Frequency and Timing of Leaf Removal Affect Growth and Podophyllotoxin Content of Podophyllum peltatum in Full Sun

• Abstract
• Introduction
• Materials and Methods
• Results
• Discussion
• Acknowledgements
• References

Abstract

Podophyllotoxin is a pharmaceutical compound found in leaves and rhizomes of American mayapple (P. peltatum L.), a species being investigated as an alternative to that of the Indian mayapple (P. emodi). Leaves alone can serve as a renewable source of podophyllotoxin (and other lignans) leaving rhizomes undisturbed to produce leaf biomass in subsequent years. It is not known, however, how frequently or severely plants can be defoliated without adversely affecting future plant growth, lignan content, or podophyllotoxin yield (g·m^-2). This study compared harvest strategies that were mild to severe in frequency and timing of leaf removal. A wild population in full sun was subjected to leaf removal treatments of varying frequency (every year, every 2^nd or 3^rd year) and timing (early or late). Control plots not previously harvested were included every year. Plots were 1.0 m² and established during spring of 2001. Duration of the study was four years. P. peltatum plants did not tolerate the most severe harvest treatment: annual harvest frequency in combination with early harvest time. Early annual harvests reduced total leaf dry mass and total leaf area in a consistent and linear manner. In contrast, plants tolerated annual harvests when conducted late in the growing season and tolerated early harvests when conducted every 2^nd or 3^rd year. The number of sexual shoots was reduced to zero by early annual harvests. Podophyllotoxin content was 2.7 to 6.5 times greater in leaves harvested early compared to those harvested late, though content was significantly greater in only two out of four years. In conclusion, we can recommend leaf removal every year from well-established P. peltatum populations grown in full sun if harvests are conducted late in the growing season. This harvest strategy ensures maximum podophyllotoxin yield without jeopardizing future leaf biomass yield. Leaves harvested early appear to have greater podophyllotoxin content, but we discourage early harvest every year. Instead, our results indicate that leaves can be harvested early every other year without reducing long-term performance of P. peltatum populations.

Introduction

American mayapple, Podophyllum peltatum L. (Berberidaceae), is being investigated as a potential domestic source of the compound podophyllotoxin. This compound is used in the manufacture of drugs used in the treatment of cancer, arthritis, and various skin conditions such as psoriasis and genital warts [1], [2], [3]. The current source of podophyllotoxin is the Indian mayapple (Podophyllum emodi Wall.; syn. P. hexandrum Royle), but the plant was considered threatened in its native habitat of the Himalayas due to excessive and destructive harvest [4], [5].

Recent research investigating the possible domestication of P. peltatum demonstrated that the species was relatively easy to manipulate with normal horticultural techniques [6], [7], [8]. Field plantings of P. peltatum accessions with high podophyllotoxin content offer the possibility of this species supplying podophyllotoxin to the pharmaceutical industry and serving as an alternative source of income for growers of specialty crops [9], [10], [11]. Naturally-occurring populations in the wild are known to vary widely in podophyllotoxin content, from zero to 56 mg·g^-1, and commercial plantings are expected to use only accessions with high levels of the compound [12]. To date, shade is the only cultural practice reported to affect podophyllotoxin content, with plants under full sun exhibiting greater podophyllotoxin contents than those under increasing levels of shade [8], [11]. Podophyllotoxin is not the only lignan found in P. peltatum, α-peltatin, β-peltatin, and epipodophyllotoxin are also present in P. peltatum leaves and are of secondary interest as pharmaceutical compounds.

A key component of establishing P. peltatum in sustainable and perennial plantings is the harvest of leaves and not rhizomes. By extracting lignans from leaves and not rhizomes, there would be no need to harvest destructively and then replant. Instead, rhizomes would be left intact to produce shoots in future years and create a renewable source of leaf biomass from a single planting. However, it is not known how severely P. peltatum plants can be defoliated without adversely affecting subsequent growth and productivity of established plantings. Botanical investigations of defoliation tend to focus on herbivory and reproductive fitness, often ignoring horticultural data such as yield. A study of partial defoliation of Trillium grandiflorum (Michaux) Salisbury, a species with a similar spring ephemeral habit as P. peltatum, reported that seed number and average seed weight were not reduced by partial leaf removal [13]. In contrast, rhizome density and total non-structural carbohydrates tended to decrease as defoliation increased. Rockwood and Lobstein [14] reported decreased sexual performance in two out of four spring ephemeral species, Jeffersonia diphylla L. and Trillium sessile L., during the second year of partial defoliation. Long-term performance of the species used in these studies was not reported. Though the evidence is weak, it seems likely that species such as these plants - herbaceous ephemerals that are unable to produce new growth in response to defoliation - would lose vigor over time as frequency and severity of defoliation increased. The overall goal of the research reported here was to explore harvest procedures that ensured long-term sustainability of perennial P. peltatum plantings. Frequency and timing of leaf removal was investigated to test the effects of harvest strategy on plant growth, leaf biomass yield (g·m^-2), podophyllotoxin content (mg·g^-1), and podophyllotoxin yield (mg·m^-2) during the four years of this study.

