Plant Cell Rep (2005) 24: 201–208 DOI 10.1007/s00299-005-0931-2
CELL BIOLOGY AND MORPHOGENESIS
Stephen C. Van Winkle · Gerald S. Pullman
Achieving desired plant growth regulator levels in liquid plant tissue culture media that include activated carbon
Received: 29 October 2004 / Revised: 6 January 2005 / Accepted: 18 January 2005 / Published online: 10 March 2005 C Springer-Verlag 2005
Abstract This paper is part of a series considering the impact of activated carbon (AC) on the composition of plant tissue culture media. Using liquid culture media for initiation of Norway spruce embryogenic tissue and eight different ACs, we present a method for achieving target plant growth regulator (PGR) levels in AC-containing medium based on sorption isotherms for individual PGRs. Linear relationships were found between PGR adsorption and specific BET (Brunauer, Emmett, Teller theory) surface area and specific total pore volume of AC. When using a new AC, this linear relationship allows one to achieve multiple PGR levels similar to historic levels through adjustment of the mass of AC based on its relative BET surface area or relative total pore volume. Target levels of PGRs and an initiation success similar to that in medium without AC were achieved with several different AC types when AC mass was adjusted on the basis of pore volume. Keywords Tissue culture . Charcoal . Adsorption . Somatic Embryogenesis Abbreviations AC Activated carbon BAP 6-Benzylaminopurine; 2,4-D 2,4-Dichlorophenyloxyacetic acid; NAA α-Naphthaleneacetic acid Communicated by G. C. Phillips S. C. Van Winkle · G. S. Pullman () Institute of Paper Science and Technology, Georgia Institute of Technology, 500 10th Street, NW, Atlanta, GA 30332-0620, USA e-mail:
[email protected] Tel.: +1-404-8945307 Fax: +1-404-8944778 e-mail:
[email protected] Present address: S. C. Van Winkle International Paper, 6283 Tri-Ridge Blvd., Loveland, OH 45140, USA
Introduction Many laboratories use activated carbon (AC) in plant tissue culture medium to improve growth or response in vitro (see Pan and van Staden 1998 for a review). AC was originally added to growth medium as a means to darken the medium and simulate soil conditions, but it has also been reported to improve most tissue culture responses including somatic embryogenesis, organogenesis, adventitious shoot production, shoot growth, and the rooting of micro-propagated tissues. The inclusion of AC in the growth medium has also had detrimental effects, such as the inhibition of root formation and the adsorption of various medium components (George and Sherington 1984; Zaghmout and Torello 1988; Pullman and Johnson 2002). A further complication is that different ACs may have different adsorption characteristics, consequently impacting the growth response in vitro differently (Pan and van Staden 1998). In addition to darkening the medium environment, the various mechanisms involved in the stimulatory or inhibitory effects of AC include: the release of substances, alteration of medium pH, adsorption of undesirable or inhibitory substances, and the adsorption of vitamins, metal ions, plant growth regulators (PGRs), and hormones, including abscisic acid and gaseous ethylene (Pan and van Staden 1998; Menzuali-Sodi et al. 1993; Van Winkle and Pullman 2003; Van Winkle et al. 2003; Pullman et al. 2005). While beneficial effects may occur when AC is added to the medium, the adsorption of specific desired ingredients may prevent these beneficial effects. Pullman and Johnson (2002). and Van Winkle et al. (2003) demonstrated that the adsorption of copper and zinc ions by AC required the addition of supplementary copper and zinc sulfate to avoid any inhibition of initiation and growth of embryogenic tissue cultures of loblolly pine. Pullman and Gupta (1991, 1994) combined 1.25–2.5 g/l AC with high levels of abscisic acid or 2,4-D, BAP, and kinetin to obtain the desired levels of PGRs for embryogenic tissue culture initiation and embryo maturation for several coniferous species. Toering and Pullman (2005) used multiple regression models that
