'Gransden' and 'Villersexel'

University of South Dakota USD RED

Honors Thesis Theses, Dissertations, and Student Projects

2017

The Old, the Young, and the Dehydrated: comparing dehydration tolerance of ‘Gransden’ and ‘Villersexel’ ecotypes of Physcomitrella patens

Katherine Bollinger University of South Dakota

Follow this and additional works at: https://red.library.usd.edu/honors-thesis

Recommended Citation Bollinger, Katherine, "The Old, the Young, and the Dehydrated: comparing dehydration tolerance of ‘Gransden’ and ‘Villersexel’ ecotypes of Physcomitrella patens" (2017). Honors Thesis. 1. https://red.library.usd.edu/honors-thesis/1

This Honors Thesis is brought to you for free and open access by the Theses, Dissertations, and Student Projects at USD RED. It has been accepted for inclusion in Honors Thesis by an authorized administrator of USD RED. For more information, please contact [email protected].

THE OLD, THE YOUNG, AND THE DEHYDRATED: COMPARING

DEHYDRATION TOLERANCE OF ‘GRANSDEN’ AND ‘VILLERSEXEL’ ECOTYPES OF

PHYSCOMITRELLA PATENS

by Katherine Bollinger

A Thesis Submitted in Partial Fulfillment Of the Requirements for the University Honors Program

______

Department of Biology The University of South Dakota December 2017

The members of the Honors Thesis Committee appointed to examine the thesis of Katherine Bollinger find it satisfactory and recommend that it be accepted.

______Karen L. Koster, Ph.D. Professor of Biology Director of the Committee

______Bernard W. M. Wone, Ph.D. Assistant Professor of Biology

______Kumudu N. Rathnayake Ph.D. Candidate in Biology

ABSTRACT

The Old, the Young, and the Dehydrated: comparing dehydration tolerance of ‘Gransden’ and ‘Villersexel’ ecotypes of Physcomitrella patens

Katherine Bollinger

Director: Dr. Karen L. Koster, Ph, D.

Plant tolerance of environmental stresses such as water deficit depends on genetic factors that can vary in response to the climate in which the plants have evolved.

Such genetically distinct populations of a species that live in different regions are referred to as ecotypes. The purpose of this research was to test the dehydration tolerance of two ecotypes of the moss species Physcomitrella patens, one from Little

Gransden, England, which has been used for decades for research, and another more recently cultured ecotype from Villersexel, France. Little Gransden receives nearly half the yearly precipitation of Villersexel, which could have led to the ‘Gransden’ ecotype becoming more dehydration tolerant than the ‘Villersexel’ ecotype. To test my hypothesis, I allowed samples containing contained both protonemata and gametophores of each ecotype to equilibrate to a range of relative humidities (RH), measured the extent of dehydration they experienced, and then assessed their ability to survive dehydration using chlorophyll fluorescence as a measure of metabolic recovery after rehydration. The data indicate that the ‘Gransden’ ecotype survived and recovered from equilibration to RH as low as 89% RH, while the

‘Villersexel’ ecotype was less tolerant of dehydration, only surviving equilibration down to 95% RH.

Keywords: Physcomitrella patens, dehydration tolerance, chlorophyll fluorescence, ecotypic variation, ‘Gransden’, ‘Villersexel’

TABLE OF CONTENTS

LIST OF FIGURES ...... vi LIST OF TABLES ...... vii ACKNOWLEDGEMENTS ...... viii INTRODUCTION ...... 1 MATERIALS AND METHODS ...... 13 RESULTS ...... 21 DISCUSSION AND CONCLUSION ...... 35 APPENDIX ...... 42 BIBLIOGRAPHY ...... 50

LIST OF FIGURES

Figure 1: The life cycle of P. patens ...... 4 Figure 2: Origin of ‘Gransden’ and ‘Villersexel’ moss ecotypes ...... 10

Figure 3: Rate of water loss of VX and GR samples over saturated KNO3 ...... 22

Figure 4: Rate of water loss of VX and GR samples over saturated MgSO4 ...... 22 Figure 5: Rate of water loss of VX and GR samples over saturated KCl ...... 23 Figure 6: Rate of water loss of VX and GR samples over saturated NaCl ...... 23

Figure 7: Rate of water loss of VX and GR samples over saturated MgCl2 ...... 24

Figure 8: Average final water content (gH2O/gDW) for VX and GR at five RH levels after 144 h equilibration period ...... 26

Figure 9: Adjusted average final water content (gH2O/gDW) for VX and GR at five RH levels after 144 h equilibration period ...... 26

Figure 10: The mean chlorophyll fluorescence (Fv/Fm) measurements during rehydration for the ‘Villersexel’ ecotype ...... 31

Figure 11: The mean chlorophyll fluorescence (Fv/Fm) measurements during rehydration for the ‘Gransden’ ecotype ...... 31 Figure 12: Survival of VX and GR ecotypes after equilibration to a range of RH ...... 33

Figure A1: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 1 of the ‘Villersexel’ ecotype ...... 44

Figure A2: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 1 of the ‘Gransden’ ecotype ...... 44

Figure A3: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 2 of the ‘Villersexel’ ecotype ...... 46

Figure A4: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 2 of the ‘Gransden’ ecotype ...... 46

Figure A5: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 3 of the ‘Villersexel’ ecotype ...... 48

Figure A6: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 3 of the ‘Gransden’ ecotype ...... 48

LIST OF TABLES

Table 1: Climatological data of Villersexel, France and Little Gransden, England ...... 8

Table 2: Control chlorophyll fluorescence (Fv/Fm) measurements of healthy ‘Gransden’ ecotype tissue prior to dehydration ...... 29

Table 3: Control chlorophyll fluorescence (Fv/Fm) measurements of healthy ‘Villersexel’ ecotype tissue prior to dehydration ...... 29

vii

ACKNOWLEDGEMENTS

This study was funded through a Council for Undergraduate Research and

Creative Scholarship (CURCS) Research Grant, provided by the University of South

Dakota. I would like to thank the council for awarding me with the opportunity and means to complete this research project.

I would also like to thank my husband, family, and friends for the continuous encouragement throughout the process of planning, preparing, researching, writing, and editing of my thesis. It is because of their constant support that I was able to persevere through the challenging times.

Next, I would like to thank Clay Lippert for researching alongside me in the

Koster laboratory. I enjoyed the time we spent together as colleagues and appreciated learning alongside you.

Lastly, I would like to thank Dr. Koster for her constant guidance, assistance, and direction. It is because of Dr. Koster’s passion that I initially became interested in research and her dedication to teaching that I learned about plant physiology and improved my scientific writing. She spent countless hours assisting me in the lab, explaining plant physiology, challenging my writing, and helping with revisions of my thesis. I cannot thank her enough.

viii

INTRODUCTION

Life as we know it would not be possible without plants. Plants provide many items we utilize in our daily life such as pharmaceuticals, food, textiles, bioenergy, and even the oxygen in the air we breathe. Given the importance of plants to life on Earth, it is alarming that plants in many parts of the world are facing increasing threats of drought stress, which could lead to an agricultural, economic, and global catastrophe (Reski and Frank, 2005; Wang et al., 2009). In response to this threat, scientists are working to develop plants that are more dehydration tolerant. Genomic research using plants that vary in their dehydration tolerance may help us further understand plant adaptations to harsher, drier conditions

(Wang et al., 2009). Early land plants, specifically the mosses, have evolved over time to survive environmental stresses, making them a valuable resource for plant stress tolerance research (Cuming et al., 2007; Prigge and Bezanilla, 2010). One moss species, Physcomitrella patens, has become a model species for studying stress responses (e.g. Frank et al., 2005; Cuming et al., 2007; Koster et al., 2010), and recent research reported the existence of ecotypes of this species with known genomic variability (Hiss et al., 2017). My study compares the dehydration tolerance of two of these ecotypes, the established ‘Gransden’ line and the more recently isolated ‘Villersexel’ line, in order to understand how differing ecotypes of this mesic bryophyte adapt to the environmental stresses of their climates.

1 Background Information on Physcomitrella patens

Moss research provides a useful avenue for studying dehydration tolerance due to the moss’s short life cycle and small size (Cove et al., 2006). Physcomitrella patens (Hedw.) Bruch & Schimp. is a moss species that has been developed for use as a model system for studying plant development, metabolism, and stress tolerance

(Kamisugi et al., 2008; Wang et al., 2009; Prigge and Bezanilla, 2010).

Physcomitrella patens is ideal for research because it requires low maintenance, uses little laboratory space, and is rapidly propagated. Researchers are able to genetically manipulate the moss and test hypotheses about gene function because of

P. patens’ completely sequenced genome and ability to undergo relatively frequent homologous recombination events (Reski and Frank, 2005; Prigge and Bezanilla,

2010; Perroud et al., 2011).

The alternation of generations life cycle, shown in Figure 1, showcases the dominant haploid phase of P. patens. The haploid phase produces gametes, which combine to produce the diploid sporophyte, while the diploid phase produces haploid spores using meiosis (Schaefer and Zrÿd, 2001, Cove, 2005; Reski and

Frank, 2005). The dominant haploid gametophyte stage of P. patens makes it an attractive model for genetic analysis because it allows for simpler Mendelian genetic crosses, targeting of genes, mutant isolations, and allele replacements (Cove, 2005;

Frank et al., 2005; Cove et al., 2006; Prigge and Bezanilla, 2010; Perroud et al.,

2011). An additional advantage of using P. patens for research is that it regenerates

2 clonally from almost all cell types, forming filamentous protonemata tissue, as seen in Fig. 1.3, after homogenization (Reski and Frank, 2005; Cove et al., 2006).