Materials and Methods

This study used a naturally-occurring population of P. peltatum located in Holly Springs, Mississippi, USA (34.820°N 89.439°W, elev. 148 m). The population was well established, had established naturally, and was located in an open sunny site adjacent to a field used for annual production of agronomic crops. The population grew in a relatively protected area alongside a severely eroded ravine. The site was also characterized by a well-established population of kudzu [Pueraria montana (Lour.) Merr. var. lobata (Willd.) Maesen & S.M. Almeida]. In addition, red buckeye (Aesculus pavia L.) and catchweed bedstraw (Galium aparine L.) competed somewhat successfully at this location. P. peltatum shoots emerged in early spring during the last week of March or the first week of April and began to senesce when kudzu vines produced new growth, about the second or third week of May. Under these conditions the P. peltatum population used in this study flourished. Voucher specimens have been collected and are being processed at the Research Institute of Pharmaceutical Sciences Herbarium at the University of Mississippi. A wild population was used in this study because of the immediate availability of a large, relatively uniform and well-established colony of plants and the years it would take to establish a stable planted population to begin the investigation. During the first years of establishment, new plantings of P. peltatum were reported to increase annually in number of shoots, leaf area, leaf biomass, and number of sexual shoots [6]. The wild population used in this study was relatively stable and plant growth characteristics increased only slightly in control plots from 2001 to 2004 (Table [1], Fig. [1]).

Research plots were established during spring of 2001. Each plot was marked with four wooden stakes placed at the corners of a 1.0 m by 1.0 m square. Plots were selected to include plants of average or slightly above average growth and density compared to the surrounding population. Plots were located away from any unusual growths or abnormalities. Leaves of sexual shoots (normally two leaves per shoot) were harvested separately from those of asexual shoots (normally one leaf per shoot). Treatments were a factorial arrangement of frequency of harvest (every year, every 2^nd or 3^rd year, or control plots not previously harvested) and time of harvest (early or late). Early harvest was defined as the earliest time leaves were fully expanded. Late harvest was defined as the time some leaves began to senesce by exhibiting yellowing but before they began to turn brown. Controls were plots that had not been previously harvested at any other time during the study. The experimental design was a randomized complete block with three blocks. Data were not available each year for all treatments due to the nature of the study (harvest every year or every 2^nd or 3^rd year) and therefore analyses were performed by year.

Leaves were collected for early harvest on 19 April 2001, 11 April 2002, 14 April 2003, and 27 April 2004. Leaves were collected for late harvest on 10 May 2001, 7 May 2002, 7 May 2003, and 6 May 2004. Leaves were harvested within each 1.0 m² plot according to shoot type and counted. Leaf area was recorded using an area meter (LI-3100 Area Meter; LI-COR; Lincoln, NE, USA). Prior to each harvest leaves were removed from a 0.5 m wide band around the outside perimeter of each plot. These leaves were discarded, but this procedure assured that rhizomes growing into the treatment area throughout the duration of the study would produce shoots that had been treated the same as shoots arising from rhizomes within each plot (see Fig. [1]S in Supporting Information).

Leaves were dried in a forced-air, constant-temperature oven (1380FM; VWR Scientific Products; Cornelius, OR, USA) at 40 °C and dry mass recorded. Samples were then sent to the National Center for Natural Products Research at the University of Mississippi to be analyzed for lignan content. Podophyllotoxin, α-peltatin, and β-peltatin were extracted and quantified according to Canel et al. [15], [16] (see Supporting Information). Data were analyzed using the Mixed Procedure of SAS (SAS Institute Inc.; Cary, NC, USA) and means separation was by LSD at P ≤ 0.05. Regression analysis was used to evaluate and compare possible linear and quadratic trends over time for plots harvested early and late every year. Control plots each year were also subjected to regression analysis to evaluate trends for early and late harvests. The Mixed Procedure was used with a first-order autoregressive correlation adjustment for repeated measures on plots over time.

