme (Thymus vulgaris L.) essential oils were tested for their ability to inhibit microbial growth in Aloe vera L. shoot cultures. Although thyme essential oil was toxic to plant growth, rosemary essential oil was effective at controlling both microbial contamination and oxidative browning at the concentrations tested (Hamdeni et al., 2022).

Antibiotic applications may be necessary when the previously listed options do not successfully remove problematic endophytes from in vitro cultures. The use of antibiotics may also improve plant growth (Thomas and Prakash, 2004; Rákosy‐Tican et al., 2011); however, plant species vary in their tolerance to various types and concentrations of antibiotics, so it is necessary to determine which antibiotic, or combination thereof, is necessary to eliminate endophytes and minimize the side effects on plants (Lata et al., 2006). Antibiotics may be applied directly in the growth medium or as a liquid placed on top of the medium, and the exposure durations are dependent upon the type and concentration of the antibiotics and the corresponding plant sensitivities (Thomas, 2004b; Kulkarni et al., 2007; Ray et al., 2017). Alternatively, explants may be soaked in antibiotic solutions (Habiba et al., 2002). Some cultures of the same genotype or of different cultivars of the same species may be more responsive to the treatments than others (Buckley et al., 1995; Thomas and Prakash, 2004; Jena and Samal, 2011; Mbah and Wakil, 2012). It is possible to determine which specific antibiotics are effective against the contaminants present by going through a process of bacterial isolation followed by testing the bacterial growth in response to a range of antibiotics, often using a disc diffusion assay (Balouiri et al., 2016).

CASE STUDY: ISOLATION, IDENTIFICATION, AND ANTIBIOTIC RESPONSE OF TWO CITRUS ENDOPHYTES

The introduction of plants into tissue culture for cryopreservation purposes is time‐ and labor‐intensive. For some crops, it is possible to cryopreserve materials harvested directly from the field, greenhouse/screenhouse, or growth chamber (Bettoni et al., 2019, 2021); however, these may exhibit higher levels of contamination during the recovery process. For the citrus plants used in the present case study, the required media were not available for the in vitro propagation of shoot tip source plants, so the shoot tips were derived directly from screenhouse‐grown plants (Volk et al., 2012, 2017, 2019). We describe two examples of our efforts to isolate and identify possible eradication methods for the bacteria that were observed during the recovery process after citrus shoot tips were cryopreserved.

From 2012 through 2017, shoot tips from 434 accessions of diverse Citrus species and some wild relatives in the USDA National Plant Germplasm System citrus collection were cryopreserved using a droplet‐vitrification method (Volk et al., 2017, 2019). Prior to long‐term storage, their viability was assessed by micrografting. The shoot tips were warmed from liquid nitrogen, diluted into sucrose solution, and then plated onto Citrus recovery medium (Woody Plant Medium [WPM] salts [McCown and Lloyd, 1981], supplemented with Murashige and Skoog [MS] vitamins, 50 g L⁻¹ sucrose, and 7 g L⁻¹ agar at pH 5.7) for up to 24 h. A small amount of tissue (~0.2 mm) was trimmed from the base of each shoot tip before the shoot tip was placed onto a seedling rootstock, and the resulting micrografted seedling was transferred into a test tube containing micrografting recovery medium (MS inorganic salts supplemented with 100 mg L⁻¹ myo‐inositol, 0.2 mg L⁻¹ thiamine HCl, 1 mg L⁻¹ pyridoxine HCl, 1 mg L⁻¹ nicotinic acid, 75 g L⁻¹ sucrose, and 7 g L⁻¹ agar at pH 5.7) for regrowth.