Figure 1. The life cycle of P. patens. (1) Sporangium, or the main body of the sporophyte, releases spores, (2) Spores start to divide to become primary protonema, (3) Protonema, or a filamentous network of cells, branches and divides to form a young bud (not shown), (4) Young bud divides to develop into a leafy shoot, or gametophore. The rhizoids, root-like structures, are indicated by the red circle, (5) The gametophore produces both the antheridium (male) and the archegonium (female), reproductive structures that produce the sperm and egg, respectively. The sperm swims to the egg and fertilizes it. The diploid

4 sporophyte then forms on the apex of the gametophore. Through meiosis in the sporangium, spores are produced and the life cycle begins again. [Figure adapted from one by Schaefer and Zrÿd, 2001. Images from Wikimedia; free for use in public domain] Defining the Dehydration Tolerance of P. patens

The purpose of this study is to compare the dehydration tolerance of two ecotypes of P. patens. Throughout history, plants have adapted to their environments, conserving genes that allow them to survive varying degrees of dehydration (Frank et al., 2005; Cuming et al., 2007). Physcomitrella patens is ideal for dehydration tolerance research because of its ability to survive an array of environmental stresses and its recently sequenced genome, which combine to enable the identification of stress adaptation genes (Frank et al., 2005; Reski and

Frank, 2005; Cuming et al., 2007). Physcomitrella patens is considered a dehydration tolerant plant species and, according to one study, can survive water potentials down to -13 MPa (Koster et al., 2010). In our study, we distinguish between dehydration tolerance and desiccation tolerance. Although there is no clear definition of dehydration tolerance, it can be categorized by the amount of water left in the plant and the extent of the damage in the membranes and macromolecules of the cell (Bryant and Wolfe, 1992; Koster and Bryant, 2006), whereas desiccation tolerance for vegetative tissues is often defined in more absolute terms as the ability to survive equilibration with air at ≤50% relative humidity (RH) or water potentials less than approximately -100 MPa or below

(Alpert, 2005; Proctor et al., 2007; Oliver et al., 2010).

The morphology of the filaments of the P. patens protonema seemingly allows it to be dehydration, but not desiccation, tolerant. Physcomitrella patens is

5 one cell layer thick (Pressel and Duckett, 2010), which enables rapid evaporation of water and equilibration to the environment (Koster et al., 2010). Although P. patens is not considered a constitutively desiccation tolerant moss, it can still survive desiccation under certain circumstances (Oldenhof et al., 2006; Cuming et al., 2007;

Koster et al., 2010). This is termed inducible desiccation tolerance (IDT), where desiccation tolerance can be turned on by environmental stresses or signals

(Proctor et al., 2007; Stark et al., 2016). Physcomitrella patens can survive desiccation after the addition of exogenous abscisic acid (ABA) (Oldenhof et al.,

2006; Koster et al., 2010) or after a slow drying process (Wang et al., 2009; Stark et al., 2016), which may induce the production of endogenous ABA (Rathnayake et al.,

2017, unpublished; Xiao et al., 2018). In many plant species with IDT, these processes trigger induction of survival mechanisms, most commonly by accumulating sugars, other solutes, and certain protective proteins (Hoekstra et al.,

2001; Oldenhof et al., 2006; Proctor et al., 2007). Dehydration and desiccation tolerant plants generally accumulate large amounts of di- and oligosaccharides

(Hoekstra et al., 2001), which aid in preventing the loss of cellular water through osmotic effects (Bryant et al., 2001; Walters and Koster, 2007).

Climatological Data of the ‘Gransden’ and ‘Villersexel’ Ecotypes of Moss

As P. patens loses water easily, it is crucial for the moss to grow in temperate zones, near riverbanks, ponds, and streams (Cove, 2005; Prigge and Bezanilla,

2010). The growing season of P. patens in the northern hemisphere typically spans from August to January, during which time the sporophyte capsules mature

6 (eFloras.org, 2017). Extended drought during this period could adversely affect reproduction and survival of the moss if the surrounding soil and air become too dry for this mesic species. Both ‘Gransden’ and ‘Villersexel’ ecotypes used in this study originated in temperate climates, but differences in annual precipitation received at each site exist, as described below and shown in Table 1 (Merkel, 2015).

The ‘Gransden’ ecotype was cultured decades ago from a spore collected near

Little Gransden, England (Engel, 1968). The climate of Little Gransden is warm and temperate. It has an average annual rainfall of 571 mm and temperature of 9.7 °C.

Little Gransden receives rainfall year-round, with even the driest month of February getting 34 mm of precipitation. August, the usual start of the growing season for P. patens, is also the wettest month of the year, peaking at 53 mm of rainfall.

Temperature is the highest in the month of July, reaching an average 16.6 °C, and decreases steadily until the month of January, when the average temperature is 2.9

°C (Table 1).

The ‘Villersexel’ ecotype originated from the town of Villersexel, France

(Kamisugi et al., 2008), which is a considerably wetter habitat than Little Gransden.

Although the climate is also warm and temperate, Villersexel, France, with 979 mm of precipitation per year, gets nearly double the annual precipitation of Little

Gransden, England. The driest month of April has an average of 69 mm of rain, while November reaches 95 mm precipitation on average. In addition, Villersexel,

France is warmer on average than Little Gransden, England in the months of April through September. The average annual temperature of Villersexel is 9.8 °C, with

7 the warmest month averaging 18.6 °C (July) and the coolest month averaging 0.9 °C

(January) (Table 1).

Table 1. Climatological data of Villersexel, France and Little Gransden, England. Average temperature (°C) and precipitation (mm) are shown for the months of January through December. Data retrieved from Climate-Data.org website (Merkel, 2015), which averages data collected between 1982-2012. No variances were given.

January February March April May June July August September October November December Average Temperature of 0.9 2.4 6 9.3 13.2 16.6 18.6 18.1 15.3 10.4 5.2 1.7 Villersexel (°C) Average Temperature of Little Gransden 2.9 3.6 6 8.7 11.9 15 16.6 16.3 14.2 10.6 6.3 4.1 (°C) Average Precipitation of 88 75 72 69 81 92 76 94 82 69 95 86 Villersexel (mm) Average Precipitation of Little Gransden 47 34 44 44 47 50 48 53 51 50 51 52 (mm)

Overall, Villersexel, France has a warmer climate and it receives more precipitation on average than Little Gransden, England. Although both Little

Gransden and Villersexel reach their highest temperature in the month of July,

Villersexel is 2 °C warmer on average. In contrast, in the month of January, Little

Gransden is 2 °C warmer on average than Villersexel. Because Little Gransden has a more consistent temperature throughout the growing season of P. patens, the greater variability in the temperature of Villersexel could contribute to a difference in its dehydration tolerance. In addition, the drastic differences in precipitation of the two climates could also be a factor in the dehydration tolerance of the moss.

Villersexel receives more precipitation throughout the P. patens growing season than Little Gransden, with an average of 514 mm of precipitation over Little

Gransden’s 304 mm precipitation in the months August through January.

8 The varying geographic locations of Little Gransden and Villersexel, as shown in Figure 2, can contribute to their differences in climate. Both Villersexel and Little

Gransden are categorized as temperate oceanic climates (Chen and Chen, 2013), but

Little Gransden is closer to the ocean, which could be a factor in its moderate temperatures. The lower amount of precipitation and the lower temperatures overall of Little Gransden may contribute to a more dehydration tolerant P. patens ecotype.

Figure 2. Origin of ‘Gransden’ and ‘Villersexel’ moss ecotypes. Little Gransden, England is located at the coordinates (52.18053, -0.14023) while Villersexel, France is at (47.55102, 6.43221). (Google Maps, 2017)

The differences in precipitation and temperature of each climate during the growing season of P. patens may have led to local adaptations at the molecular level, allowing one ecotype to better survive drought and desiccating conditions.

Genomes for both ecotypes of moss have been sequenced (von Stackelberg et al.,

2006; Kamisugi et al., 2008; Rensing et al., 2008), which makes it possible to

10 compare their sequences and look for differences that might lead to differences in stress tolerance. According to a comparative genomics website

(www.genomevolution.org/coge), 2,670,127 single nucleotide polymorphisms

(SNPs) were found to differ between ‘Gransden’ and ‘Villersexel’, demonstrating that the two ecotypes of moss are genetically distinct (Meyberg, 2016; Hiss et al., 2017).

SNPs represent variations in the genetic code between different organisms or species (Brookes, 1999). Each SNP indicates the location that a single nucleotide is altered in the DNA sequence (Hiss et al., 2017). Knowing that the P. patens ecotypes are genetically distinct means that there may be a genetic basis for any differences in dehydration tolerance. Moreover, it will enable the future identification of additional candidate genes that may contribute to dehydration tolerance based on

SNPs that lead to different gene expression or different isoforms of protein being made.

Rationale and Hypothesis

Knowing that the two ecotypes of moss are genetically distinct, and that they originate from varying climates, we can hypothesize they may have evolved genetically over time and that their responses to dehydration will differ, allowing one ecotype to better tolerate dehydration. The importance of comparing ecotypes is that if one ecotype is more equipped to survive dehydration, further genomic research can be conducted, which would give additional insight into what genes are regulated to help the plant survive dehydration.

11 The purpose of this research was to test the dehydration tolerance of two ecotypes of the moss species P. patens. Little Gransden, England receives nearly half the yearly precipitation of Villersexel, France, which might have led the ‘Gransden’ ecotype to be evolve more dehydration tolerance than the ‘Villersexel’ ecotype. To test my hypothesis, I allowed samples of each ecotype to equilibrate to a range of water contents, then assessed their ability to survive dehydration using chlorophyll fluorescence as a measure of metabolic recovery after the rehydration. The data indicate that the ‘Gransden’ ecotype survived and recovered from equilibration to

89% RH, while the ‘Villersexel’ ecotype was less tolerant of dehydration, only surviving equilibration with 95% RH, but not to 89% RH.

