Introduction

Propagating plants in vitro usually requires a seal for the culture vessel that prevents contamination and the dehydration of the explants and media. However, the type of seal or lid may limit gas exchange between the in vitro and ex vitro environments, leading to morphophysiological disorders that cause high plant mortality during acclimatization (Nguyen and Kozai 2005; Alvarez et al. 2012).

In conventional in vitro propagation systems, the most common seals used for culture vessels are threaded caps, aluminum foil, plastic film (polyvinyl chloride), and standard polypropylene lids, which all act as barriers, limiting gas exchange. In these systems, there are low transpiration and photosynthesis rates, restricted water and nutrient absorption, and high dark respiration rates, which reduces the explant growth rate (Kozai et al. 1997). In addition, the high relative humidity (RH) inside the culture vessel reduces the deposition of epicuticular waxes as well as the development of functional stomata, which can lead to losses during acclimatization (Chandra et al. 2010). For these reasons, it is important to aerate the vessel in order to minimize differences between the inner and outer atmospheres (Zobayed 2005, 2006).

The use of different seals, which allow greater gas exchange, may increase explant growth as well as enable a photoautotrophic propagation system. The modification of the microenvironment, brought about by gas exchange, maintains CO2 concentrations suitable for stimulating photosynthesis and reduces both the ethylene concentration and RH inside the culture vessel (Kozai and Kubota 2001).

In order to improve the ventilation of in vitro culture vessels, lids with membranes that allow higher gas exchange have been used, which reduced RH inside the vessels and increased the plants transpiration and water and nutrient absorption (Kozai 2010; Xiao et al. 2011). Additionally, the use of gas permeable membranes has reduced the hyperhydricity percentage of in vitro propagated plants (Kozai et al. 1997).

As early as 1965, a simple tissue-culture vessel permitting gas exchange was proposed (Kordan 1965). It was based on screw-cap vials with a centered hole (0.8 cm in diam) in the lid, covered with polyethylene film. Currently, several membranes are commercially available, which promote gas exchange in in vitro culture vessels (Zobayed 2005).

A range of species has been successfully propagated in vitro using MilliSeal® membranes to improve gas exchange, such as Azadirachta indica (Rodrigues et al. 2012), Solanum melongena (Ribeiro et al. 2009), Pfaffia glomerata (Iarema et al. 2012), Uniola paniculata (Valero-Aracama et al. 2007), Hypericum perforatum (Couceiro et al. 2006), Alha gigraecorum (Zobayed et al. 2006), and Eucalyptus tereticornis (ShaValli Khan et al. 2002. However, these membranes are expensive, limiting their commercial application. For this reason, reducing the cost of the membranes would be ideal for large scale, in vitro, autotrophic propagation projects (Nguyen and Kozai 2005).

Previous works in our laboratory found that Pfaffia has marked growth in response to increased gas exchange using commercial porous membranes. The goal of the present work was to compare the efficiency of low-cost membranes with MilliSeal® membranes by evaluating the morphogenesis and in vitro growth of Pfaffia glomerata (Spreng.) Pedersen. This species is a medicinal plant that produces phytoecdysteroids, and is popularly known in Brazil as fáfia or ginseng brasileiro. To the best of our knowledge, this is the first article to report on the combined use of microporous and polytetrafluoroethylene (PTFE) tapes as inexpensive, alternative membranes that can be used to promote gas exchange during in vitro propagation of chlorophyllous explants.

Materials and methods

Plant material and culture conditions

Nodal segments (average 2 cm) of Pfaffia glomerata (hereafter referred to as Brazilian ginseng) with two pre-existing axillary buds were excised from plantlets maintained in vitro on MS (Murashige and Skoog 1962) medium under photomixotrophic conditions (25 ± 2 ºC air temperature and 60 μmol m−2 s−1 irradiance provided by two tubular fluorescent lamps [cool white] with a 16-h day−1 photoperiod).