During the recovery process, it was observed that 42 of the 434 Citrus accessions were infected with bacteria that grew on and/or around the shoot tips and recovering micrografted plants. The bacteria exhibited light to heavy growth, with a low to significant impact on the recovery of the shoot tips. The bacteria were clear, white, off‐white, or orange/tan, and appeared hazy, slimy, stringy, or sudsy. The bacteria displaying light growth were usually visible around the recovering shoot tips on the Citrus recovery medium prior to micrografting. The shoot tips could often successfully regenerate despite this light bacterial growth, which may be similar to the endophytic microbiota seed‐to‐shoot transmissions that occurred in Citrus and did not inhibit growth, as observed by Faddetta et al. (2021). It appeared as if the bacteria were incorporated into the micrografts, and the shoot tips healed and recovered as normal, even if some light bacterial growth remained around the seedling roots in the micrografting recovery medium. On the other hand, moderate to heavy bacterial growth on or around the recovering shoot tips often overwhelmed the tissue and hindered development and recovery after micrografting.

Isolation

Citrus shoot tips were cryopreserved and then warmed for the regrowth assessment. Our case studies include the bacterial isolate RCRC4244_C1 from shoot tips of the Citrus hybrid RCRC4244 plated on Citrus recovery medium overnight, and RCRC4210_C1 from the Citrus hybrid RCRC4210 shoot tips plated on Citrus recovery medium overnight and then micrografted onto ‘Carizzo’ seedling rootstocks grown in micrografting recovery medium for seven days. Beige bacterial growth was observed on the recovering RCRC4244 shoot tips and on the surface of the Citrus recovery medium immediately surrounding the shoot tips (Figure 2). Gas‐producing bacterial growth was observed around the base of the RCRC4210 micrograft rootstock and roots, and a slight bluish tinge was produced in the micrografting recovery medium (Figure 3). The bacteria were sampled with a loop and immediately streaked onto both nutrient agar and micrografting recovery medium to identify the culture conditions that provided robust microbial growth. The plates were incubated at 25°C until colonies were observed, and a single colony was selected to restreak and then inoculate liquid nutrient broth for freezer stocks and antibiotic testing. Small white colonies of RCRC4244_C1 were apparent on the nutrient agar plates after 48 h, while similar smaller colonies grew very slowly on micrografting recovery medium. Compact white to pinkish colonies of RCRC4210_C1 were visible on both the nutrient agar plates and the micrografting recovery medium after 48 h. Both isolates grew sufficiently, after shaking for 48 h at 25°C in liquid Nutrient Broth (Difco; Becton, Dickinson and Company, Franklin Lakes, New Jersey, USA), for the preparation of glycerol freezer stocks.

Figure 2.

Antibiotic disc diffusion assays and bacterial growth response of isolate RCRC4244_C1 during a shoot tip recovery. (A) Antibiotic discs placed on the surface of nutrient agar spread with isolate RCRC4244_C1. Some zones of inhibition had very distinct borders (for example, red arrow), while others were less distinct (for examples, lavender arrows). Discs: – = negative control; p = penicillin; r = rifampicin; s = streptomycin; t = tetracycline; v = vancomycin. (B) Antibiotic discs placed on the surface of nutrient agar spread with isolate RCRC4244_C1. Some zones overlapped, making measurement difficult (for example, blue arrow). Discs: – = negative control; a = ampicillin; e = erythromycin; g = gentamicin; h = chloramphenicol; x = cefotaxime. (C) Bacterial growth on and surrounding Citrus hybrid RCRC4244 shoot tips (circled in red) and on recovery medium after removal of the shoot tips. Scale bar = 1 cm.

Figure 3.

Antibiotic disc diffusion assays on nutrient agar plates and bacterial growth present during in vitro culture of micrografted Citrus hybrid RCRC4210. (A) Antibiotic discs placed on the surface of nutrient agar spread with isolate RCRC4210_C1. Discs: – = negative control; g = gentamicin; r = rifampicin; s = streptomycin; t = tobramycin; v = vancomycin. (B) Antibiotic discs placed on the surface of nutrient agar spread with isolate RCRC4210_C1. Discs: – = negative control; b = tobramycin; k = kanamycin. (C) Bacterial growth present during in the vitro culture of micrografted Citrus hybrid RCRC4210 on the surface of citrus media (a), surrounding rootstock roots (b), in gas‐filled pockets surrounding the roots (c), and as a bluish tinge imparted to the growth media by the bacterial growth (d). Scale bar = 1 cm.