202 Table 1 Activated carbon characterization summary
Activated carbon
Lot number
Moisture content (%)
Total pore volume (cm3 /g)
BET surface area (m2 /g)
Sigma C-5260 (N1) (N2) Sigma C-9157 (T1) (T2) Norit S×2 Norit S×4 Darco G60 Merck (Acid-washed, non-USP)
124563 124677 1723656 672894 1739890 2873499 35789 536782
7.4 5.6 8.3 8.4 3.7 8.2 7.8 5.6
0.540 0.482 0.833 0.730 0.535 0.442 0.679 0.645
702 753 1,105 949 736 668 931 872
inputted initial 2,4-D, AC concentration, and time in days to explain 85–88% of the variation in available PGRs in the medium. A problem frequently encountered when using AC is the ability of AC to adsorb plant hormones and growth regulators. Many researchers using AC are unaware of the levels of hormones or PGRs that are actually available to the plant tissue. Most often the successful application of AC is with in vitro procedures for rooting, shoot elongation, or anther culture. The common factor in all of these applications is that the removal or decline of exogenous hormone is usually desirable. Obtaining target levels of adequate free exogenous hormone may be difficult. For example, when Ebert and Taylor (1990) added 2.5 g/l AC to agar or Phytagelsemisolid medium along with 22.1 mg/l 2,4-D, after 3 days only about 1 mg/l was available to the plant tissue. Even less hormone was available in liquid medium. Similar results were obtained with BAP (Ebert et al. 1993). When 11.3 mg/l BAP was added to gelled medium, less than 2% of the BAP was available to plant tissue after 3 days. Further, the amount of BAP added to the medium influenced the availability of 2,4-D. The main objective of the research reported here was to gain a better understanding of the adsorption of multiple PGRs such that target exogenous concentrations can be achieved. In our work the levels present in medium without AC, 2 mg/l 2,4-D and 1 mg/l BAP were targeted to be available in medium containing moderate to high amounts of AC. A further objective was to develop a methodology to compensate media to accommodate different ACs. This need arises regularly in the laboratory when the AC supply is depleted and a new supply used. Materials and methods AC characterization Six different grades of AC were used. Two of these were supplied by Sigma in the form of an untreated powder (C-5260; St. Louis, Mo.), designated “N” type, and as an acid-washed, tissue-culture tested powder (C-9157), designated “T” type. Two production lots were used, designated “1” and “2”, respectively. Other ACs included Norit Sx2, S×4, and Darco G60 (J.T. Baker), and Merck, acid-washed non-USP grade (EMD Chemicals, Gibbstown, N.J.). The
Brunauer, Emmett, and Teller (BET) specific surface area (Shaw 1992) and total pore volume were determined by nitrogen adsorption (analyzed by Westvaco, Charleston, S.C., using a Coulter, Omnisorp 100; Coulter Electronics, Hialeah, Fla.). Moisture content was calculated from mass loss after a period of 12 h at 150◦ C. These data are summarized in Table 1. Medium and explant preparation To facilitate the rapid removal of AC from the medium by filtering and the study of AC medium component adsorption, we developed a liquid initiation medium for Norway spruce (Pullman and Peter 2002; Van Winkle and Pullman 2003). The culture media were formulated from tissue-culture grade reagents supplied through Sigma, with the exception of casamino acids, which were supplied through Difco (Detroit, Mich.). Media were formulated using nanopure deionized water. The liquid media were based on media for initiating somatic embryogenic cultures from Norway spruce, as reported by Verhagen and Wann (1989), a modified 1/2-strength Brown and Lawrence (BLG) formulation (Amerson et al. 1985) with the pH adjusted