Four explants were inoculated without leaves in glass jars (12 cm high × 5 cm internal diameter) containing 50 mL of basal MS medium, MS vitamins (0.5 mg L−1 nicotinic acid, 0.5 mg L−1 piridoxine.HCl, 0.1 mg L−1 thiamine.HCl and 2 mg L−1 glycine), 100 mg L−1 myo-inositol and 30 g L−1 sucrose. All culture media were adjusted to pH 5.7 and solidified with 7 g L−1 granulated agar (Merck® KGaA, Darmstadt, Germany). The salts of MS medium were added to the culture medium as powdered pre-mix (Sigma Chemical Company, St. Louis, MO, USA, Cat no. M5519). The culture medium was sterilized by autoclaving at 121 °C and 1.5 atm for 15 min.

Comparison of sealing systems

The six treatments consisted of different types of flasks seals: (RP) autoclavable rigid polypropylene lid; (MJ) RP with two 10 mm-diameter holes covered with a fluoroporo hydrophobic membrane [Polytetrafluoroethylene (PTFE); MilliSeal® Air Vent, Tokyo, Japan]; (C2 and M2) RP with two 10 mm-diameter holes covered with a membrane (Fig. 1a) composed of two layers of tape microporous C (Cremer®) or M (Missner & Missner®) and a PTFE (Amanco®) of 0.05 ± 0.01 mm thickness; (C3 and M3) the same as for C2 and M2 (Fig. 1b) but comprised of three layers of microporous tape and PTFE (Amanco®). Plants grown in containers with RP were used as a negative control because this is the condition found in conventional in vitro propagation systems, whereas plants grown in flasks covered with MJ-sealed lids were considered a positive control (Fig. 2a). During the experiments no subcultures were performed.

Fig. 1
figure 1

Schematic diagram of the assemblage of the membranes with microporous tape and PTFE film. a Membrane with two layers of microporous tape C or M and a single layer of PTFE (Amanco®) (C2 and M2). b Membrane with three layers of microporous tape C or M, and a single layer of PTFE (C3 and M3)

Full size image
Fig. 2
figure 2

Aspect of the in vitro cultures of Pfaffia glomerata grown using different types of seals (after 35 days). a: Top view of the flask lids with different types of seals. b Overview of the development of in vitro-derived plants grown using the different seal types. c Detail of root system development as affected by the seal type. d Detail of the gelled medium retraction due to water loss as affected by the seal types. Bar = 3 cm. RP autoclavable rigid polypropylene lid. MJ RP with two orifices covered by the PTFE MilliSeal® membrane. M3, M2, C3 and C2: RP with two orifices covered with the M2, M3, C2 or C3 membrane

Full size image

All membranes were sterilized by autoclaving them in the same conditions with culture medium.

The amount of gas exchange provided by each membrane type was determined as described by Fujiwara and Kozai (1995). Therefore, the headspace of each flask containing a different membrane type was saturated with a mixture of carbon dioxide (CO2) at a concentration of 2.5 %. The readings of the inner CO2 concentration were made with a Headspace Gas Analyzer 6600 (Ilinois© Instruments, Johnsburg, IL, USA). The amount of gas exchange per hour (N’) for each sealing condition was estimated by the following equation: \( N^{\prime } = \frac{1}{T}\ln \frac{{C_{1} - C_{out} }}{{C_{2} - C_{out} }} \) (Fujiwara and Kozai 1995), where T represents the time (in hours) between readings 1 and 2; C1 and C2 correspond to the inner CO2 concentrations in the flasks at times 1 and 2; and Cout is the CO2 concentration in the environment external to the flask.

Water vapor loss rate (WVLR)

The flasks with the different seal types were characterized according to the water vapor loss rate (WVLR) at 25 °C and 30 % RH by means of the gravimetric method. For this, flasks sealed with rigid lids (RP), with or without membranes and 20 mL of MS medium, were weighed on an analytical balance (0.001 g) every 24 h and weight losses for a week were recorded.

Scanning electron microscopy

Three samples of each material used in the membranes for gas exchange were analyzed using a scanning electron microscope (Hitachi Tabletop TM 3000, Japan).