MATERIALS AND METHODS

Cultivation Methods

Both the ‘Gransden’ 2013 and ‘Villersexel’ ecotypes of P. patens (Hedw.)

Bruch & Schimp. were provided by Dr. Stuart McDaniel from the University of

Florida. The two ecotypes were cultured using the technique of Cove et al. (2009), as follows. Using sterile technique in a biosafety hood, moss samples of approximately 3-4 weeks post-homogenization, comprising primarily protonemata, were homogenized in water and pipetted onto cellophane discs, obtained from AA

Packaging Ltd., England, courtesy of Dr. Mel Oliver (USDA-ARS), layered on agar culture medium in 9 cm diameter Petri dishes. All instruments and supplies had been sterilized in an autoclave prior to use.

The initial trials of ‘Gransden’ and ‘Villersexel’ were grown on BCDAT medium, containing ammonium tartrate, while the latter trials (Trials 2 and 3) were grown on BCD medium, which only has nitrate as a source of nitrogen (Cove et al.,

2009). The species of nitrogen available influences development, affecting the abundance of the two types of protonema: caulonema and chloronema, which differ in their morphology and growth rate (Cove et al., 2006). Caulonema cells are rapidly growing and less branched than chloronema, which are slow-growing and more branched (Cove et al., 2006). The addition of ammonium tartrate promotes chloronema branching and decreases the formation of caulonema (Cove et al.,

13 2006). However, caulonema filaments, and not chloronema, give rise to the buds that form gametophores (Cove et al., 2006). Therefore, gametophore formation is delayed in cultures grown in the presence of ammonium tartrate.

After propagation, both ecotypes of moss were placed in a Percival 136-LLX

(Perry, IA) growth cabinet at 24 C with a 16 h light: 8 h dark cycle at a light intensity of 50-75 µmol·m-2·s-1. Moss was grown for 4-6 weeks, producing samples that contained both protonemata and gametophores prior to use in dehydration tolerance trials.

Sample Dehydration

For experimental trials, the 9 cm cellophane discs covered with moss tissue were cut into 2x2 cm squares, yielding 6 samples per plate. Three samples were removed from the cellophane and placed on pre-weighed aluminum foil squares, while the other three samples were kept on cellophane. All 6 samples were then placed in small plastic Petri dishes measuring 4 cm diameter and 5 mm height. The samples placed on aluminum foil were used for measurements of water content since foil, unlike cellophane, does not absorb water and therefore does not confound moss water content determinations. Previous tests in the Koster lab showed that moss dehydration rates on foil did not differ significantly from rates on cellophane

(Niemann and Koster, unpublished data). Samples kept on cellophane were used to determine the survival of each ecotype after drying to equilibrium with different RH levels.

14 Controlled RH chambers were made to equilibrate samples of P. patens to different levels of RH. Each chamber was made using 500 mL screw-top Nalgene jars. Saturated salt solutions were made for each chamber to achieve distinct RH levels. The RH levels, water potential values at 25 C, and the corresponding salts used to create the salt solutions were: 93% (-10 MPa) KNO3; 89% (- 16 MPa) MgSO4;

86% (- 21 MPa) KCl; 75% (- 40 MPa) NaCl; 33% (- 153 MPa) MgCl2 (Rockland, 1960;

Winston and Bates, 1960). Samples were equilibrated at 24 C, which should have negligible effect on the predicted RH values, based on the data tables found in the literature, which show no difference greater than 1% between RH values at 20 and

24 C (Rockland, 1960; Winston and Bates, 1960). Five RH chambers were made for each ecotype so that each chamber only contained samples of one ecotype. All RH chambers were prepared 24 h before the beginning of a trial to allow the chambers to come to equilibrium at their designated RH. A Petri dish of saturated salt solution was prepared by gradually adding water to a layer of salt until a slurry was created, approximately 1 cm deep with no more than 1 mm of liquid water on top. This slurry buffers the RH within a narrow range, as long as excess water absorbed from samples is removed and water is added to the high RH chambers that lost water due to evaporation each time the lids were taken off. A Petri dish containing the appropriate salt slurry was then placed at the bottom of each chamber and the lid closed. After 24 h, the six 2x2 cm moss samples in their small Petri dishes were placed inside a single RH chamber on top of a plastic mesh, which allowed separation of the dishes from the salt solutions.

15 To ensure accuracy of the RH chambers, SHT31 RH probes (Sensirion, Staefa

ZH, Switzerland), which measure at ±2% RH accuracy, were taped inside the lid of each chamber. Five humidity probes were used and were transferred every 24 h between each set of chambers to monitor humidity. Water was taken out or added if chamber RH levels were inconsistent with the values established in the literature for each salt. Data were transmitted via Bluetooth and recorded each time and day sample weights were recorded.

The initial weights of the samples on aluminum foil were recorded before samples were placed in the RH chambers. Sample weights were measured using a

Mettler MT-10 microbalance with readability to 1 μg. After moss samples were placed in their respective RH chambers, the chambers were placed in the incubator with a 16 h light: 8 h dark cycle, at 24 C, to start the dry down period. Weights of the 3 samples on foil were measured quickly at 24 h intervals to monitor water loss at each RH. The samples were held in the RH chambers for 6 d (144 h), by which time constant weights were attained, indicating that the moss had equilibrated to the RH of the chamber. Only samples at 93% RH (KNO3) for Trial 3 of both

‘Villersexel’ and ‘Gransden’ stayed in the salt chambers another 24 h to ensure full equilibration. The samples at 168 h showed that even after an additional 24 h dry down period, the samples were already fully equilibrated, losing no additional water. After weights were recorded at 144 or 168 h, the three samples on aluminum foil were placed in the oven at 70 C for 48 h to remove remaining water from the moss tissue. Oven-dried samples were transferred to the microbalance in containers with approximately 1 cm silica gel to prevent rehydration prior to

16 recording the dry weight. Water contents on a dry matter basis were calculated using the formula

(퐹푊 − 퐷푊) 푊퐶 = 퐷푊 where WC is the water content in g H2O/ g dry matter (dm), FW is the fresh weight of moss after equilibration to RH, and DW is the weight of the moss after oven drying.

Rehydration

After equilibration in the RH chambers for 144 h, moss samples for viability measurements were placed, still on their cellophane, in a 9 cm Petri dish on two sheets of Whatman #1 filter paper that had been soaked with 2 mL sterile water.

Three samples were placed together in each Petri dish. The dishes were then sealed, wrapped in aluminum foil, and placed back into the incubator for 1 h to rehydrate. Rehydration was done in darkness to limit any potential photobleaching.

Chlorophyll fluorescence

Survival of moss samples was primarily determined through measurements of the chlorophyll fluorescence ratio Fv/Fm. Chlorophyll fluorescence (Fv/Fm) measurements indicate the maximal possible activity of Photosystem II reaction centers in the electron transport chain, thereby providing a gauge of metabolic recovery and survival of the moss tissue (Maxwell and Johnson, 2000; Greenwood and Stark, 2014). In short, some of the light energy that is absorbed in the chlorophyll molecules is re-emitted as light, which can then be measured by a

17 chlorophyll fluorometer. The amount of fluorescence emitted depends on how effectively chlorophyll in the photosystems uses absorbed light to stimulate the electron transfer reactions that start the photosynthetic light reactions (Maxwell and Johnson, 2000). The ratio Fv/Fm compares the variable fluorescence (Fv) to the maximal fluorescence (Fm) and is calculated according to the formula:

퐹 (퐹 −퐹 ) 푣 = 푚 표 퐹푚 퐹푚

where Fo represents the minimum fluorescence yield and Fm is the maximum fluorescence yield. The difference between maximum fluorescence and minimum fluorescence equals Fv (Maxwell and Johnson, 2000). Using a ‘light doubling’ technique, a high intensity and quick flash of light closes the PSII reaction centers.

During the short flash of light, the fluorescence reaches its maximum fluorescence

(Fm) and emits the fluorescence from energy that is not used in the electron transport chain during illumination. Comparing the maximum fluorescence to the fluorescence in the absence of light (Fo) indicates the efficiency of photochemical quenching, and therefore, the activity of PSII. When the photosystem is damaged, it emits more fluorescence, so Fm and Fo would be high. As the photosystem repairs itself, the Fo decreases because more energy can be diverted into the electron transport, leading to a higher Fv/Fm measurement. Higher Fv/Fm measurements denote healthier moss tissue and therefore better survival than samples that may have lower measurements.

Before samples of moss were dried down, the control chlorophyll fluorescence ratio Fv/Fm was measured at 0.7 s exposure using an Opti-Sciences

18 Modulated Fluorometer model OS-100 (Opti-Sciences Inc., Hudson, NH) with an 8 mm diameter probe. Measurements were taken on 3 separate, randomly selected areas of moss. If samples contained patches lacking growth, fluorescence measurements were avoided in those areas.

After 1 h of rehydration in 1 mL of sterile water on filter paper, moss samples were again measured for chlorophyll fluorescence, using the parameters described earlier, to gauge survival of the moss tissue. Chlorophyll Fv/Fm ratios were recorded at three randomly selected locations on each 2x2 cm sample of moss, avoiding areas of no growth. After Fv/Fm measurements were taken, moss samples were removed from the filter paper and placed on sterile BCD or BCDAT medium in 9 cm Petri dishes. The dishes were covered in aluminum foil for 24 h at 24 C to prevent photobleaching.