Identification

Endophytes have traditionally been identified in tissue culture using classical microbiological techniques (Leifert et al., 1994; Reed et al., 1997; Dunaeva and Osledkin, 2015); however, molecular techniques now offer the potential for rapid identification using a variety of methods (Souza et al., 2013; Lau and Botella, 2017; Gómez‐Lama Cabanás et al., 2021), including the real‐time sequence analysis (Kong et al., 2021) of microorganisms in cultures (Gómez‐Lama Cabanás et al., 2021), pools of cultures (De Souza et al., 2016), and directly from plant material (Chalupowicz et al., 2019).

Our bacterial isolates were sequenced using the Oxford Nanopore Technologies (ONT) MinION Nanopore sequencer with the What's In My Pot (WIMP) real‐time sequence analysis method from Oxford Nanopore (Oxford, United Kingdom). Briefly, DNA was extracted from pelleted cells collected from liquid cultures of the endophytes. Barcoded sequencing libraries were created as per the instructions for the SQK‐LSK108 sequencing kit with the EXP‐NBD103 native barcoding kit, and the pooled barcoded libraries were sequenced for 48 h using an ONT MinION sequencer with a FLO‐MIN106 flow cell and analyzed using the real‐time ONT WIMP protocol (revision 3.2.1) on the ONT EPI2ME platform.

Sequences from isolate RCRC4244_C1 showed very low homology to any organisms available in the National Center for Biotechnology Information (NCBI) RefSeq database used by the WIMP protocol (Juul et al., 2015). These sequences were then compared with a local copy of the NCBI non‐redundant sequence database using the MegaBLAST algorithm (Morgulis et al., 2008) and again showed very low homology, which meant RCR4244_C1 remained unidentified. Isolate RCRC4210_C1 was identified with high confidence to be Pseudomonas aeruginosa, but it was not possible to identify the specific strain.

Responses to antibiotics

The antibiotic responses of the identified bacterial contaminants were determined using a modified Kirby–Bauer disk diffusion assay (Balouiri et al., 2016). Selected bacteria were grown as liquid cultures in either Difco Nutrient Broth or liquid micrografting recovery medium (as above, without agar) overnight at 25°C, and then spread evenly over the surface of 100 × 15 mm plates containing nutrient agar (Difco Nutrient Broth with 1.5% agar) or micrografting recovery medium (Volk et al., 2012). BD Sensi‐Discs (Becton, Dickinson and Company) infiltrated with 12 different antibiotics (ampicillin [10 µg], chloramphenicol [30 µg], cefotaxime [30 µg], erythromycin [15 µg], gentamicin [10 µg], kanamycin [30 µg], penicillin [10 IU], rifampicin [5 µg], streptomycin [10 µg], tetracycline [30 µg], tobramycin [10 µg], and vancomycin [30 µg]) were placed on the plates. The plates were then incubated at 25°C for one to five days. When zones of inhibition were visible, the diameter of the clear area surrounding each antibiotic disc was measured (Figures 2 and 3, Table 1).

Table 1.

Zones of inhibition surrounding antibiotic‐impregnated discs on plates of RCRC4244_C1 and RCRC4210_C1 bacteria, showing their responses to antibiotic exposure.

	RCRC4244_C1	RCRC4210_C1
	Zone of inhibition (mm)	Zone of inhibition (mm)
Antibiotic	Nutrient agar	Nutrient agar	Recovery medium
Ampicillin (10 µg)	12	0	0
Chloramphenicol (30 µg)	12	0	0
Cefotaxime (30 µg)	20	0	0
Erythromycin (15 µg)	14	0	0
Gentamicin (10 µg)	10	8	14
Kanamycin (30 µg)	12	12	0
Penicillin (10 IU)	0	0	0
Rifampicin (5 µg)	14	12	16
Streptomycin (10 µg)	12	0	0
Tetracycline (30 µg)	14	20	16
Tobramycin (10 µg)	10	12	14
Vancomycin (30 µg)	0	0	0