Results

Harvest frequency only slightly affected growth of P. peltatum plants during the four years of this study. Significant effects due to harvest frequency were detected only for total leaf dry mass during 2003 and ratio of sexual shoots during 2003 and 2004 (Table [1]). In each case, plants harvested every year produced significantly less total leaf dry mass and fewer sexual shoots than plants harvested every other year, every third year, or the controls. Harvest time affected P. peltatum growth to a greater extent than harvest frequency. Harvest time significantly affected total leaf area during 2001 and 2002, total leaf dry mass during 2002 and 2003, shoot number during 2001, and ratio of sexual shoots during 2001, 2002, and 2003 (Table [1]). Significant interactions between harvest frequency and harvest time occurred during 2003 and 2004 for total leaf area and during 2004 for total leaf dry mass (Table [1]). In each case, plants harvested early every year produced significantly less total leaf area and total leaf dry mass than any of the other treatment combinations. Shoot number was not affected by harvest frequency or time.

Lignan content and podophyllotoxin yield were generally not affected by harvest frequency or time (Table [2]). Exceptions occurred during 2001 and 2003 when plants harvested early produced significantly greater podophyllotoxin content than those harvested late. In addition, during 2001 plants harvested every other year produced lower podophyllotoxin yield than those harvested every year or every third year.

Discussion

Our results clearly demonstrate that P. peltatum plants did not tolerate the most severe of harvest treatments: annual harvest frequency in combination with early harvest time. Shoot growth was negatively impacted during the four years of this study when leaves were removed from plants at the beginning of each growing season. Compared to annual harvests made late in the season, and compared to control plants harvested early or late, annual harvests made early significantly reduced total leaf dry mass (Fig. [1]) and total leaf area (data not shown) in a consistent and linear manner. These results were expected. P. peltatum plants did not produce new leaves in response to defoliation and, therefore, did not have the opportunity to recover from early leaf removal during the year in which they were defoliated. Several years of leaf removal would be expected to weaken plants by depleting rhizomes and reduce long-term performance compared to unharvested plants allowed to benefit from repeated growing seasons with foliage intact. The negative impact of leaf removal clearly affected leaf size and weight but not number. Plant number did not significantly change during the study.

Our results also demonstrate that P. peltatum plants can be harvest annually if harvested late in the growing season. Annual harvest frequency in combination with late harvest time proved to be less severe than early harvest every year and was not different from control plots that had not been harvested previously. This indicates that P. peltatum leaves can be harvested every year as long as leaves are retained by plants for a sufficient length of time after full leaf expansion. P. peltatum plants also tolerated early harvests when made in combination with harvest frequencies greater than every year such as every 2^nd or 3^rd year. This indicates that P. peltatum plants can tolerate early harvest if allowed to recover for one or two years before again being harvested.

Lubbers and Lechowicz [13] reported that partial leaf removal at time of flowering of T. grandiflorum reduced rhizome density and total non-structural carbohydrates at the end of the growing season. Long-term performance of these plants was not reported, but it is possible that reduced levels of storage carbohydrates in rhizomes caused by repeated defoliation could result in long-term reductions of leaf area and leaf dry weight similar to our results reported here. It seems reasonable to assume that repeated and significant reductions in carbohydrate content and loss of rhizome mass would also reduce overall growth, and the effects would be obvious after only a few growing seasons.

Rockwood and Lobstein [14] reported that partial defoliation tended to reduce sexual performance of J. diphylla and T. sessile. These species have a similar growth habit as P. peltatum. Our results demonstrate an almost immediate effect of early and severe leaf removal on reproductive fitness. Though not statistically significant, the ratio of sexual shoots to total shoots was reduced to zero by early annual harvests (Table [1]).

During the first year of the study, plants harvested early appeared to produce about twice the leaf area of plants harvested late. We discovered that the second harvest of 2001 was scheduled too late and leaves had already begun to decline and form brown, dry margins. Leaf area was underestimated because some of the dry leaf tissue was not recorded by the area meter. As a result, early harvest appeared to produce significantly greater leaf area than late harvest. Under these conditions dry mass was a more accurate measure of treatment differences. In contrast to 2001, plants harvested early during 2002 produced significantly less leaf area than plants harvested late. We discovered at this time that, again, harvest timing was important. When leaves were harvested slightly too early they were not fully expanded and tended to underestimate leaf area and leaf dry mass.

Podophyllotoxin contents were 2.7 to 6.5 times greater every year of this study in leaves harvested early compared to those harvested late, though content was significantly greater in only two out of four years. To our knowledge this is the first report of differences in podophyllotoxin content due to physiological leaf age, and these results may encourage growers to harvest early. Podophyllotoxin yield, however, was not greatly affected by treatments and treatment combinations during this study. During 2001, podophyllotoxin yield was greater among plants harvested every year and every 3^rd year compared to those harvested every 2^nd year, although this difference could not have been due to treatments because it occurred during the first year of the study. Contents of α-peltatin and β-peltatin were not significantly affected at any time during this study.