Growth parameters

The plant height (cm), leaf area (cm2 plant−1), leaf number, fresh and dry weight of shoot (SFW and SDW) and root system (RFW and RDW) (g. plant−1) were assessed after 35 days of in vitro cultivation. The samples for dry mass determination were dried until constant weight was reached (around 48 h). Seven plants per treatment were randomly sampled to estimate the leaf area.

Photosynthetic pigments

Determination of carotenoids and chlorophyll a and b followed Wellburn (1994). Four leaf discs (7 mm diameter) from each treatment were cut from the third fully expanded leaf (from the shoot tip) and placed in a vial with 5 mL DMSO solution (saturated with CaCO3) at room temperature for 48 h (Santos et al. 2008). Absorbance at 665, 645 and 480 nm was determined using a spectrophotometer (Genesys 10UV, Thermo Scientific).

Plant acclimatization

Fifteen plants of Brazilian ginseng of each treatment were acclimatized at 35 days by transferring them to an ex vitro environment, in plastic cups (300 mL capacity) that contained approximately 260 cm3 of a commercial substrate (Bioplanta®, São Paulo). The acclimatization stage was initially conducted in a laboratory, for 2 weeks, and the in vitro-derived plants were subsequently transferred to a greenhouse.

Statistical analysis

The characteristics were evaluated using Bartlett’s test for variance homogeneity and the results of dry matter and water vapor loss rate were transformed into \( \sqrt x \) and submitted to analysis of variance and Dunnett’s test at 5 % significance. All statistical analyses were performed with SAS 9.1 (SAS Institute 2003). The experimental design was a randomized block with six treatments and nine repetitions. Each experimental unit consisted of four explants per jar.

Results

Contamination was not observed in any of the treatments. The different membranes remained intact after being autoclaved five times.

Sealing types that allowed higher gas exchange resulted in higher growth

The type of seal significantly (p < 0.05) influenced gas exchange between the inner and outer environment of the flask, and consequently the growth of the in vitro-derived plants (Fig. 2).

At 35 days, shoot cultures grown in flasks sealed with membranes were higher than those grown in flasks that lacked membranes (negative control, RP). Among the membrane systems assayed there were no significant differences (p > 0.05) compared to the positive control-MJ (Fig. 3a), and the highest means of height were obtained for treatments MJ, M3, M2 and C3 (Fig. 3a).

Fig. 3
figure 3

Growth characteristics of Pfaffia glomerata plantlets propagated in vitro using different types of seals (after 35 days). a Height. b Leaf number. c Shoot fresh weight (SFW). d Root fresh weight (RFW). e Shoot dry weight (SDW). f Root dry weight (RDW). g Leaf area. Different letters denote statistical differences calculated with Dunnett’s test, at 5 % probability; upper case and lower case letters compare means with the negative (RP) and positive (MJ) controls, respectively. RP: autoclavable rigid polypropylene lid. MJ RP with two orifices covered by the PTFE membrane MilliSeal®. M3, M2, C3 and C2: RP with two orifices covered with the M2, M3, C2 or C3 membrane

Full size image

A decrease in leaf number per shoot was observed for flasks with lids that had porous membranes (Figs. 2d, 3b); however the treatment C2 did not differ from the negative control. The treatments with membranes did not differ from the MJ positive control (Fig. 3b). Therefore, the greatest number of leaves was recorded in flasks without membranes (RP), and the in vitro-derived plants grown in flasks with membranes (M2 and M3) had the lowest average number of leaves.

Average values for shoot fresh mass did not differ with treatments with porous membranes as compared with the positive control (Fig. 3c); however, the positive control differed from RP. Likewise, the fresh weight of the roots (RFW) did not differ (p > 0.05) among treatments with membranes and the control MJ (Fig. 3d), but MJ and M3 differed from the negative control, leading to a higher accumulation of RFW (Fig. 2c). Intermediate RFW values were found for treatments C2, C3 and M2, and did not differ (p > 0.05) from RP.