After 24 h, the aluminum foil was removed, and the chlorophyll fluorescence measurements were again recorded for all samples of both ecotypes. One mL of sterile water was added to the BCD plates and each plate was sealed with surgical tape (3M Micropore 1.25 cm width) in order to keep the samples sterile while still allowing gas exchange to occur. Chlorophyll fluorescence measurements were repeated every 24 h for three days. The aluminum foil was removed after 24 h and the plates were placed in the dark for 10 min for dark adaptation before Fv/Fm measurements were taken.

19 Visual assessments were also used to gauge moss survival. Living moss cells retain chlorophyll, so they appear green, allowing a quick assessment of the moss viability after recovery from dehydration (Koster et al., 2010).

Statistical analyses were performed using Microsoft Excel software. T-tests were used to compare water content and survival data between ecotypes at each RH value, while ANOVAs were used to assess ecotypic differences between the drying curves. Details of each statistical analysis are given along with the data.

RESULTS

To assess the ‘Villersexel’ (VX) and ‘Gransden’ (GR) ecotypes’ ability to tolerate dehydration, samples were dried to varying degrees. Weights were recorded during the dehydration period to calculate water content, and chlorophyll fluorescence was measured throughout the rehydration period to determine survival of the tissue. Using the water content and fluorescence data can give a valuable comparison of the survival of the ‘Villersexel’ and ‘Gransden’ lines at five

RH levels.

Equilibration period

Figures 3-7 display the water content measurements over the 144 h dehydration period for both ecotypes of moss at five RH levels. The water content measurements denote the amount of water lost from the samples as they equilibrated to the surrounding atmospheric humidity. In addition, the following drying curves compare how quickly each ecotype equilibrated to the varying RH levels.

KNO3 25

)

/ 20

O 2 VX 93% RH

(

n GR 93% RH

o 10

t a 5

0 0 50 100 150 Equilibration Time (h)

Figure 3: Rate of water loss of VX and GR samples over saturated KNO3. Each line shows the mean water contents (gH2O/gDW) for moss samples measured during equilibration to the chamber RH. Three biological replicate samples were measured in each experiment, and three experimental trials were conducted. Error bars represent standard deviation (n=3). Saturated KNO3 should generate an RH of 93% at 25C.

MgSO4 25

)

/ 20

O 2 VX 89% RH

(

n GR 89% RH

o 10

t a 5

0 0 50 100 150 Equilibration Time (h)

Figure 4: Rate of water loss of VX and GR samples over saturated MgSO4. Each line shows the mean water contents (gH2O/gDW) for moss samples measured during equilibration to the chamber RH. Three biological replicate samples were measured in each experiment, and three experimental trials were conducted. Error bars represent standard deviation (n=3). Saturated MgSO4 should generate an RH of 89% at 25C. 22

KCl 25

)

g 20

O 2 VX 86% RH

(

e GR 86% RH

o 10

t a 5

0 0 50 100 150 Equilibration Time (h)

Figure 5: Rate of water loss of VX and GR samples over saturated KCl. Each line shows the mean water contents (gH2O/gDW) for moss samples measured during equilibration to the chamber RH. Three biological replicate samples were measured in each experiment, and three experimental trials were conducted. Error bars represent standard deviation (n=3). Saturated KCl should generate an RH of 86% at 25C.

NaCl 25

)

g 20

O 2 VX 75% RH

(

t GR 75% RH

o 10

t a 5

0 0 50 100 150 Equilibration Time (h)

Figure 6: Rate of water loss of VX and GR samples over saturated NaCl. Each line shows the mean water contents (gH2O/gDW) for moss samples measured during equilibration to the chamber RH. Three biological replicate samples were measured in each experiment, and three experimental trials were conducted. Error bars represent standard deviation (n=3). Saturated NaCl should generate an RH of 75% at 25C. 24

MgCl2 25

)

/ 20

H VX 33% RH

(

e GR 33% RH

o 10

a 5

0 0 50 100 150 Equilibration Time (h)

Figure 7: Rate of water loss of VX and GR samples over saturated MgCl2. Each line shows the mean water contents (gH2O/gDW) for moss samples measured during equilibration to the chamber RH. Three biological replicate samples were measured in each experiment, and three experimental trials were conducted. Error bars represent standard deviation (n=3). Saturated MgCl2 should generate an RH of 33% at 25C.

The drying curves, shown in Figures 3-7, reveal that GR samples equilibrated faster than VX samples when placed in the KNO3 salt chamber, but dehydrated at a similar rate to VX samples in the MgSO4, KCl, NaCl, and MgCl2 salt chambers. This is supported by statistical analysis using two-factor (ecotype x drying time) ANOVA tests with replication on data from each salt chamber (i.e. from each figure), which show that the only chamber that showed a significant difference in the drying curves between the two ecotypes was KNO3, with a P-value of 0.0009. For KNO3, GR samples had equilibrated by 72 h, but VX samples did not equilibrate until 120 h, as indicated in Figure 5. The GR and VX samples that were placed in MgSO4, KCl, NaCl, and MgCl2 salt chambers showed similar water loss rates and all had P-values

26 ranging from 0.1 to 0.9, showing no significant differences between the drying curves of each ecotype. At 89% RH, both ecotypes equilibrated by 72 h. At 86% RH,

GR samples equilibrated by 48 h and VX samples equilibrated at 72 h, but the

AVOVA test showed that the two sets of data were not statistically different. Both ecotypes of moss equilibrated by 48 h in the NaCl RH chambers and within 24 h in the MgCl2 RH chambers, as shown in Figures 6 and 7.

The time it took for the moss to fully equilibrate positively correlates with the RH level. For example, at higher RH levels, it took both ecotypes much longer to lose all of their cellular water, whereas at lower RH levels, the all water content was lost by 24 h. The rate at which water was lost also can be connected to the moss’s ability to withstand intracellular water loss during dehydration conditions and the moss’s morphological structure, which will be further described in the discussion.

Final Water Contents

The final water contents were measured after the 144 h equilibration period in order to assess the amount of water left in the tissue on a dry matter basis.

Figures 8 and 9 compare the final water content for both ecotypes of P. patens.

Figure 8 displays the mean of all three trials, while Figure 9 shows the adjusted average by using rearranged data for chambers in which the RH was measured higher or lower than the expected RH (described more fully below).

1 Average

) 0.9 Gransden

D 0.8

O 0.7 Villersexel

( 0.6

e 0.5

o 0.4

e 0.3

0.2

i 0.1

F 0 93% 89% 86% 75% 33% Relative Humidity

Figure 8: Average final water content (gH2O/gDW) for VX and GR samples at five RH levels after 144 h equilibration period. Three experimental trials, each with three biological replicates, were repeated for each ecotype at five RH levels. Each bar shows the average of all three experimental trials at one RH level. Error bars represent standard deviations (n=3).

1.0 Adjusted Average

) 0.9

W Gransden

D 0.8

O 0.7 Villersexel

g 0.6

(

n 0.5

n 0.4

r 0.3

a 0.2

l 0.1

F 0.0 93% 89% 86% 75% 33% Relative Humidity

Figure 9: Adjusted average final water content (gH2O/gDW) for VX and GR samples at five RH levels after 144 h equilibration period. Three experimental trials, each with three biological replicates, were repeated for each ecotype at five RH levels. Samples where measured RH was 3% higher or lower than expected RH were combined with data of the closest RH chamber. Each bar shows the average of all the experimental trials at one RH level. Error bars represent standard deviations (n= 2-4). 28

29 Figure 8 shows the average final water content of the three trials at each humidity level. Statistical analysis using unpaired, two-tailed Student’s T-tests with

α = 0.05, reveal that there are no significant differences between equilibrium water contents for the two ecotypes at each RH.

The adjusted average graph, shown in Figure 9, uses recategorized data in order to more accurately represent the final water content measurements per RH.

In some cases, RH monitors indicated that the expected RH was not maintained by the salt slurry in the chamber. In these cases, the relative humidities were greater than 3% above or below expected values, so sample water contents were averaged with data from other chambers with similar measured RH values, instead of being grouped with other data from the same salt solution. This changed four data points.

VX Trial 2 over MgSO4 (expected 89% RH) measured 93% RH, so the data from this trial were combined with VX 93% RH data (KNO3). GR Trial 3 over KNO3 (expected

93% RH) and MgSO4 (expected 89% RH) both measured at 85% RH, so data were combined with the GR 86% RH data (KCl). In addition, the GR Trial 3 over KCl

(expected 86% RH) measured at 88% RH, so the data were combined with the 89%

RH data. The result of adjusting these samples is that there is less variability in final water contents within the trials and between ecotypes. The P-value from T-tests changed from 0.28 to 0.81 at 93% RH and from 0.12 to 0.47 at 86% RH.

Rearranging the data points narrowed differences in water contents between the ecotypes because inaccuracies of the RH were factored out.

30 Survival Determined by Chlorophyll Fluorescence Measurements

Control chlorophyll fluorescence measurements were taken before the dry- down period and were used as a comparison for moss throughout the rehydration period. After the 144 h equilibration period, initial chlorophyll fluorescence ratios

(Fv/Fm) were recorded on rehydrated samples of moss and measurements were repeated every 24 h throughout the rehydration period.

In addition, visual assessments were used to determine the survival of the moss tissue based on the amount of green tissue.

Control Chlorophyll Fluorescence Measurements

Control fluorescence values indicate that the moss was alive and not under measurable stress at the start of the dehydration period (Tables 2-3). Plant tissue in good health typically has measurements with a theoretical maximum of 0.832 ±

0.004 Fv/Fm (Henrique, 2009), and values ranging between approximately 0.6 and

0.8 have been reported for unstressed mosses (e.g. Proctor and Smirnoff, 2000;

Mayaba et al. 2001; Proctor et al., 2007). Testing the viability of the moss before the equilibration period is essential so that we know the Fv/Fm measurements taken during the rehydration period are not skewed by the initial health of the tissue.

