Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of temperature, photoperiod, and leaf age on foliar senescence in western larch (larix occidentalis… Rosenthal, Selma Iduna 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1995-983434.pdf [ 3.5MB ]
JSON: 831-1.0088906.json
JSON-LD: 831-1.0088906-ld.json
RDF/XML (Pretty): 831-1.0088906-rdf.xml
RDF/JSON: 831-1.0088906-rdf.json
Turtle: 831-1.0088906-turtle.txt
N-Triples: 831-1.0088906-rdf-ntriples.txt
Original Record: 831-1.0088906-source.json
Full Text

Full Text

THE EFFECTS OF TEMPERATURE, PHOTOPERIOD, AND LEAF AGEON FOLIAR SENESCENCE iN WESTERN LARCH(LARIX OCCIDENTALIS NUTr.)bySELMA IDUNA ROSENTHALA.B., Bowdoin College, 1985M.S., George Washington University, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Botany)We accept this thesis as conformingto the required standardTHE UNIVERSfiEY OF BRITISH COLUMBIAApril 1995© Selma Iduna Rosenthal, 1995In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of_________________The University of British ColumbiaVancouver, CanadaDate______DE-6 (2/88)11ABSTRACTDeciduous trees integrate environmental signals to determine the onset and rate ofautumnal foliar senescence. In this thesis I studied that process by characterizingenvironmentally-induced foliar senescence in western larch (Larix occidentalis Nutt.). Iobtained measurements from seedlings senescing outdoors and from those induced to senescein environmentally controlled chambers. The data were examined using graphs and regressions.Findings from outdoor growing seedlings were similar to results from seedlings that senescedin growth chambers, suggesting that both experimental designs allow one to study thesenescence process effectively.I quantified the timing and process of senescence as it varied with leaf age andenvironmental conditions. In most experiments, senescing needles displayed a decline inpigment levels first, followed by carbon assimilation and Rubisco amount, and finally,chlorophyll a/b ratio and quantum yield. Warmer air temperature delayed the onset ofchlorophyll and photosynthetic decline. Extended photoperiod delayed the onset of chlorophylldecline but did not affect the timing of photosynthetic. decline. Neither air temperature orphotoperiod directly affected the onset of decline in chlorophyll a/b ratio. Instead, the initialdecline in pigment and assimilation rates may have caused the subsequent decline in chlorophylla/b ratio. Finally, increased leaf age accelerated the onset of decline of all measures ofsenescence.Environmental conditions affected not only the onset but also the rate of senescence.While extended photoperiods delayed the onset of pigment loss, the rate of photosyntheticdecline, once it began, was unaffected by extended photoperiod. In contrast, seedlings grown111in 8°C above ambient warmer soil had higher photosynthetic rates per unit chlorophyll thancontrol seedlings while seedlings grown in 3°C warmer soil did not display differentphotosynthetic rates from control seedlings.Using new methods developed in this thesis, I provide evidence that leaves integratetemperature and photoperiod signals differently. I show that it is possible to quantify the relativeimportance of these conditions in stimulating autumn chioroplast senescence. Furthermore, ifdifferences in the weather and age of the leaf are accounted for, it is possible to predict thetiming of pigment loss and rate of photosynthetic decline during autumn.ivTABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS ivLIST OF TABLES viLIST OF FIGURES viiACKNOWLEDGEMENTS viiiCHAPTER 1: INTRODUCTION 1CHAPTER 2: A LITERATURE REVIEW: ENVIRONMENTAL CONDITIONS INDUCiNGLEAF SENESCENCE 72.1 INTRODUCTION 72.2 ENVIRONMENTAL CONDITIONS 82.2.1 Light Level and Quality 82.2.2 Photoperiod 122.2.3 Water 132.2.4 Temperature 162.2.5 Mineral nutrition 182.2.6 Carbon dioxide 222.2.7 Pollution 242.3 CONCLUSIONS 28CHAPTER 3: ONSET OF FOLIAR SENESCENCE DURING AUTUMN: THE EFFECTS OFSOIL TEMPERATURE AND PHOTOPERIOD 323.0 ABSTRACT 323.1 INTRODUCTION 323.2 MATERIALS AND METHODS 343.2.1 Experimental design 343.2.2 Pigment and photosynthesis measurements 353.2.3 Chlorophyll a fluorescence measurements 363.2.4 Determination of breakpoints 373.3 RESULTS 383.3.1 Environmental conditions and pigment loss 383.3.2 Gas Exchange 493.3.3 Chlorophyll a fluorescence and quantum yield 563.4 DISCUSSION 63VCHAPTER 5: THE ONSET OF FOLIAR SENESCENCE IN ENVIRONMENTALLYCONTROLLED CHAMBERS: THE EFFECTS OF AIR TEMPERATURE,PHOTOPEPJOD, AND LEAF AGE 865.0 ABSTRACT 865.1 INTRODUCTION 865.2 MATERIALS AND METHODS 885.2.1 Experimental design 885.2.2 Pigment