to 5.8. A control medium without AC is shown in Table 2. The AC Basal formulation (Table 2) was used for all media containing AC. Initial experiments included AC (T1) at 1.25 g/l and elevated PGRs at 100 mg/l 2,4-D and 90 mg/l BAP. Subsequent media with AC included the AC Basal medium as shown and varying AC and PGR levels as noted in the text. Media and explants were prepared as described in detail in Van Winkle (2000). Embryogenic tissue was initiated from mature seed (F.W. Schumaker Co.) embryos of Norway Spruce (Picea abies L., Karst.) as described by Van Winkle and Pullman (2003). Cultures were grown on 10 ml medium dispensed into 15-ml polystyrene Petri plates and sealed with Parafilm. Support to the growing cultures was provided by a polyester pad (Poly-fil batting; Fairfield Processing, Danbury, Conn.)/black mixed cellulose ester membrane (Ahlstrom no. 8613–0425, 47 mm) combination. Initiation occurred when an explant produced one or more somatic embryos observed using an optical dissecting microscope. Data were analyzed by analysis of variance, and significant differences between treatment means were determined by the Least Significant Differences Multiple Range Test
203 Table 2 Comparison of liquid culture media formulations Component (mg/l) Mineral components (mg/l) pH 5.8 KNO3 KH2 PO4 KCl CaCl2 ·2H2 O MgSO4 ·7H2 O KI H3 BO3 MnSO4 ·H2 O ZnSO4 ·7H2 O Na2 MoO4 ·2H2 O CuSO4 ·5H2 O CoCl2 ·6H2 O FeSO4 ·7H2 O Na2 EDTA Organic components Sucrose myo-Inositol Casamino acids l-Glutamine Thiamine HCl Pyridoxine HCl Nicotinic Acid l-Asparagine Activated carbon 2,4-D BAP
AC Basala Without ACb 50 85 372.5 220 160 0.415 3.1 8.45 8.6 0.125 0.25 0.0125 13.9 18.65
50 85 372.5 220 160 0.415 3.1 8.45 4.3 0.125 0.0125 0.0125 13.9 18.65
10,000 50 500 750 0.15 0.15 0.75 50
10,000 50 500 750 0.05 0.05 0.25 50 0 2 1
PGR analysis The liquid growth medium samples for PGR analysis were collected using a sterile, disposable syringe and then passed through a syringe filter (Acrodisc, 0.2 µm) to remove the AC. The first few milliliters through the filter were discarded. The samples were analyzed by high-pressure liquid chromatograph (HPLC; HewlitPackard 1090) equipped with a UV diode array detector allowing monitoring at two different wavelengths (230 nm and 270 nm). Separation was achieved using a reversed phase column (Sigma-Aldrich Sperisorb ODS-25 µm, 250×4.6 mm) with isocratic elution. The mobile phase consisted of a 50/50 methanol/aqueous KOH-acetate buffer (0.01 M) adjusted to pH 6.0. At a flow rate of 0.5 ml/min, 2,4-D eluted near 9.3 min and BAP eluted near 24 min. Results and discussion
c c c
a
Liquid: modified 1/2-BLG (Verhagen and Wann 1989); all ACcontaining media included elevated Cu (20×), Zn (2×), and vitamins (3×) b Liquid: modified 1/2-BLG (Verhagen and Wann 1989) c 2,4-D, BAP and AC levels varied as indicated in the text
at the 5% level of significance using statgraphics plus v4.0. For the AC adsorption experiments, media were formulated at double strength and adjusted to pH 5.8. AC was added to a separate, equal volume of water, and the pH adjusted to 5.8. After combining the AC slurry and medium into a standard Erlenmeyer flask, we again monitored the suspension and adjusted for pH prior to autoclaving. Control media were diluted with water to the same extent as pH-adjusted media containing AC. Autoclaving was performed for 22 min at 122◦ C. After autoclaving, the flasks were promptly removed to a shaker table in a thermostated room maintained at 21◦ C. Shaker table speed was set to ensure that AC moved within the flask, but not so high as to create a vortex within the flask. Samples for hormone analysis were taken after 2.5 days, prior to the addition of glutamine (our practice is to add filter-sterilized glutamine to the medium just prior to dispensing media to small Petri plates). Experimental treatments were typically replicated three times.