The accumulation of the shoot dry weight significantly (p < 0.05) increased in membrane-based treatments, which enabled gas exchange, although it differed significantly from the negative control RP (Fig. 3e). The higher mean values were obtained for M3 and M2, and did not differ from MJ. Again, means for C2 and C3 biomass accumulation presented intermediate values, and the lowest values were for the control RP. The dry matter accumulation in the root system of the in vitro-derived plants also increased (p < 0.05) in the presence of membranes (Fig. 3f), with the highest averages observed in M2 and M3, and did not differ from MJ. The lowest dry matter accumulation was observed for RP, however it did not differ from treatments C2 and C3.

Leaf area was significantly influenced (p < 0.05) by the type of seal used (Fig. 3g). The higher values were found for in vitro-derived plants grown using the MJ and C3 seals. Treatments C2, M2 and M3 had intermediate values, and the presence of the membrane did not promote significant increases in leaf area when compared to the control RP (Fig. 3g). The leaf area in in vitro-derived plants grown in flasks with membranes showed no significant difference with MJ (Fig. 3g). However, the shoots that developed in treatment RP had the lowest mean leaf area.

In general, the lower means for growth characteristics were observed for treatments lacking membranes (Fig. 3), except for the number of leaves and SFW. While in flasks with both higher gas exchange and water vapor loss rate, larger growth values for the evaluated traits were observed, except for leaf area (Fig. 3).

M3 provides similar water vapor loss rate in flasks compared to MJ

Significant differences were observed (p < 0.05) between the sealing systems evaluated during 7 days compared to the negative (RP) and positive (MJ) controls (Fig. 4). Higher WVLR were observed for flasks with membranes that showed greater permeability to water vapor, and M3 did not differ from the positive control MJ. The negative control (RP) presented the lowest WVLR. Among the new membranes evaluated, M3 performed better when compared to the PTFE membrane (MJ). The sealing system with M2 showed greater water loss, differing from MJ (Figs. 2b, d; 4). The seals using membranes C2 and C3 showed intermediate values of water loss, differing from both negative and positive controls. Treatment M2 performed better than M3, indicating that the number of layers of tape used in the preparation of microporous membranes influenced the WVLR. Importantly, as the WVLR increased the culture medium became more dehydrated. The amount of gas exchange per hour, which was estimated by the use of CO2 in each flask, was similar to WVLR, as follows: 0.03 (TR), 0.86 (MJ), 0.77 (M3), 0.76 (M2), 0.22 (C3), and 0.20 (C2).

Fig. 4
figure 4

Characterization of water vapor loss rate (WVLR) in flaks sealed with different membranes. The WVLR was determined at 25ºC and 30 % relative humidity. Different letters denote statistical differences calculated with Dunnett’s test, at 5 % probability; upper case and lower case letters compare means with the negative (RP) and positive (MJ) controls, respectively. RP: autoclavable rigid polypropylene lid. MJ: RP with two orifices covered by PTFE membrane MilliSeal®. M3, M2, C3 and C2: RP with two orifices covered with the M2, M3, C2 or C3 membrane

Full size image

Sealing types promote an increase in content of photosynthetic pigments

The content of photosynthetic pigments of leaves varied significantly (p < 0.05) among the types of seals in relation to the positive and negative controls (after 35 days of in vitro culture). The content of chlorophyll a among treatments ranged from 14 to 50 μg cm−2, chlorophyll b from 5 to 17 μg cm−2, total chlorophyll from 19 to 66 μg cm−2, and carotenoids from 3 to 9 μg cm−2 (Fig. 5). The best performances in relation to photosynthetic pigments were attributed to the MJ, M2 and M3 seals, which did not differ statistically (p > 0.05). The in vitro-derived plants cultured in flasks sealed with MJ and M3 membranes produced three times more chlorophyll than the RP treatment (negative control). The C2 and C3 systems showed intermediate performance, differing from the negative and positive controls (Fig. 5). The in vitro-derived plants grown in flasks sealed with RP had the lowest values for these characteristics. The estimate of the total chlorophyll content by SPAD showed a pattern similar to that estimated by extraction with DMSO (Fig. 5c, e).