In conclusion, we can recommend leaf removal every year from well-established P. peltatum populations grown in full sun if harvests are conducted late in the growing season, after plants are allowed to accumulate sufficient resources for the next season. This harvest strategy ensures maximum leaf biomass yield (g·m^-2), podophyllotoxin content (mg·g^-1), and podophyllotoxin yield (mg·m^-2) without jeopardizing long-term performance of the population. Leaves harvested early, however, appear to have greater podophyllotoxin content. We strongly discourage early harvest every year because this quickly reduced vigor and productivity. Instead, our results indicate that leaves can be harvested early every other year without reducing long-term performance.

Acknowledgements

This research was supported by USDA NRI Competitive Grant Program 71.2, Award #2002 - 01 525, and Specific Cooperative Agreement No. 58 - 6408 - 7-012.

References

• 1 Bedows E, Hatfield G M. An investigation of the antiviral activity of Podophyllum peltatum .  J Nat Prod. 1982;  45 725-9
  CrossrefPubMedSearch in Google Scholar
• 2 Jackson D E, Dewick P M. Tumor-inhibitory aryltetralin lignans from Podophyllum pleianthum .  Phytochemistry. 1985;  24 2407-9
  CrossrefPubMedSearch in Google Scholar
• 3 Lerndal T, Svensson B. A clinical study of CPH 82 vs. methotrexate in early rheumatoid arthritis.  Rheumatology (Oxford). 2000;  39 316
  CrossrefPubMedSearch in Google Scholar
• 4 Foster S. Medicinal plant conservation and genetic resources: Examples from the temperate Northern hemisphere.  Acta Hortic. 1993;  330 67-73
  PubMedSearch in Google Scholar
• 5 Rai L K, Prasad P, Sharma E. Conservation threats to some important medicinal plants of the Sikkim Himalaya.  Biol Conserv. 2000;  93 27-33
  PubMedSearch in Google Scholar
• 6 Cushman K E, Maqbool M. Propagule type and planting time affect subsequent mayapple growth.  Hort Science. 2005;  40 640-4
  PubMedSearch in Google Scholar
• 7 Cushman K E, Maqbool M, Gerard P D. Mulch type, mulch depth, and rhizome planting depth for field-grown American mayapple.  Hort Science. 2005;  40 635-9
  PubMedSearch in Google Scholar
• 8 Cushman K E, Maqbool M, Lata H, Bedir E, Khan I A, Moraes R M. Podophyllotoxin content and yield of American mayapple leaves in sun and shade.  Hort Science. 2005;  40 60-3
  PubMedSearch in Google Scholar
• 9 Meijer W. Podophyllum peltatum - Mayapple a potential new cash-crop plant of Eastern North America.  Econ Bot. 1974;  28 68-72
  PubMedSearch in Google Scholar
• 10 Moraes R M, Burandt Jr C, Ganzera M, Li X, Khan I, Canel C. The American Mayapple revisited - Podophyllum peltatum - still a potential cash crop?.  Econ Bot. 2000;  54 471-6
  PubMedSearch in Google Scholar
• 11 Moraes R M, Momm H G, Silva B, Maddox V, Easson G L, Lata H. et al . Geographic information system method for assessing chemo-diversity in medicinal plants.  Planta Med. 2005;  71 1157-64
  Thieme ConnectPubMedSearch in Google Scholar
• 12 Moraes R M, Bedir E, Barret H, Burandt C, Canel C, Khan I. Evaluation of Podophyllum peltatum L accessions for podophyllotoxin production.  Planta Med. 2002;  68 341-4
  Thieme ConnectPubMedSearch in Google Scholar
• 13 Lubbers A E, Lechowicz M J. Effects of leaf removal on reproduction vs. belowground storage in Trillium grandiflorum .  Ecology. 1989;  70 85-96
  PubMedSearch in Google Scholar
• 14 Rockwood L L, Lobstein M B. The effects of experimental defoliation on reproduction in four species of herbaceous perennials from northern Virginia.  Castanea. 1994;  59 41-50
  PubMedSearch in Google Scholar
• 15 Canel C, Dayan F E, Moraes R M, Burandt C. Enhanced yield of podophyllotoxin from natural products through in situ conversion methods. US Patent 6,143,304; 2000. 
• 16 Canel C, Dayan F E, Ganzera M, Khan I A, Rimando A, Burandt C L. et al . High yield of podophyllotoxin from leaves of Podophyllum peltatum by in situ conversion of podophyllotoxin 4-O-β-D-glucopyranoside.  Planta Med. 2001;  67 97-9
  Thieme ConnectPubMedSearch in Google Scholar