Preliminary initiation results Prior experiments in this laboratory using semi-solid media indicated that embryogenic tissue initiation in Norway spruce varied with AC type (data not shown). The initial tissue culture experiments with liquid media were performed using a growth medium for Norway spruce initiation that included the AC Basal formulation (Table 2) plus AC-T1 at 1.25 g/l, 100 mg/l 2,4-D, and 90 mg/l BAP. Low initiation percentages were observed with this AC-containing media; less than 20% for medium with AC and high PGR levels versus more than 40% initiation for the control medium without AC. Preliminary PGR adsorption results An adsorption experiment was carried out using a set of eight different ACs. Five of these were carbons that were referenced in the tissue culture literature. Sigma C-9157 is a tissue-culture grade AC that we have used extensively; Sigma C-5260 is a non-acid washed carbon that has also been used in our laboratory. Two different samples from different production lots were compared for each of the two Sigma AC grades. Media contained AC Basal, 100 mg/l 2,4-D, 90 mg/l BAP, and 1.25 g/l of a test AC. Figure 1 shows the available PGR as a function of BET surface area, which for these carbons, ranged from 665 m2 /g to 1,105 m2 /g (Table 1). For each carbon type, available BAP after 2.5 days in the liquid phase was significantly lower than the approximate 1 mg/l measured at the same time in control medium without AC. For three of the carbons, the 2,4-D concentrations were close to the control level (2 mg/l). Neither BAP nor 2,4-D changed more than 5% as a result of autoclaving or time. A different adsorption experiment was carried out where AC-containing medium was prepared with 0.3 g/l AC (T1), 65 mg/l BAP, and 12 mg/l 2,4-D. Medium without AC was also measured after autoclaving at time zero. During the
204
natural logarithmic (ln) function. The slope of the ln function varies with the inverse of time: the slope decreases with time but never reaches zero, or true equilibrium. The adsorption period of 2.5 days represented a practical compromise between the common laboratory practice of using media the day after formulation and the more ideal practice with respect to sorption equilibria of allowing 7 days or more for adsorption to occur. Single-hormone isotherms (apparent)
Fig. 1 Available BAP (a) or 2,4-D (b) versus BET surface area in liquid tissue culture medium formulated with eight different activated carbons. Media contained 1.25 g/l AC, 100 mg/l 2,4-D, and 90 mg/l BAP. Data were collected after 2.5 days by HPLC. Each data point represents a single replication; some replications overlap for a carbon
Fig. 2 Available BAP (a) or 2,4-D (b) versus time in tissue culture growth media formulated with 0.3 g/l AC (T1), 65 mg/l BAP and 12 mg/l 2,4-D. Data points at time zero indicate AC-control samples (media without AC) after autoclaving. Error bars represent the 90% confidence interval based on three replications
first day following formulating of the AC-containing media, over 90% of the initial BAP and 75% of the initial 2,4-D were adsorbed (Fig. 2). The 2,4-D and BAP concentrations gradually declined further over the remainder of the 36-day period examined. Adsorption appeared to follow a decay-curve response, which is often fit using a
Three ACs (T1, Merck, and Sx4) were chosen for further study to represent the range in specific BET surface area depicted in Fig. 1. Adsorption isotherm data were collected for each PGR for the three ACs. Data were generated from separate trials with media containing a single hormone after an equilibration period of 2.5 days. Available BAP was measured for media containing 10, 50, 75, and 100 mg/l BAP, with each carbon present at 0.3 g/l. Additional BAP sorption points were generated with 30 mg/l BAP and 0.3 g/l Sx4 AC and 100 mg/l BAP with 0.2 g/l of T1 or Merck AC. The available 2,4-D data were generated with 10, 20, 40, and 80 mg/l of 2,4-D and 0.3 g/l AC. A liquid-solid adsorption isotherm is a graphical depiction of an equilibrium mass balance for a solute removed from bulk solution to the solid/liquid interface of a sorbent at constant temperature. It is commonly depicted as specific adsorption (e.g., milligram sorbate per gram sorbent) as a function of the bulk phase equilibrium concentration. For the work presented here, the data collected at 2.5 days were treated as if they were equilibrium data, and a constant temperature was assumed. Actual temperature was representative of typical practice, and air temperatures fluctuated over a range of 3◦ C. The term “apparent” is applied here to distinguish our data from true equilibrium isotherm data. Isotherm data are most often collected from pure solutions or relatively simple systems—i.e., a series of aqueous solutions with increasing ionic strength. For this research, the adsorption isotherm data for individual PGRs were collected from complete media to avoid the analytical complexities of modeling the impact on adsorption for the various medium components. Apparent isotherms for BAP and 2,4-D adsorption onto three different ACs are presented in Fig. 3. For BAP, the apparent isotherm was based on six BAP-AC combinations for each AC type, whereas, for 2,4-D five 2,4-D-AC combinations were sufficient. The HPLC data for 2,4-D were in very good agreement with previous data generated over a broader range using UV measurements on partial media (Van Winkle 2000), although lower levels of AC were employed due to a solubility limit for BAP. Because successful tissue culture media for somatic embryogenesis typically include PGR levels below 5 mg/l, additional isotherm data were unnecessary. The data in Fig. 3 were fitted using the Freundlich isotherm equation (Shaw 1992): x/m = k F cn/1
205
Fig. 3 Apparent isotherms representing adsorbed BAP/g AC versus available BAP (a) and adsorbed 2,4-D/g AC versus available 2,4-D (b) in liquid tissue culture media. Data were generated from separate trials with media containing a single hormone after an equilibration period of 2.5 days. The BAP data were generated from solutions of 10, 50, 75, and 100 mg/l BAP for each carbon present at 0.3 g/l. Additional BAP sorption points were generated from solutions with 100 mg/l BAP and 0.2 g/l AC for T1 and Merck AC and 30 mg/l BAP for 0.3 g/l of Sx4 AC. The 2,4-D data were generated from initial concentrations of 10, 20, 40, and 80 mg/l 2,4-D with 0.3 g/l AC. Error bars represent the 90% confidence interval based on three replications. The data have been fit using the Freundlich isotherm equation. Note that the y-axes are drawn to different scales
where x/m is the adsorption of solute per gram of AC, c is the apparent equilibrium concentration remaining available in solution, and kF and n are empirically derived constants related to sorptive capacity and strength of adsorption, respectively. Comparing the y-axis scales from Fig. 3, it may be seen that for each AC BAP was adsorbed to a level of about twice that of 2,4-D at any apparent-equilibrium point. The relative affinity and sorptive capacity of the carbons for the two PGRs were reflected in the magnitudes of the respective Freundlich constants (Table 3).