Fig. 5
figure 5

Photosynthetic pigments from P. glomerata plantlets propagated in vitro under different gas exchanges (after 35 days of culture). a Chlorophyll a. b Chlorophyll b. c Total chlorophyll. d Carotenoids. e Chlorophyll content determined by SPAD. Different letters denote statistical differences calculated with Dunnett’s test, at 5 % probability; upper case and lower case letters compare means with the negative (RP) and positive (MJ) controls, respectively. RP autoclavable rigid polypropylene lid. MJ: RP with two orifices covered by PTFE MilliSeal® membrane. M3, M2, C3 and C2: RP with two orifices covered with the M2, M3, C2 or C3 membrane

Full size image

Microporous tape of M3 and M2 showed wider pores and less adhesive

When the membranes were analyzed using scanning electron microscopy (SEM), we found that the materials used in the production of the new membranes had very different physical characteristics, particularly in the arrangement, size and number of pores (Fig. 6a, b). The PTFE MilliSeal® membrane (MJ) had a uniform arrangement of PTFE fibers and smaller pores compared to the PTFE Amanco® membrane. However, the latter had a higher accumulation of residues on the fibers and larger pores (Fig. 6e, f).

Fig. 6
figure 6

Scanning electron microscopy (SEM) images of the surface of materials used in the membranes. a, c Adaxial surface of the microporous tape made by Cremer® (a) and Missner & Missner® (c), without adhesive (bar = 200 μm). b, d Abaxial surface of the microporous tape made by Cremer® (b) and Missner & Missner® (d), with adhesive (bar = 200 μm). e, f Polytetrafluoroethylene membrane by MilliSeal® (e) and Amanco® (f) (bar = 10 μm). Or orifice, Ad adhesive

Full size image

SEM revealed that the arrangement of fibers that make up the microporous tape was irregular (Fig. 6a, c). The microporous tape Missner & Missner® possessed wider pores, in higher numbers, on the non-adhesive side when compared to the same side of the Cremer® tape (Fig. 6b, d), which resulted in a greater capacity for gas exchange. On the Cremer® tape, the abundance of adhesive blocked the pores, which reduced the permeability.

No differences were observed in acclimatization for all conditions

During the acclimatization phase to ex vitro conditions, the in vitro-derived plants grown under different conditions showed 100 % survival. At the time of initial transfer to a substrate, a greater dehydration was observed in plantlets derived from the RP treatment. However, after 30 days of acclimatization there were no morphological differences in the in vitro-derived plants grown using the different seals.

Discussion

In this study we verified that the type of seal influenced the growth of Brazilian ginseng, and showed that higher gas exchange positively increased the growth of the in vitro-derived plants. Concurrently, the aeration of the culture flasks with PTFE commercial membranes (MilliSeal Air Vent®, Tokyo, Japan) favored the growth of Brazilian ginseng plantlets under both photomixotrophic and photoautotrophic conditions (Iarema et al. 2012). The stimulation of gas exchange between the external and internal environment by the low-cost, alternative membrane system favors the effective uptake by plants of nutrients from the culture medium, increasing the growth of explants (Arigita et al. 2010).

The present study showed that in vitro-derived plants had greater growth in treatments where there were greater losses of water vapor (Figs. 2, 3c). Moreover, the increase in gas exchange between the inner flask atmosphere and the external environment can reduce hyperhidricity in explants cultured in vitro (Ivanova and Van Staden 2010). Indeed, the in vitro-derived plants of Brazilian ginseng grown using different seals showed no symptoms of hyperhidricity.

In vitro propagation systems using membranes that allow gas exchange between the external and internal atmosphere of the flasks increases the natural ventilation, providing that the CO2 concentration is adequate, which results in an increased rate of photosynthesis and growth (Kozai 2010). In the present study, in vitro-derived plants of Brazilian ginseng grown in flasks with membranes showed higher dry matter accumulation in the shoots and roots (Figs. 2c, 3), indicating the importance of gas exchange in in vitro morphogenesis. Several studies report that an increase in gas exchange, in vessels used in in vitro plant propagation, improves the development of the explants (Sha Valli Khan et al. 2002; Couceiro et al. 2006; Zobayed et al. 2006; Valero-Aracama et al. 2007; Ribeiro et al. 2009; Iarema et al. 2012; Rodrigues et al. 2012).