and photosynthesis measurements 895.2.3 Determination ofbreakpoints and the onset of senescence 905.2.4 Regression model 915.3 RESULTS 935.4 DISCUSSION 100CHAPTER 6: THE RATE OF PHOTOSYNTHETIC DECLINE DURING AUTUMN LEAFSENESCENCE: THE EFFECTS OF SOIL TEMPERATURE AND PHOTOPERIOD105CHAPTER 7: CONCLUSIONS 128CHAPTER 4: FURTHER CHARACTERIZATION OF THE ONSET OF FOLIARSENESCENCE DURING AUTUMN: THE TIIvIING OF RUBISCO DEGRADATION4. AND METHODSRESULTS AND DISCUSSION7171• 7173746.0 ABSTRACT6.1 INTRODUCTION6.2 MATERIALS AND METHODS6.2.1 Experimental design6.2.2 Development ofmodel6.3 RESULTS6.4 DISCUSSION105106107107112116124CHAPTER 8: REFERENCES 132viLIST OF TABLESTable 3-1. Chlorophyll and chlorophyll a/b ratio in Larix occidentalis needles of differentsenescence stages 46Table 3-2. Internal : ambient CO2 concentration in Larix occidentalis needles of differentsenescence stages 57Table 5-1. Number of days until decline of chlorophyll (C), chlorophyll a/b ratio (AB), orphotosynthetic rates (P; measured either as carbon assimilation or oxygen evolution) forseedlings senescing under different photoperiod and temperature regimes 96Table 5-2. The effect of Larix occidentalis needle age and environmental conditions on thenumber of days in the growth chamber until chlorophyll m2, photosynthesis (measuredas CO2 assimilation or 02 evolution) and chlorophyll a/b ratio declined 98Table 6-1. Regression analysis of photosynthetic rates in Larix occidentalis seedlings, forautumn 1991 and 1992 117viiLIST OF FIGURESFigure 3-1. Temperature at the experimental site during autumn 40Figure 3-2. Environmental site conditions during autumn 42Figure 3-3. Chlorophyll and chlorophyll a/b ratio during autumn 45Figure 3-4. Determination ofdate of decline for chlorophyll, photosynthesis and chlorophyll a/bratio for naturally senescing seedlings during autumn 48Figure 3-5. Photosynthetic parameters for seedlings senescing under outdoor conditions duringautumn 51Figure 3-6. Photosynthetic parameters for seedlings senescing under warm soil during autumn.53Figure 3-7. Stomatal conductance (mmol m2 s’) during autumn 55Figure 3-8. Oxygen evolution per leaf area (20°C, 50 mL L’ CO2, 660nm light) at differentirradiances for naturally senescing seedlings in different phases of senescence 59Figure 3-9. Quantum yield ofPSll (c1),) calculated as in Genty et al. 1989, compared to oxygenquantum yield ((1)02) 62Figure 3-10. Chlorophyll a fluorescence quenching as a function of absorbed PFD (i.tmol m2 s’).65Figure 4-1. Carbon assimilation and Rubisco amounts during autumn in naturally senescing Larixneedles. a) Control and b) warm soil seedlings 76Figure 4-2. Immunoblot oflarge subunit ofRubisco in autumn senescing Larix needles, materialprepared with several protease inhibitors but not leupeptin 79Figure 4-3. Immunoblot of large subunit ofRubisco in autumn senescing Larix needles, materialprepared with several protease inhibitors including leupeptin 81Figure 4-4. Large subunit of Rubisco and proteolytic fragments in autumn senescing larchneedles prepared in buffers with different protease inhibitors 83Figure 5-1. Chlorophyll mg m2(a), chlorophyll a/b ratio (b) and CO2 assimilation m2 (c) fromwestern larch (Larix occidentalis) seedlings placed into environmentally controlledchambers 95Figure 6-1. Carbon assimilation rates for control (filled circles) and warm soil (open circles)seedlings in (a) 1992 and (b) 1991 109Figure 6-2. Chlorophyll content in control (filled circles) and warm soil (open circles) seedlingsin (a) 1992 and (b) 1991 111Figure 6-3. Air temperature (°C) at needle harvest 120Figure 6-4. Minimum soil temperatures (°C) for control (lower line) and warm soil (upper line)seedlings in a) 1992 and b) 1991 122yinACKNOWLEDGEMENTSThis research was in part funded by a grant from the Natural Sciences and EngineeringResearch Council ofCanada and a GREAT award from the Science Council ofBritish Columbiaand Weyerhaeuser Canada. I thank Dr. Valerie LeMay for comments on the regression modelin Chapter 6. I also thank members of my committee, Drs. Edith Camm, Beverley Green, andlain Taylor, for helpful discussions throughout my research and in preparing this manuscript.I am especially grateful to Edith Camm for introducing me to the subject of this thesis. Finally,I would like to thank my husband and parents for their support and patience throughout thisproject.1CHAPTER 1: INTRODUCTIONSenescence is the process of getting old. This simplistic statement does not account forthe observations that young tissue can show symptoms of senescence such as chlorophyll loss,protein degradation, and eventually