gram of AC on a per-liter basis. Typical tissue culture practice incorporates AC levels of 1.25 g/l or 2.5 g/l. An estimate of the BAP concentration necessary to accommodate 1.25 g of carbon T1 would be 1.25×225 mg/l or 281 mg/l plus the 1 mg/l remaining available. Similarly, an AC level of 2.5 g/l would require an initial BAP concentration of about 564 mg/l. Acid (HCl) is commonly used to dissolve BAP to formulate stock solutions of 1 mg/ml. The pH of this solution was approximately 3. We observed precipitation when the pH of a BAP stock solution was raised to 5.8, the target pH for the medium. The BAP remaining in solution was measured by UV270 absorbance as 165 mg/l. Based on the BAP isotherm for T1, this solubility limit imposed an experimental “mass ceiling” for AC T1 of about 0.66 g/l. It should be noted that complete media are complex solutions, and it is expected that the solubility of BAP in complete media at pH 5.8 differs from that in pure stock solution. Gelling agents may also impact solubility. In contrast, the solubility for 2,4-D was not a limitation as it exceeded 1 g/l across the pH range of interest. Relative sorption and partitioning behavior of solutes have been related to their activity in the aqueous phase (Schwarzenbach et al. 1993). Aqueous solubility is also a manifestation of that activity and may prove to be a predictor of sorptive potential for the various tissue culture medium components (Van Winkle 2000). Based on the solubility of BAP, further experimentation proceeded using media formulated with a reduced level of AC. Initiation results with the Norway spruce bioassay suggested that when AC was added at levels as low as 0.1 g/l, adsorption of a necessary component occurred resulting in reduced initiation (Table 4). An AC level above 0.1 but less than 0.66 g/l was desired: 0.3 g/l was chosen as a compromise level that still allowed formulation of double-strength BAP-containing media, which was useful during media formulation. It should be noted that our tissue-culture grade AC contained moisture levels of about 7.2%; 0.3 g/l nominal was equivalent to 0.28 g/l on an oven-dry basis. Approximation of initial BAP needed to achieve target (1 ppm BAP)
Achieving target PGR levels (2 ppm 2,4-D; 1 ppm BAP)
Using a carbon level of 0.3 g/l, we made an initial attempt to achieve target hormone levels of 1 mg/l and 2 mg/l for BAP
Referring again to the apparent isotherms (Fig. 3), the BAP trace for AC T1 indicates that to achieve an available level of 1 mg/l after 2.5 days, 225 mg/l would be adsorbed per
Table 4 Norway spruce initiation success after 9 weeks for liquid media varying in AC content
Table 3 Results from Freundlich curve fitting of apparent isotherm data Freundlich constants Carbon type BAP n
kf
2,4-D N
kf
Norit S×4 Merck TI (Sigma)
100 198 252
4.9 5.6 4.8
52 78 89
5.4 9.9 10
Replication Withouta 10 mg/l AC AC
50 mg/l AC
100 mg/l 250 mg/l AC AC
1 2 3 Mean
4 1 5 3.3 a
0 0 0 0.0 b
a
3 6 5 4.7 a
5 3 5 4.3 a
0 0 0 0.0 b
Liquid media (designated “Without AC”, Table 2) were used with 2 mg/l NAA, 1 mg/l BAP, and standard vitamin levels. Ten explants per replication were used. Means followed by the same letter are not significantly different at P=0.05 based on the multiple range test
206 Table 5 PGR levels and resultant adsorption to reach target levels of 2 mg/l 2,4-D and 1 mg/l BAP with AC present at 0.3 g/l Plant growth regulator