31 Experimental Experimental Experimental Trial

Trial 1 Trial 2 3 Plate 1 0.693 ± 0.016 0.693 ± 0.029 0.625 ± 0.023 Plate 2 0.692 ± 0.007 0.666 ± 0.019 Plate 3 0.656 ± 0.038 0.679 ± 0.024 Plate 4 0.671 ± 0.019 0.703 ± 0.017 Plate 5 Mean ± SD 0.693 0.678 ± 0.018 0.669 ± 0.033

Table 2: Control chlorophyll fluorescence (Fv/Fm) measurements of healthy ‘Gransden’ ecotype tissue prior to dehydration. Each value is the mean of three separate readings on one plate of moss tissue, before 6 samples were cut for drying experiments. The means of each experimental trial were calculated along with standard deviations (n=3).

Experimental Experimental Experimental Trial 1 Trial 2 Trial 3 Plate 1 0.681 ± 0.015 0.639 ± 0.025 0.686 ± 0.025 Plate 2 0.604 ± 0.044 0.674 ± 0.021 Plate 3 0.605 ± 0.005 0.687 ± 0.014 Plate 4 0.620 ± 0.027 0.686 ± 0.008

Plate 5 0.659 ± 0.022 Mean ± SD 0.681 0.617 ± 0.016 0.678 ± 0.012

Table 3: Control chlorophyll fluorescence (Fv/Fm) measurements of healthy ‘Villersexel’ ecotype tissue prior to dehydration. Each value is the mean of three separate readings on one plate of moss tissue, before 6 samples were cut for drying experiments. The means of each experimental trial were calculated along with standard deviations (n=3).

Tables 2 and 3 reveal that both GR and VX samples had mean Fv/Fm ratios above 0.600 and so could be considered alive and healthy before dehydration. It is also seen that the Fv/Fm ratios of VX samples in Trial 2 were lower than ratios from other trials of VX, and this was confirmed using an unpaired, two-tailed Student’s T- test (P = 0.00034).

Chlorophyll Fluorescence Measurements Throughout Rehydration

After the 144 h dehydration period, the moss was placed back on BCD agar plates to start the rehydration period. Fv/Fm was measured after 1 h, and measurements were repeated every 24 h for 3 days. Monitoring the chlorophyll fluorescence data over time during rehydration helped determine the point at which the tissue was no longer viable, the ability of moss tissue to survive at each RH level, and which ecotype had more viable tissue.

33 Adjusted Average VX 0.8

0.7 VX 93% RH

m 0.6

F VX 89%

e 0.5 RH

c s 0.4

e VX 86%

o RH

F 0.3

y VX 75%

h RH

p 0.2

o l VX 33% h 0.1 C RH

0 0 24 48 72 Time Rehydrated (h)

Figure 10: The mean chlorophyll fluorescence (Fv/Fm) measurements during rehydration of the ‘Villersexel’ ecotype. Samples where measured RH was 3% higher or lower than expected RH were combined with data of the closest RH chamber. The adjusted mean changed four data points: VX Trial 2 at 89%, GR Trial 3 at 93%, 89%, and 86% RH. Error bars represent standard deviation (n= 2-4).

Adjusted Average GR 0.8

0.7 GR 93% RH

m 0.6

F GR 89%

e 0.5 RH

c s 0.4

e GR 86%

o RH

F 0.3

y GR 75%

h RH

p 0.2

o l GR 33% h 0.1 C RH

0 0 24 48 72 Time Rehydrated (h)

Figure 11: The mean chlorophyll fluorescence (Fv/Fm) measurements during rehydration of the ‘Gransden’ ecotype. Samples where measured RH was 3% higher or lower than expected RH were combined with data of the closest RH chamber. The adjusted mean changed four data points: VX Trial 2 at 89%, GR Trial 3 at 93%, 89%, and 86% RH. Error bars represent standard deviation (n= 2-4). 34 Figures 10 and 11 reveal the adjusted mean Fv/Fm measurements for Trials

1-3. The adjusted means includes four changed data sets, as described above for water content. Data from each trial can be seen in the Appendix.

Figures 10-11 reveal that the chlorophyll fluorescence measurements are similar between ecotypes, with the exception of the moss exposed to 89% RH. All data sets from moss at the highest RH have large variability, indicating that there was extensive trial-to-trial variability in the Fv/Fm ratios of the moss during recovery, as will be discussed further below. Nonetheless, some trends can be detected. Unlike VX samples, the GR samples with a measured 89% RH recovered some of their initial photosynthetic capacity, averaging Fv/Fm ratios of 0.410 at 48 h and 0.325 at 72 h, while VX samples never had Fv/Fm ratios greater than 0.2 during rehydration after equilibration to 89% RH (Fig. 10). For all three trials at a measured 89% RH, VX samples did not show increasing Fv/Fm measurements throughout the rehydration period and hovered around 0.100 (Fig. 10). The difference in the chlorophyll fluorescence measurements at 89% RH most distinguishes the two ecotypes and could indicate ‘Gransden’’s ability to tolerate lower humidity levels. Based on Fv/Fm ratios, neither GR nor VX ecotypes survived

RH levels lower than 89% (Figs. 10-11). Consistent with this, no GR samples in Trial

3 recovered (Fig. A6); however, the RH values measured for the chambers in this trial did not exceed 88%. The samples for both ecotypes at 86%, 75%, and 33% RH consistently maintained Fv/Fm ratios of 0.100 and below during rehydration (Figs.

10-11), suggesting that they were too dehydrated to revive.

35 Percentage of Initial Chlorophyll Fv/Fm Ratio

To showcase the difference between the health of the dehydrated and rehydrated moss to the health of the moss initially, recovery as a percentage of the initial chlorophyll fluorescence ratio was calculated. This calculation also minimizes effects of variability in the absolute values of the initial Fv/Fm ratios among the trials, as was noted for the VX samples in Trial 2 (Table 3). We can consider this measure of recovery as a quantitative indicator of tissue survival.

Adjusted Average Percent of Initial Fluorescence

120%

)

a r 100%

f 80%

o VX Average

e GR Average v 60%

(

40%

u 20%

0% 33% RH 75% RH 86% RH 89% RH 93% RH

Figure 12: Survival of VX and GR ecotypes after equilibration to a range of RH. Survival was calculated as the percentage of the initial, pre- dehydration Fv/Fm ratio that was recovered after 3 d rehydration. The adjustment changed two data points as described in the text. The number of samples differs for some means due to the rearrangement of data. Error bars represent standard deviation (n= 2-4).

Figure 12 shows GR samples recovered a high percentage of their initial fluorescence after dehydrating at 93% RH and 89% RH. VX samples recovered more than 60% of their pre-drying chlorophyll fluorescence ratios during recovery

36 from drying at 93% RH, but no other samples recovered more than 10% of their

Fv/Fm values (Fig. 12). Unpaired, two-tailed T-tests comparing data at each RH reveal no significant differences between recovery of GR and VX samples at any RH.

The greatest apparent difference between the ecotypes is seen for survival data after dehydration at 89% RH (Fig. 12). The data for the GR samples were highly variable at this RH, as indicated by the large standard deviation of the mean seen in

Fig. 11. Individual trials (Appendix, Figs. A4-A6) showed variable responses for GR at 89% RH: samples from Trial 1 had 50% survival, samples from Trial 2 had 87% survival, and samples from Trial 3 had only 5% survival. In contrast, the VX samples at 89% RH had a maximal recovery indicating only 5% survival (Figs. 12, A1-A3).

Another way of quantifying survival is to look at individual samples in each trial. Based on the recovery of chlorophyll fluorescence ratios during rehydration, with 50% of the initial ratio taken as the minimum criterion for survival, five out of nine samples of the GR ecotype survived dehydration to equilibrium with 89% RH, whereas no samples of the VX ecotype survived this RH.

Visual Observations of Survival

In addition to the chlorophyll fluorescence measurements, visual assessments were made to indicate survival. Visual estimates suggest that moss samples with high Fv/Fm measurements had more than 30% of the tissue remaining green, as described by Koster et al. (2010). Moss tissue that became brown during rehydration had low Fv/Fm measurements.

DISCUSSION AND CONCLUSION

My results suggest that the ‘Gransden’ and ‘Villersexel’ ecotypes of P. patens differ in their dehydration tolerance, specifically at 89% RH, possibly due to the climatological differences in each ecotypes’ origin. Unlike samples of the

‘Villersexel’ ecotype, samples of the ‘Gransden’ ecotype survived at 89% RH, making it more dehydration tolerant. Specifically, among the three experimental trials, five out of nine GR ecotype samples equilibrated at 89% RH recovered chlorophyll fluorescence ratios similar to their pre-drying values. This ratio is similar to that reported by Koster et al. (2010), who studied the ‘Gransden’ ecotype and found approximately 50% survival at 91% RH. No ‘Villersexel’ samples survived dehydration to the RH value in the current study.

Analyzing the water contents for each ecotype determines the extent of dehydration, the potential cellular and membrane damage, and the resistance to water loss at various humidity levels (Koster et al., 2010). The final water contents of ‘Gransden’ and ‘Villersexel’ ecotypes in my study are comparable to those of

Koster and colleagues’ 2010 study, where P. patens at 95% RH had water contents of

0.64 g H2O/g DW and 0.32 g H2O/g DW at 91% RH. In that study, samples with water contents less than 0.4 g H2O/g DW lost their green color and did not resume growth (Koster et al., 2010). These findings are similar to those of my study. On average, ‘Gransden’ samples equilibrated at 89% RH had water contents of 0.623 g

38 H2O/g DW and recovered about 50% of their initial chlorophyll fluorescence ratio after rehydration, while samples at 86%, 75% and 33% RH had final water contents less than 0.4 g H2O/g DW, lost their green color, and did not survive. When comparing the final water contents though, there were no significant differences between the two ecotypes at any RH level (Fig. 9).