Dr. Kent Cushman

Southwest Florida Research & Education Center

University of Florida

Institute of Food and Agricultural Sciences

Immokalee

FL 34142

USA

Phone: +1-239-658-3429

Fax: +1-239-658-3469

Email: kcushman@ufl.edu

References

• 1 Bedows E, Hatfield G M. An investigation of the antiviral activity of Podophyllum peltatum .  J Nat Prod. 1982;  45 725-9
  CrossrefPubMedSearch in Google Scholar
• 2 Jackson D E, Dewick P M. Tumor-inhibitory aryltetralin lignans from Podophyllum pleianthum .  Phytochemistry. 1985;  24 2407-9
  CrossrefPubMedSearch in Google Scholar
• 3 Lerndal T, Svensson B. A clinical study of CPH 82 vs. methotrexate in early rheumatoid arthritis.  Rheumatology (Oxford). 2000;  39 316
  CrossrefPubMedSearch in Google Scholar
• 4 Foster S. Medicinal plant conservation and genetic resources: Examples from the temperate Northern hemisphere.  Acta Hortic. 1993;  330 67-73
  PubMedSearch in Google Scholar
• 5 Rai L K, Prasad P, Sharma E. Conservation threats to some important medicinal plants of the Sikkim Himalaya.  Biol Conserv. 2000;  93 27-33
  PubMedSearch in Google Scholar
• 6 Cushman K E, Maqbool M. Propagule type and planting time affect subsequent mayapple growth.  Hort Science. 2005;  40 640-4
  PubMedSearch in Google Scholar
• 7 Cushman K E, Maqbool M, Gerard P D. Mulch type, mulch depth, and rhizome planting depth for field-grown American mayapple.  Hort Science. 2005;  40 635-9
  PubMedSearch in Google Scholar
• 8 Cushman K E, Maqbool M, Lata H, Bedir E, Khan I A, Moraes R M. Podophyllotoxin content and yield of American mayapple leaves in sun and shade.  Hort Science. 2005;  40 60-3
  PubMedSearch in Google Scholar
• 9 Meijer W. Podophyllum peltatum - Mayapple a potential new cash-crop plant of Eastern North America.  Econ Bot. 1974;  28 68-72
  PubMedSearch in Google Scholar
• 10 Moraes R M, Burandt Jr C, Ganzera M, Li X, Khan I, Canel C. The American Mayapple revisited - Podophyllum peltatum - still a potential cash crop?.  Econ Bot. 2000;  54 471-6
  PubMedSearch in Google Scholar
• 11 Moraes R M, Momm H G, Silva B, Maddox V, Easson G L, Lata H. et al . Geographic information system method for assessing chemo-diversity in medicinal plants.  Planta Med. 2005;  71 1157-64
  Thieme ConnectPubMedSearch in Google Scholar
• 12 Moraes R M, Bedir E, Barret H, Burandt C, Canel C, Khan I. Evaluation of Podophyllum peltatum L accessions for podophyllotoxin production.  Planta Med. 2002;  68 341-4
  Thieme ConnectPubMedSearch in Google Scholar
• 13 Lubbers A E, Lechowicz M J. Effects of leaf removal on reproduction vs. belowground storage in Trillium grandiflorum .  Ecology. 1989;  70 85-96
  PubMedSearch in Google Scholar
• 14 Rockwood L L, Lobstein M B. The effects of experimental defoliation on reproduction in four species of herbaceous perennials from northern Virginia.  Castanea. 1994;  59 41-50
  PubMedSearch in Google Scholar
• 15 Canel C, Dayan F E, Moraes R M, Burandt C. Enhanced yield of podophyllotoxin from natural products through in situ conversion methods. US Patent 6,143,304; 2000. 
• 16 Canel C, Dayan F E, Ganzera M, Khan I A, Rimando A, Burandt C L. et al . High yield of podophyllotoxin from leaves of Podophyllum peltatum by in situ conversion of podophyllotoxin 4-O-β-D-glucopyranoside.  Planta Med. 2001;  67 97-9
  Thieme ConnectPubMedSearch in Google Scholar

Dr. Kent Cushman

Southwest Florida Research & Education Center

University of Florida

Institute of Food and Agricultural Sciences

Immokalee

FL 34142

USA

Phone: +1-239-658-3429

Fax: +1-239-658-3469

Email: kcushman@ufl.edu

• Supplementary Material