Activated carbon typea S×4 Merck T1
Initial BAP (mg/l) Adsorbed BAP (mg/g AC) Initial 2,4-D (mg/l) Adsorbed 2,4-D (mg/g AC)
32 114 8 19.4
55 118 10 26.4
65 234 12 34.9
a
Fig. 4 Available 2,4-D versus available BAP in medium containing T1 AC with three PGR combinations predicted from apparent isotherm data to provide target levels of PGRs. Trials were conducted using complete media with AC (0.3 g/l). Media were formulated with initial concentrations of 68 mg/l and 65 mg/l for BAP and 2, 10, and 12 mg/l for 2,4-D. Data were collected after 2.5 days. The target levels were chosen as 1 mg/l BAP and 2 mg/l 2,4-D based on successful media without AC. Each data point represents a single replication. The axes indicate available PGR: a functional relationship between the axes is not intended
and 2,4-D, respectively. Based on the findings with respect to the relative levels of adsorption between BAP and 2,4D, the first approximation was made assuming that there was no competition between the PGRs—i.e., the entire AC sorption volume was occupied by BAP. At an available PGR level of 1 mg/l, the adsorption of BAP was estimated to be 225 mg/g AC. Allowing for one remaining mg/l of BAP in solution, 0.3 g/l AC resulted in a calculated initial concentration of 68.5 mg/l. The initial concentration of 2,4D was 2 mg/l, assuming that no adsorption would occur. The results from this and subsequent trials using different initial PGR levels are depicted in Fig. 4. The data in Fig. 4 are presented in a way that allows both hormone levels to be observed: a functional relationship between the axes is not intended. The “target” PGR levels are defined by the intersection at 1 mg/l BAP and 2 mg/l 2,4-D. The initial experiment with 0.3 g/l AC, 68 mg/l BAP, and 2 mg/l 2,4-D resulted in solutionphase average PGR levels, based on two replications, of about 3.25 mg/l and 0.6 mg/l, for BAP and 2,4-D, respectively, and were close to, but not quite equal to the 2.5-day targets. Subsequent initial combinations were chosen by studying the slope of the BAP isotherm and trial and error with respect to 2,4-D. Medium with T1 AC at 0.3 g/l, 65 mg/l BAP, and 10 mg/l 2,4-D averaged 0.75 mg/l available BAP and 1.25 mg/l available 2,4-D. The successful combination of initial PGR levels that produced approximate target levels was found to be 65 mg/l and 12 mg/l for BAP and 2,4-D, respectively, averaging 0.93 mg/l available BAP and 1.85 mg/l available 2,4-D after 2.5 days. The process was repeated for the two remaining ACs for which apparent isotherm data had been generated. The successful initial PGR levels are presented in Table 5. The sorption data were based on the actual PGR levels present in the control, rather than the nominal dispensed levels, and the oven-dry mass of AC (moisture content varied from about 4.5% to 8%).
The sorption data have been corrected to reflect 2.5-day PGR levels actually measured in the control sample (approximate 5% loss in 2,4-D) and AC moisture content (4.5–8%)
Correlation between porosity and adsorption of PGRs The sorption data, corrected for moisture content, are presented in Figs. 5 and 6 as functions of specific total pore volume and specific BET surface area. An excellent coefficient of determination, R2 , between porosity and 2.5-day adsorption was found.