Due to the difference in Little Gransden, England’s yearly precipitation, which is only half that of Villersexel, France, it is possible that the drier climate contributes to the ‘Gransden’ ecotype’s ability to survive lower RH levels. Similar findings have been shown in other plant species, such as the grass Dactylis glomerata and giant cane Arundo donax, where variations in climate zones led to differences in drought and dehydration tolerance among ecotypes. In one study,

Kallida et al. (2012) examined six ecotypes of Dactylis glomerata from various climates of Morocco. After exposing each ecotype to drought conditions from 21 until 33 days, they determined that the ecotypes had varying dehydration tolerance and that only D1 and D6 ecotypes, which were from areas of low hydration and low soil water content, were able to survive a 33 day drought with only 1% soil moisture

(Kallida et al., 2012). A similar study by Ahrar et al. (2017) used two ecotypes of

Arundo donax: a Bulgarian (BG) ecotype, which originated from a drier climate, and an Italian (IT) ecotype, which originated from a more mesic environment. After 4-6 weeks of severe drought stress, photosynthesis was significantly higher in the BG ecotype and the intrinsic water use efficiency was much higher in BG than in the IT ecotype (Ahrar et al., 2017). Ahrar and colleagues stated that the BG ecotype most likely developed a better strategy to protect the photosynthetic apparatus and leaf

39 chloroplasts (Ahrar et al., 2017). Both studies reveal that differences in drought tolerance can be attributed to the climatological differences in each ecotypes’ origin.

The drying rate of the moss can also have an impact on its dehydration tolerance (Stark et al., 2016; Xiao et al., 2018). Plants with extremely slow drying times sometimes have better dehydration tolerance than plants that are rapidly equilibrated (Stark et al., 2016; Xiao et al., 2018). Because ‘Villersexel’ ecotype samples took longer to equilibrate than ‘Gransden’ ecotype samples at 93% RH (Fig.

3), we might expect to see improved survival and higher chlorophyll fluorescence ratios in the ‘Villersexel’ ecotype. But, at 93% RH, there was no significant difference in the survival as measured by recovery of initial fluorescence (Fig. 12).

It is doubtful that the difference in chlorophyll fluorescence and survival of the moss ecotypes at 89% RH can be attributed to the rate at which the moss dried because the equilibration period was not long enough to have made a difference between the two ecotypes. According to Greenwood and Stark (2014), to greatly improve the dehydration tolerance, the equilibration period should be prolonged to >150 h.

Samples in my study all equilibrated to 89% RH by 72 h, so the drying rate should not have changed the dehydration tolerance of either ecotype.

The only difference in drying rate measured in my study was in the KNO3 chamber (93% RH), in which ‘Villersexel’ samples took longer than ‘Gransden’ samples to reach equilibrium. The difference in drying rate may have been a result of difference in growth morphology between the ecotypes (Hiss et al., 2017). The gametophore stems in the ‘Villersexel’ line appeared to be denser, increasing the

40 boundary layer resistance to diffusion of water out of the moss. According to Fick’s

Law, the rate of diffusion is determined by the density of the moss clump, relative humidity, the stillness of the air, and the temperature (Taiz et al., 2014). Fick’s Law for water vapor diffusion is shown by the following equation,

퐽푤푣 = Δ푐푤푣/푅푤푣

where Jwv is the flux of water vapor out of the clump of moss, Δcwv is the water vapor concentration gradient, which is dependent on the RH, and Rwv is the resistance to water vapor movement. Because of ‘Villersexel’’s denser clump morphology, the boundary layer resistance to water diffusion to the air of the chamber would be greater than that of ‘Gransden’. This would decrease the overall rate of diffusion of water across the surface, increasing the time it took ‘Villersexel’ to equilibrate, specifically at 93% RH (Fig. 3). At lower humidity levels, where the density of the moss had less effect on the diffusion rate because the water vapor concentration gradients were much steeper, both ecotypes of moss equilibrated at similar rates.

Other studies using P. patens have also shown that the drying rate depends on the RH (Koster et al., 2010; Xiao et al., 2018). According to a Koster et al. (2010) study, ‘Gransden’ samples at 97% RH and 91% RH equilibrated at 144 h, which is slower than drying rates in my study, where ‘Gransden’ at both 93% RH (KNO3) and

89% RH (MgSO4) equilibrated at 72 h (Figs. 3 and 4). In the Koster et al. study, at

86% RH, ‘Gransden’ samples equilibrated by 120 h, whereas my 86% RH ‘Gransden’ samples equilibrated by 48 h. Lastly, both the 75% and 68% RH ‘Gransden’ samples

41 in the Koster et al. (2010) study equilibrated by 72 h, whereas the 75% RH

‘Gransden’ samples in my study equilibrated by 48 h. ‘Gransden’ ecotype samples equilibrated faster at all RH levels in the current study when compared to the samples in the Koster et al. 2010 study. Xiao et al. (2018) also found slower drying rates for ‘Gransden’ samples exposed to both 20% and 50% RH; however, samples of moss in this study were equilibrated with water added so that high RH values

(70-90%) were maintained in the chamber for about 72 h before dropping. In the study by Xiao et al. (2018), the very slow drying rates allowed the plants to develop increased tolerance of desiccation.

The recovery of initial chlorophyll fluorescence ratios validates the

‘Gransden’ ecotype’s ability to revive after drying at 89% RH. When comparing the chlorophyll fluorescence prior to drying to the chlorophyll fluorescence after rehydration and recovery, the mean values for both ecotypes reached above 60% of the initial Fv/Fm ratio after mild dehydration at 93% RH (Fig. 12). For samples dried at 89% RH, only ‘Gransden’ samples recovered to over 40% of the initial Fv/Fm, whereas ‘Villersexel’ samples did not revive (Fig. 12). This indicates that both ecotypes could repair PSII after mild dehydration, but only ‘Gransden’ could recover after more severe water loss.

The chlorophyll fluorescence measurements for both ecotypes in my study were comparable to those of other studies of moss desiccation tolerance. In a study conducted by Mayaba et al. (2001), Fv/Fm of the moss Atrichum androgynum after desiccation and rehydration ranged from 0.400 to 0.500. Survival of the moss in

42 that study was improved by a pre-treatment with the stress hormone abscisic acid

(ABA). Another study conducted by Beckett et al. (2000) also pretreated the moss A. androgynum with ABA and measured the desiccation tolerance using chlorophyll fluorescence. During rehydration, the moss had chlorophyll fluorescence ratios ranging from 0.600 to 0.650. Although my study did not use ABA treatments, the

Fv/Fm data for moss that survived desiccation in the studies by Mayaba et al. (2001) and Beckett et al. (2000) were comparable to those of the Physcomitrella dehydrated at 89% and 93% RH. After 72 h of rehydration, samples of the

‘Villersexel’ ecotype had adjusted average Fv/Fm measurements of 0.418 after recovery from drying at 93% RH (Fig. 10). For lower RH levels, ‘Villersexel’ samples did not survive, and their Fv/Fm measurements were 0.000 (Fig. 10). After 72 h of rehydration, samples of the ‘Gransden’ ecotype had mean Fv/Fm measurements of about 0.524 after drying at 93% RH and 0.324 at 89% RH (Fig. 11). The chlorophyll fluorescence measurements reveal that ‘Gransden’ samples did not revive as well as

‘Villersexel’ samples at 93% RH, but had better survival at 89% RH.

Other studies of moss with intrinsic desiccation tolerance by Proctor and colleagues (Proctor and Smirnoff, 200; Proctor et al., 2007) found much more rapid recovery of chlorophyll fluorescence ratios during rehydration than were seen in our study. In those studies, full recovery after desiccation was measured in less than 24 h, suggesting that the intrinsically tolerant moss species experience less damage during water loss than experienced by constitutively tolerant P. patens.

43 Various human factors affected the accuracy of the research. Moss used in

Trials 2 and 3 was not grown with ammonium tartrate, and as a result, had increased caulonema growth, rather than chloronema. This is because the chemical form of nitrogen used by Physcomitrella influences development, with nitrate leading to more caulonema and ammonium leading to more chloronema (Cove et al.,

2006). Since caulonema more frequently gives rise to gametophores rather than protonema (Cove, 2005), this change in the medium composition led to a less dense growth of gametophores in later trials. In addition, RH was not measured in Trial 1, due to the lack of RH meters. Because of the lack of RH meters for Trial 1, it is not certain that the RH levels were accurate.

Future studies should randomize the samples of moss in RH chambers, so that any variability among RH chambers with the same salt solutions affects both ecotypes equally. In addition, RH should be closely monitored in order to assure accuracy in the measurements.

The results reported in this thesis support the hypothesis that the ‘Gransden’ ecotype is more tolerant of dehydration than the ‘Villersexel’ ecotype. However, because of the possible sources of error noted above, additional replication will be needed in order to affirm this conclusion.

APPENDIX

Chlorophyll Fluorescence Data: Individual Trials

Data from individual experimental trials showed variability in the chlorophyll fluorescence recovery during rehydration. Those data are described here for reference.