In conventional in vitro propagation systems, the elevated humidity inside the vessel reduces transpiration rates and may reduce the transport of ions dependent on mass flow. The in vitro-derived plants of Brazilian ginseng, grown in sealed flasks with MJ and M3, showed an increase in the values of growth characteristics compared to other treatments, and these two types of seals enabled greater loss of water from the system (Fig. 4). Growth stimulation may be related to better nutrition of the in vitro-derived plants. This is related to the increased transport of ions absorbed from the culture medium and is also based on the flow of the xylem sap, which is stimulated by a decrease in relative humidity (in vitro) due to the loss of water because of transpiration. It is believed that an increase of transpiration in plant leaves is required to transport nutrients in the xylem sap over long distances (Tanner and Beevers 2001). Indeed, for P. glomerata it has been observed that an increase in gas exchange in the culture flasks provides the most effective conditions for the absorption of nutrients (Iarema 2008).

Regarding the content of photosynthetic pigments, an increase in gas exchange caused an increment in the biosynthesis of pigments in the leaves of the in vitro-derived plants of Brazilian ginseng. Similar results have been reported for other species (Ivanova and Van Staden 2010; Mohamed and Aldason 2010). A decrease in chlorophyll content in explants cultured in flasks with reduced gas exchange may be related to a high accumulation of ethylene, as demonstrated using Brassica oleracea where chlorophyll content was increased by inhibiting ethylene with silver nitrate (10 μM) and increasing gas exchange in the culture flask (Zobayed et al. 1999). Here, the lower content of photosynthetic pigments in the Brazilian ginseng in vitro-derived plants was associated to the sealing systems that provided lower moisture losses from the flasks (Fig. 2). This suggests that less water loss from the vessels results in more moisture in the microenvironment of the flasks as well as reduced content of photosynthetic pigments. Studies with other plant species support this hypothesis, where the increase in relative humidity in vitro can reduce the content of photosynthetic pigments (Chanemougasoundharam et al. 2004; Cha-um et al. 2010).

The reduced content of photosynthetic pigments in in vitro-derived plants can commit the acclimatization to ex vitro conditions (Chanemougasoundharam et al. 2004). The accumulation of chlorophyll in the Brazilian ginseng in vitro-derived plants grown in flasks with membranes M3 and MJ was similar. Despite the acclimatization rates of 100 % survival of the in vitro-derived plants in all treatments, greater growth in vitro may result in higher vigor during acclimatization due to favorable morphophysiological traits.

The microporous tape Missner & Missner® (M3 and M2) possessed wider pores, in higher numbers on the non-adhesive side when compared to the same side of the Cremer® tape (C3 and C2), as well as less adhesive on the other side (Figs. 6b, d), which resulted in a greater capacity for gas exchange. The presence of more adhesive on the Cremer® tape resulted in the blocking of pores, which reduced gas exchange.

The results reported here reinforce other reports in the literature on the effects of gas exchange in morphogenesis in vitro, and widen the possibilities of using alternative and efficient materials to promote in vitro growth of plants by increasing gas exchange during in vitro propagation.

Among the membranes tested, the M3 membrane was similar to the MJ membrane when comparing the traits of Brazilian ginseng in vitro-derived plants cultivated in vitro. However, the M3 membrane has a unit cost of production that is ten times lower than the commercial MJ membrane. The cost of the M3 membrane is around US $0.082 while the commercial membrane costs US $0.860. In addition, the homemade membranes proposed here are easily prepared and used, which avoids issues related shipping times for imported goods, such as the MJ membranes. Thus, the use of the M3 membrane to promote gas exchange in in vitro culture systems is attractive, because of its low cost, making it a possible alternative for large-scale, in vitro commercial propagation systems that are based on photomixotrophic or photoautotrophic growth patterns.