Fig. 5 Sorption of 2,4-D and BAP as a function of total AC pore volume. Data are shown for three different ACs. Media included AC at 0.3 g/l and the initial PGR levels listed in Table 5, and were complete except for glutamine. PGR levels were measured 2.5 days after medium preparation
Fig. 6 Sorption of 2,4-D and BAP as a function of BET surface area. Data are shown for three different ACs. Media included AC at 0.3 g/l and the initial PGR levels listed in Table 5, and were complete except for glutamine. PGR levels were measured 2.5 days after medium preparation
207
Extending the correlation with porosity
Table 6 Norway spruce initiation success after 9 weeks for media without and with two different types of activated carbon
The isotherm method of achieving target levels, though effective, is cumbersome in practice. The sorption versus porosity data suggest that the mass of a new AC could be adjusted on the basis of relative BET surface area or total pore volume to achieve the same sorptive potential with little or no adjustment in the initial PGR levels. This correlation between sorption and porosity was explored further using additional ACs. Pore volume and BET surface area measurements for these carbons are presented in Table 1. The range in specific BET surface area extended from 665 m2 /g for Norit Sx4 to 1,105 m2 /g for Sigma C9157 (T1). Significant variation between production lots was noted and is normal for common commercial AC grades. This range in surface area is typical of carbons used commercially for water treatment (Baker et al. 1992). Note that the total pore volume gave a different ranking for carbons with lower surface area. An adsorption experiment was conducted with media containing 65 mg/l BAP, 12 mg/l 2,4-D, and AC mass levels that were predicted to give a similar sorptive potential to that of T1, based on the ratios of total pore volumes. Results for seven different ACs are presented in Fig. 7. The overall achievement of PGR target levels was quite good. Two carbons (N2 and Sx2) disagreed slightly and displayed somewhat elevated 2,4-D levels. Because both PGRs were high for N2, simply increasing the AC mass would improve the agreement. The ratio of N2 to T1 mass to achieve better agreement was estimated to be 1.8, as interpolated from the isotherms and pore volume data. For Sx2, the elevated 2,4-D was attributed to an elevated pH level: the Sx2 pH drifted to 6.3, which was significantly higher than that of the other ACs, which averaged 5.5. The importance of pH will not be discussed here but has been considered in a previous investigation (Van Winkle 2000). It should be noted that the ash content of AC may have a significant impact on the adsorption of organic compounds from a solution (Diamadopoulos et al. 1992). For tissue culture applications, there are additional concerns when
Replicationa
Without AC
T1 AC
S×4 AC
1 2 3 Mean
5 7 7 6.3 a
8 3 5 5.3 a
5 8 7 6.7 a
Fig. 7 PGR levels available in media formulated with different AC types and masses. Using T1 initial PGR levels and AC mass, adjustments for other carbon masses were based on ratios of pore volumes between the test AC and T1 (e.g., the mass of T2 was 1.14-fold the mass of T1). BAP and 2,4-D initial levels were 65 and 12 mg/l, respectively, with target levels of 1 and 2 mg/l, respectively. Data were collected after 2.5 days. Error bars depict the 90% confidence interval based on three replications
a
Replications consisted of ten explants. Media with AC contained 0.3 g/l and were formulated with initial PGR levels that resulted in similar levels to the “Without AC” after 2.5 days (see Tables 2 and 5). Means followed by the same letter are not significantly different at P=0.05 (ANOVA)
using AC with a high ash content with respect to the mineral composition of the medium. Acid washing removes much of this ash content from AC. For the carbons described in this paper, ash content was not a significant concern (Van Winkle 2000). Tissue culture confirmation Experiments were conducted with the Norway spruce initiation bioassay using two different ACs (T1, Sx4; 0.3 g/l each) and the initial PGR levels that resulted in 1 mg/l and 2 mg/l of available BAP and 2,4-D, respectively, after 2.5 days. Initiation results are shown in Table 6. The initiation success for the AC-containing systems was greatly improved from that of the initial trials with 1.25 g/l T1, relative to the control (initial trials gave one-half of the 40% initiation that occurred with the control). With the adjustments in AC type and mass, Norway spruce initiation did not differ significantly (P=0.05) between medium with and without AC present. Conclusion The preliminary experimental levels of AC (1.25 g/l), BAP (90 mg/l), and 2,4-D (100 mg/l) resulted in low initiation and sub-optimal levels of available BAP versus the control level of 1 mg/l. BAP adsorbed to about twice the level of 2,4-D for any given PGR concentration. The low solubility of BAP in stock solution at pH 5.8 (165 mg/l) imposed a practical experimental limit on the mass of AC that could be added to the medium. Adsorption isotherm techniques were applied to the liquid Norway spruce embryogenic tissue initiation medium. Though this was a complex, non-equilibrium system, the equilibrium-based approach was instrumental in achieving desired hormone levels after 2.5 days and resulted in the production of target levels of 1 mg/l and 2 mg/l for BAP and 2,4-D, respectively, in media containing three different ACs (0.3 g/l). Tissue culture experiments using the improved media with different ACs resulted in significantly improved initiation success rates. PGR adsorption correlated well with both AC pore volume and BET surface area. By adjusting the mass of AC,
208
based on the ratio of specific total pore volumes or specific BET surface areas of the carbons, similar available PGR concentrations were obtained in liquid medium containing different ACs. Acknowledgements The authors thank the member companies of the Institute of Paper Science and Technology (IPST) for their generous support. Our thanks are also extended to Westvaco Company for determinations of AC BET and pore volumes. Portions of this work were used by S. Van Winkle in partial fulfillment of the requirements for the Ph.D. degree at IPST.