Figures A1 and A2 show Fv/Fm measurements throughout the rehydration period for both ‘Gransden’ and ‘Villersexel’. The figures reveal that the Fv/Fm measurements were very similar between both ecotypes of moss, except for samples that had been placed in 89% RH chambers. For ‘Villersexel’, the samples that had equilibrated to 93% RH started the rehydration period with Fv/Fm at about

0.450, but values increased to around 0.700 at 72 h, showing substantial recovery.

‘Gransden’ samples that had equilibrated to 93% RH started with lower chlorophyll fluorescence than ‘Villersexel’ at Fv/Fm equal to about 0.350, but values increased to almost 0.700 by the end of the rehydration period, also demonstrating revival of the moss.

46 Trial 1 VX 0.8

0.7 VX 93% RH

m 0.6

v VX 89%

F 0.5

e RH

c 0.4 VX 86%

e RH

o 0.3

u l VX 75%

l l 0.2 RH

p VX 33% o 0.1 r RH

C 0 0 24 48 72 Time Rehydrated (h)

Figure A1: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 1 of the ‘Villersexel’ ecotype. Three measurements were taken on each of three samples after the 144 h equilibration period, starting after 1 h of being rehydrated on BCD medium. Recordings were repeated every 24 h for 3 days. Each point represents the mean Fv/Fm for the three replicate measurements of three samples and error bars represent standard deviations (n=3).

Trial 1 GR 0.8

0.7 GR 93% RH

m 0.6

v GR 89%

F 0.5

e RH

c 0.4 GR 86%

e RH

o 0.3

F GR 75%

l l RH

y 0.2

o GR 33% r 0.1

o RH

C 0 0 24 48 72 Time Rehydrated (h)

Figure A2: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 1 of the ‘Gransden’ ecotype. Three measurements were taken on each of three samples after the 144 h equilibration period, starting after 1 h of being rehydrated on BCD medium. Recordings were repeated every 24 h for 3 days. Each point represents the mean Fv/Fm for the three replicate measurements of three samples and error bars represent standard deviations (n=3). 47

48 The ecotypes differ most drastically after equilibration to 89% RH, where the

Fv/Fm decreased in ‘Villersexel’ and increased in ‘Gransden’ over time. Both ecotypes started with Fv/Fm around 0.100 but ‘Villersexel’ dropped down to 0.000 by 72 h, while ‘Gransden’ increased considerably, reaching Fv/Fm of 0.600 by 48 h, only to decline slightly by 72 h.

As for the samples that had been dehydrated at 86%, 75%, and 33% RH, the

Fv/Fm measurements during rehydration were very similar between the two ecotypes. Chlorophyll fluorescence for the three lowest humidity levels originated at about 0.050 after 1 h of rehydration, and declined slightly until reaching 0.000 at

72 h for all samples.

Figure A3: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 2 of the ‘Villersexel’ ecotype. Three measurements were taken on each of three samples after the 144 h equilibration period, starting after 1 h of being rehydrated on BCD medium. Recordings were repeated every 24 h for 3 days. Each point represents the mean Fv/Fm for the three replicate measurements of three samples and error bars represent standard deviations (n=3).

Trial 2 GR 0.8

0.7 GR 93% RH

m 0.6

v GR 89%

F 0.5 RH

c 0.4 GR 86%

e RH

o 0.3

F GR 75%

l l 0.2 RH

o GR 33%

r 0.1

o RH

C 0 0 24 48 72 Time Rehydrated (h)

Figure A4: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 2 of the ‘Gransden’ ecotype. Three measurements were taken on each of three samples after the 144 h equilibration period, starting after 1 h of being rehydrated on BCD medium. Recordings were repeated every 24 h for 3 days. Each point represents the mean Fv/Fm for the three replicate measurements of three samples and error bars represent standard deviations (n=3). 50

51 The data in Figures 12 and 13 reveal similar findings as in Figures 10 and 11.

In Trial 2, the figures reveal that the Fv/Fm measurements are very similar between both ecotypes of moss, excluding samples that were placed in 89% RH chambers.

‘Villersexel’ started with higher Fv/Fm measurements for samples that had equilibrated to 93% RH, but by 72 h, they both recovered similarly, with chlorophyll

Fv/Fm measurements at 0.600.

Similarly to Trial 1, the ecotypes differed in their recovery for the samples held over MgSO4 in Trial 2. It is important to note that the ‘Villersexel’ samples held over MgSO4 had an expected RH of 89%, but measured at 93% RH during the dehydration period. Both ecotypes had chlorophyll fluorescence measurements around 0.100, but only ‘Gransden’ recovered. ‘Villersexel’ declined in Fv/Fm, where

‘Gransden’ recovered slightly, reaching 0.400.

The three lowest humidities for both ecotypes had Fv/Fm that fluctuated between the 0.000 and 0.100 range, showing higher chlorophyll fluorescence measurements than Trial 1, but still no indicated photosynthetic recovery.

52 Trial 3 VX 1 0.9 VX 93% 0.8 RH

/ 0.7 v VX 89%

e 0.6 RH

c 0.5 VX 86%

e RH

o 0.4

VX 75% l 0.3

y RH

p 0.2

r VX 33%

l 0.1 RH

h C 0 0 24 48 72 Time Rehydrated (h)

Figure A5: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 3 of the ‘Villersexel’ ecotype. Three measurements were taken on each of three samples after the 144 h equilibration period, starting after 1 h of being rehydrated on BCD medium. Recordings were repeated every 24 h for 3 days. Each point represents the mean Fv/Fm for the three replicate measurements of three samples and error bars represent standard deviations (n=3).

Trial 3 GR 0.8

0.7 GR 93% RH

m 0.6

v GR 89%

F 0.5 RH

e 0.4 GR 86%

c s RH

e r 0.3

l GR 75%

l RH

l 0.2

p 0.1 GR 33%

r RH

h 0 C 0 24 48 72 Time Rehydrated (h)

Figure A6: Chlorophyll fluorescence (Fv/Fm) measurements during rehydration for Trial 3 of the ‘Gransden’ ecotype. Three measurements were taken on each of three samples after the 144 h equilibration period, starting after 1 h of being rehydrated on BCD medium. Recordings were repeated every 24 h for 3 days. Each point represents the mean Fv/Fm for the three replicate measurements of three samples and error bars represent standard deviations (n=3).

53 The data for Trial 3 reveal much lower Fv/Fm measurements during rehydration than were found in Trials 1 and 2. For samples that were held over a

KNO3 salt slurry, ‘Villersexel’ reached a Fv/Fm of 0.400 by 72 h while ‘Gransden’ only increased to 0.100 at 24 h and then decreased until to 0.000 at 72 h. Again, it is important to note that the ‘Gransden’ KNO3 chamber had a measured RH of 85%, rather than the expected 93% RH. The low chlorophyll fluorescence for samples held over KNO3 indicates the poor health of the moss; death of moss at 93% RH is not common. The samples of moss in the KNO3 chamber for ‘Gransden’ Trial 3 had

Fv/Fm measurements of 0.000 by 72 h, which skewed the overall mean for

‘Gransden’ KNO3.

In Trial 3, ‘Villersexel’ samples that had equilibrated to 89%, 86%, 75%, and

33% continued the trend as seen in Trials 1 and 2. But, in ‘Gransden’ the Fv/Fm measurements for all humidity levels remained under 0.140. Again, the low chlorophyll fluorescence measurements in Trial 3 of ‘Gransden’ skewed the average data for all three trials.

BIBLIOGRAPHY

Ahrar, M., Doneva, D., Tattini, M., Brunetti, C., Gori, A., Rodeghiero, M., Wohlfahrt, G., Biasioli, F., Varotto, C., Loreto, F., Velikova, V. (2017). Phenotypic differences determine drought stress responses in ecotypes of Arundo donax adapted to different environments. Journal of Experimental Botany 68(9), 2439-2451. doi: 10.1093/jxb/erx125.

Alpert, P. (2005). The Limits and Frontiers of Desiccation-Tolerant Life. Integrative & Comparative Biology 45(5), 685-695.

Beckett, R.P., Csintalan, Z., Tuba, Z. (2000). ABA treatment increases both the desiccation tolerance of photosynthesis, and nonphotochemical quenching in the moss Atrichum undulatum. Plant Ecology 151, 65–71.

Brookes, A.J. (1999). The essence of SNPs. Gene 234(2), 177-186.

Bryant, G., Koster, K.L., Wolfe, J. (2001). Membrane behaviour in seeds and other systems at low water content: the various effects of solutes. Seed Science Research 11, 17–25. doi: 10.1079/SSR200056

Bryant, G., Wolfe, J. (1992) Interfacial forces in cryobiology and anhydrobiology. Cryo-Letters 13, 23–36.

Chen, D., Chen, H.W. (2013). Using the Köppen classification to quantify climate variation and change: An example for 1901–2010. Environmental Development 6, 69-79. doi:10.1016/j.envdev.2013.03.007

Cove, D. (2005). The Moss Physcomitrella patens. Annu. Rev. Genet. 39, 339–358. doi: 10.1146/annurev.genet.39.073003.110214

55 Cove, D., Bezanilla, M., Harries, P., Quatrano, R. (2006). Mosses as Model Systems for the Study of Metabolism and Development. Annual Review Of Plant Biology 57(1), 497-520. doi:10.1146/annurev.arplant.57.032905.105338

Cove, D.J., Perroud, P.F., Charron, A.J., McDaniel, S.F., Khandelwal, A., Quatrano, R.S. (2009) Culturing the moss Physcomitrella patens. Cold Spring Harb Protoc. doi: 10.1101/pdb.prot5136

Cuming, A.C., Cho, S.H., Kamisugi, Y., Graham, H., Quatrano, R.S. (2007). Microarray analysis of transcriptional responses to abscisic acid and osmotic, salt, and drought stress in the moss, Physcomitrella patens. New Phytologist 176(2), 275-287. doi:10.1111/j.1469-8137.2007.02187.x

'eFloras (2017). Physcomitrella patens. Published on the Internet http://www.efloras.org [accessed 13 September 2017]. Missouri Botanical Garden, St. Louis, MO & Harvard University Herbaria, Cambridge, MA.