The bacterial growth characteristics differed between the nutrient agar and micrografting recovery media plates, as well as across the antibiotic treatments. Isolate RCRC4244_C1, an unidentified bacterium, grew very slowly on the micrografting recovery medium, and its growth was insufficient for successful disc diffusion testing. When the assay was performed using nutrient agar, all antibiotics tested, except for penicillin and vancomycin, produced zones of inhibition, with cefotaxime having the greatest zone of inhibition (20 mm; Figure 2, Table 1). On the nutrient agar, the growth of isolate RCRC4210_C1, Pseudomonas aeruginosa, was inhibited by gentamicin, kanamycin, rifampicin, tetracycline, and tobramycin. The disc impregnated with tetracycline had the largest inhibitory zone of 20 mm. When the test was carried out on micrografting recovery medium, the inhibitory effect of gentamicin, rifampicin, and tobramycin increased slightly, while the effect of tetracycline decreased slightly, and the effect of kanamycin disappeared altogether (Figure 3, Table 1).

The next steps are to identify antibiotic treatment concentrations that are effective for bacterial elimination yet do not significantly damage the host plants (Cornu and Michel, 1987; Leifert et al., 1991). These tests should consider the effect of the plant culture conditions (e.g., pH, carbon source, nutrient availability, and temperature) on antibiotic responses (Falkiner, 1997). The antibiotic treatment could take place during culture initiation, or possibly during the shoot tip recovery process.

DISCUSSION

Plant endophytes prevent the success of many tissue culture efforts, including recovery after shoot tip cryopreservation. A number of different strategies have emerged to eliminate or minimize the deleterious effects of endophytes, some of which may be more effective than others, depending on plant species and cultivar, source of materials, propagation methods, and treatment tolerances. Due to the challenges resulting from contamination in tissue‐cultured plants, it may be worthwhile to thoroughly test cultures using microbial screening assays during the introduction and initial growth stages to confirm they do not harbor bacterial contaminants prior to investing labor and resources into long‐term culture maintenance and use. Although this adds significant upfront costs, the eventual cost of repeating introductions for contaminated cultures may be even higher. Antibiotic treatments may be necessary when new source materials are not available. With the increased availability of rapid DNA‐sequencing technologies, it has become possible to readily identify bacterial contaminants and then propose possible antibiotic regimes for targeted elimination strategies. This information, as well as testing the efficacy of antibiotics on the isolated bacteria, may streamline the clean‐up process to produce healthy plants.

Traditionally, tissue culture labs have sought to eliminate any signs of visible bacteria to improve micropropagation success (Thomas, 2010); however, some positive effects have been reported for endophytes in tissue‐culture plants (Murthy et al., 1999; Pohjanen et al., 2014; Quambusch et al., 2016; Kanani et al., 2020). For example, microbial symbionts play a role in plant responses to cold and freezing stresses (Acuña‐Rodríguez et al., 2020). A better understanding of plant–endophyte relationships may reveal opportunities to improve micropropagation, and possibly even shoot tip cryopreservation and other methods to achieve more vigorous and healthy plants.

AUTHOR CONTRIBUTIONS

G.M.V. and A.D.H. conceived the research and designed the experiments; R.B., A.C.A.O., and A.D.H. performed all the experiments. A.D.H. analyzed the data and G.M.V., R.B., A.C.A.O., and A.D.H. wrote the manuscript. All authors approved the final version of the manuscript.

Volk, G. M. , Bonnart R., de Oliveira A. C. A., and Henk A. D.. 2022. Minimizing the deleterious effects of endophytes in plant shoot tip cryopreservation. Applications in Plant Sciences 10(5): e11489. 10.1002/aps3.11489

DATA AVAILABILITY STATEMENT

Sequence data have been deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) database with BioProject PRJNA799443. Isolate RCRC4244_C1 is BioSample SAMN25166301, SRA accession SRR17698257, and Isolate RCRC4310_C1 is BioSample SAMN25166293, SRA accession SRR17698265.