References Amerson HV, Frampton LJ Jr, McKeand SE, Mott RL, Weir RJ (1985) Loblolly pine tissue culture. In: Tissue culture in forestry and agriculture. Basic Series, vol 32. Plenum Press, New York, pp 271–287 Baker FS, Miller CE, Tolles DE (1992) Activated carbon. In: Kirk-othmer encyclopedia of chemical technology, vol 4., 4th edn. Wiley & Sons, New York Diamadopoulos E, Samaras P, Sakellaropoulos GP (1992) The effect of activated carbon properties on the adsorption of toxic substances. Water Sci Technol 25:153–160 Ebert A, Taylor F (1990) Assessment of the changes of 2,4dichlorophenoxyacetic acid concentrations in plant tissue culture media in the presence of activated charcoal. Plant Cell Tissue Organ Cult 20:165–172 Ebert A, Taylor F, Blake J (1993) Changes of 6-benzylaminopurine and 2,4-dichlorophenoxyacetic acid concentrations in plant tissue culture media in the presence of activated charcoal. Plant Cell Tissue Organ Cult 33:157–162 George EF, Sherington PD (1984) Plant propagation by tissue culture. Exergetics, Hants, UK Menzuali-Sodi A, Panizza M, Serra G, Tognoni F (1993) Involvement of AC in the modulation of abiotic and biotic ethylene levels in tissue culture. Sci Hortic 54:49–57 Pan JJ, van Staden J (1998) The use of charcoal in in vitro culture—a review. Plant Growth Regul 26:155–163
Pullman GS, Gupta PK (1991) Method for reproducing coniferous plants by somatic embryogenesis using adsorbent materials in the development stage. U.S. Patent No. 5034326, July 23, 1991 Pullman GS, Gupta PK (1994) Method for reproducing conifers by somatic embryogenesis using mixed growth hormones for embryo culture. U.S. Patent No. 5294549, March 15, 1994 Pullman GS, Johnson S (2002) Somatic embryogenesis in loblolly pine (Pinus taeda L): improving culture initiation rates. Ann For Sci 59:663–668 Pullman GS, Peter G (2002) Methods of initiating embryogenic cultures in plants. U.S. Patent No. 6,492,174B1. Issued December 10, 2002 Pullman GS, Gupta PK, Timmis R, Carpenter C, Kreitinger M, Welty E. (2005) Improved Norway spruce somatic embryo development through the use of abscisic acid combined with activated carbon. Plant Cell Rep (in press) Schwarzenbach RP, Gschwend PM, Imboden DM (1993) Environmental Organic Chemistry. John Wiley & Sons, New York Shaw DJ (1992) Colloid and surface chemistry. Butterworth Heinemann, Boston Toering A, Pullman GS (2005) Modeling available 2,4dichlorophenoxyacetic acid in a tissue culture medium containing activated carbon. Plant Cell Tissue Organ Cult (in press) Van Winkle S (2000) The effect of activated carbon on the organic and elemental composition of plant tissue culture medium. PhD thesis, Institute of Paper Science and Technology, Atlanta, Ga. Van Winkle S, Pullman GS (2003) The combined impact of pH and activated carbon on the elemental composition of plant tissue culture media. Plant Cell Rep 22:303–311 Van Winkle S, Johnson S, Pullman GS (2003) The impact of Gelrite and activated carbon on the elemental composition of two conifer embryogenic tissue initiation media. Plant Cell Rep 21:1175–1182 Verhagen SA, Wann SR (1989) Norway spruce somatic embryogenesis: high-frequency initiation from light-cultured mature embryos. Plant Cell Tissue Organ Cult 16:103–111 Zaghmout OMF, Torello WA (1988) Enhanced regeneration in long-term callus cultures of red fescue by pretreatment with activated carbon. HortScience 23:615–616