Engel, P.P. (1968). The Induction of Biochemical and Morphological Mutants in the Moss Physcomitrella patens. American Journal of Botany 55, 438–446.

Frank, W., Ratnadewi, D., Reski. R. (2005). Physcomitrella patens is highly tolerant against drought, salt and osmotic stress. Planta 220, 384–394. doi:10.1007/s00425-004-1351-1

Greenwood, J.L., Stark, L.R. (2014). The rate of drying determines the extent of desiccation tolerance in Physcomitrella patens. Functional Plant Biology 41, 460–467. doi:10.1071/FP13257

Henriques, F.S. (2009) Leaf Chlorophyll Fluorescence: Background and Fundamentals for Plant Biologists. Botanical Review 75(3), 249-270. doi:10.1007/sl2229-009-9035-y

Hiss, M., Meyberg, R., Westermann, J., Haas, F.B., Schneider, L., Schallenberg- Rüdinger, M., Ullrich, K.K., Rensing, S.A. (2017). Sexual reproduction, sporophyte development and molecular variation in the model moss Physcomitrella patens: introducing the ecotype Reute. The Plant Journal 90, 606–620. doi: 10.1111/tpj.13501

56 Hoekstra, F.A., Golovina, E.A., Buitink, J. (2001). Mechanisms of plant desiccation tolerance. Trends Plant Sci 6, 431–438.

Kamisugi, Y., von Stackelberg, M., Lang, D., Care, M., Reski, R., Rensing, S. A., Cuming, A.C. (2008). A sequence-anchored genetic linkage map for the moss, Physcomitrella patens. Plant Journal 56(5), 855-866. doi:10.1111/j.1365- 313X.2008.03637.x

Kallida, R., Shaimi, N., Volaire, F. (2012). Dehydration tolerance and drought survival summer dormant Moroccan ecotypes cocksfoot (Dactylis glomerata L.). In: New approaches for grassland research in a context of climate and socio- economic changes (Acar Z., López-Francos A., Porqueddu C. eds.). pp. 77- 82.

Koster, K.L., Balsamo, R.A., Espinoza, C., Oliver, M.J. (2010). Desiccation sensitivity and tolerance in the moss Physcomitrella patens: assessing limits and damage. Plant Growth Regulation 62, 293–302. doi:10.1007/s10725-010- 9490-9

Koster, K.L., Bryant, G. (2006). Dehydration in Model Membranes and Protoplasts: Contrasting Effects at Low, Intermediate and High Hydrations. In Cold Hardiness in Plants (Chen, T.H.H., et al., eds.). CABI, Wallingford, UK, pp. 219– 234.

Maxwell, K., Johnson, G.N. (2000). Chlorophyll fluorescence – a practical guide. Journal of Experimental Botany 51, 659–668. doi:10.1093/jexbot/51.345.659

Mayaba, N., Beckett, R.P., Csintalan, Z., Tuba, Z. (2001). ABA Increases the Desiccation Tolerance of Photosynthesis in the Afromontane Understorey Moss Atrichum androgynum. Annals of Botany 88(6), 1093- 1100. doi:10.1006/anbo.2001.1552

Merkel, Alexander. “Climate: Little Gransden.” Climate-Data.org, 9 Aug. 2015, en.climate-data.org/location/101109/.

Merkel, Alexander. “Climate: Villersexel.” Climate-Data.org, 9 Aug. 2015, en.climate- data.org/location/486160/.

57 Meyberg, Rabea. “SNPs ‘Villersexel’ vs ‘Gransden’ .” CoGe: ExperimentView, CoGe, 3 Nov. 2016, genomevolution.org/coge/ExperimentView.pl?eid=9128

Oldenhof, H., Wolkers, W.F., Bowman, J.L., Tablin, F., Crowe, J.H. (2006). Freezing and desiccation tolerance in the moss Physcomitrella patens: An in situ Fourier transform infrared spectroscopic study. Biochimica et Biophysica Acta 1760(8), 1226-1234. doi:10.1016/j.bbagen.2006.03.025

Oliver, M.J., Cushman, J.C., Koster, K.L. (2010). Dehydration Tolerance in Plants. Plant Stress Tolerance, Methods in Molecular Biology (R. Sunkar eds.). 639, 3-24. doi:10.1007/978-1-60761-702-0_1

Perroud, P., Cove, D.J., Quatrano, R.S., McDaniel, S.F. (2011). An experimental method to facilitate the identification of hybrid sporophytes in the moss Physcomitrella patens using fluorescent tagged lines. New Phytologist 191(1), 301-306. doi:10.1111/j.1469-8137.2011.03668.x

Pressel, S., Duckett, J.G. (2010). Cytological insights into the desiccation biology of a model system: moss protonemata. New Phytologist 185(4), 944-963. doi:10.1111/j.1469-8137.2009.03148.x

Prigge, M. J., Bezanilla, M. (2010). Evolutionary crossroads in developmental biology: Physcomitrella patens. Development 137, 3535-3543. doi:10.1242/dev.049023

Proctor, M.C.F., Ligrone, R., Duckett, J.G. (2007) Desiccation tolerance in the moss Polytrichum formosum: physiological and fine-structural changes during desiccation and recovery. Annals of Botany 99: 75-93.

Proctor, M.C.F., Oliver, M.J., Wood, A.J., Alpert, P., Stark, L.R., Cleavitt, N.L., Mishler, B.D. (2007). Desiccation-Tolerance in Bryophytes: A Review. The Bryologist 110, 595–621. doi:10.1639/0007-2745(2007)110[595:DIBAR]2.0.CO;2

Proctor, M.C.F., Smirnoff, N. (2000) Rapid recovery of photosynthesis on rewetting desiccation-tolerant mosses: chlorophyll fluorescence and inhibitor experiments. Journal of Experimental Botany 51: 1695-1704.

58 Rathnayake KN, BWM Wone, DC Alexander, KL Koster. 2017. The acquisition of desiccation tolerance in the moss Physcomitrella patens: a metabolomic analysis. Plant Biology 2017 Abstract.

Rensing, S.A., Lang, D., Zimmer, A.D., Terry, A., Salamov, A., Shapiro, H., Yamaguchi, K. (2008). The Physcomitrella Genome Reveals Evolutionary Insights into the Conquest of Land by Plants. Science 319, 64-69. doi:10.1126/science.1150646

Reski, R., Frank, W. (2005). Moss (Physcomitrella patens) functional genomics -- Gene discovery and tool development, with implications for crop plants and human health. Briefings In Functional Genomics & Proteomics 4(1), 48-57.

Schaefer, D.G., Zrÿd, J.P. (2001). The Moss Physcomitrella patens, Now and Then. Plant Physiol. 127, 1430-1438.

Taiz L., Zeiger, E., Møller, I.M., Murphy, A. (2014). Plant Physiology and Development, 6th ed. Sinauer Press, Sunderland, MA, USA. pp. 110-111. von Stackelberg, M., Rensing, S.A., Reski, R. (2006) Identification of genic moss SSR markers and a comparative search in twenty-four algal and plant gene indices reveal species-specific rather than group-specific characteristics of microsatellites. BMC Plant Biol. 6, 9.

Stark L.R., Greenwood J.L., Slate M.L., Brinda, J.C. (2016) Syntrichia norvegica shoots exhibit a complex inducible response to desiccation: separating the effects of rate of drying and water content. Botany 95(5), 481-491.

Walters, C., Koster, K. L. (2007). Structural Dynamics and Desiccation Damage in Plant Reproductive Organs. In Plant Desiccation Tolerance (Jenks, M.A. and Wood, A.J., eds.). Blackwell Publishing, Iowa, USA, pp. 251–280.

Wang, X. Q., Yang, P. F., Liu, Z., Liu, W. Z., Hu, Y., Chen, H., … He, Y. K. (2009). Exploring the Mechanism of Physcomitrella patens. Desiccation Tolerance through a Proteomic Strategy. Plant Physiology 149(4), 1739–1750. doi:10.1104/pp.108.131714

59 Xiao, L., Yobi, A., Koster, K.L., He, Y., Oliver, M.J. (2018). Desiccation tolerance in Physcomitrella patens: Rate of dehydration and the involvement of endogenous abscisic acid (ABA). Plant, cell & environment. 41(1), 275-284 doi: 10.1111/pce.13096

IMAGE CREDITS

Fig. 1.1: “Physcomitrella patens (a, 144649-481016) 6963” by Hermann Schachner is licensed under CC CC0 1.0 Universal Public Domain Dedication.

Fig. 1.2: “Physcomitrella patens (a, 144649-481016) 6964” by Hermann Schachner is licensed under CC CC0 1.0 Universal Public Domain Dedication.

Fig. 1.3: “Physcomitrella Protonema” by Anja Martin, Labor Ralf Reski is licensed under CC CC0 1.0 Universal Public Domain Dedication.

Fig. 1.4: “Physcomitrella” by Pirex is licensed under English Wikipedia.

Fig. 1.5: “Physcomitrella Sporophyte” by Ralf Reski is licensed under CC Attribution- Share Alike 1.0 Generic.

Fig 2: Google Maps. (2017). Villersexel and Little Gransden. Retrieved from https://www.google.com/maps/d/edit?mid=1RJ7KO7uLK35EBP8L5qsXO8p qmro&ll=49.96262625891726%2C1.0640768203124935&z=6