cellular death. Furthermore, it does not account for thevariation found in the rate and sequence of events during senescence. Since the late 1800’s, butespecially in the last twenty years, many researchers have examined plant senescence at all levels,included organisms, organs, and organelles, in an effort to understand the processes ofsenescence. The interest in senescence can be attributed to the fact that it is a fundamentalprocess that affects the life cycle of a plant.Studies on plant senescence have varied considerably, both with respect to the methodsused and the insights obtained. In part, that variation reflects the fact that monocarpic. plants(plants that die after flowering) and polycarpic plants (plants that flower, grow vegetatively, andthen can flower again before dying) provide different systems for examining senescence. Amongmonocarpic plants the interaction between nutrients during the shift from vegetative toreproductive stages is important as a senescence trigger. On the other hand, among polycarpicplants, age-related deterioration and stress conditions are the most common causes of senescence.Most often, a combination of conditions trigger senescence. For example, autumnal leafsenescence in deciduous trees occurs when autumn weather and photoperiod triggers leafsenescence as the tree becomes dormant for the winter season.The study of senescence in artificial systems is common. For both monocarpic andpolycarpic plants, senescence of detached organs such as fruits, flowers, cotyledons, and leaveshave received attention because complications caused by inter-organ communication have been2removed. Furthermore, the study of detached leaves has been a popular method to characterizesenescence because the system is easily reproduced and senescence generally occurs more quicklythan under natural conditions. In addition, it is possible to place leaves in darkness or in solutionscontaining stimulators of senescence and to record the observed effects on senescence. Whilestudies ofdetached leaves have been invaluable in defining events during senescence, researchershave also begun to focus on less artificial systems.In this dissertation I examine the onset and process of autumnal foliar senescence inattached leaves. Foliar senescence can be triggered through many types of stimuli. Internalfactors accelerating foliar senescence also include altered source-sink relations while externalfactors include natural (seasonal) and man-made environmental changes, and stresses fromnutrient deficiency and pathogen invasion. Leaf senescence is commonly a result of acombination of these events.Given the importance of the senescence process to plant development, a large literaturenot surprisingly reflects attempts to understand senescence. Nevertheless, many gaps remain.Biochemical studies have demonstrated that certain processes, such as damage from oxidativestress and protein degradation, occur during all types of senescence (Smart 1994, Pell and Dann1991, Brown et al. 1991, Kelly and Davies 1988, Stoddart and Thomas 1982). On the otherhand, other degradative processes appear to vary with plant species and the conditions underwhich leaf senescence occurs. The causes of that variation have received little attention and arenot well understood.One likely reason for differences observed in senescing systems is that plant senescencewas sensitive to differences in environmental conditions among those studies. For example, the3impacts ofenvironmental attributes on autumn senescence have been difficult to evaluate becausenumerous environmental conditions change simultaneously during autumn making it difficult toidentify effects ofindividual environmental signals on leaf degradation. Changes in environmentalconditions also affect many different features of senescing leaves in ways that are both complexand cumulative over time. As a result, isolation and quantification of individual environmentaleffects on plant senescence have proved particularly difficult.In my research I have sought to fill some of the gaps in the senescence literature byevaluating the degree to which different environmental attributes affect senescence. Toaccomplish that goal, I developed several new methods that enabled me to isolate and quantifythe effects of environmental conditions and leaf age on assimilation rates and chlorophyll levelsduring leaf senescence. I obtained measurements from both autumn senescing western larchneedles and needles senescing in controlled environments. The data were analyzed using bothgraphical and regression analyses. Results demonstrated that it was possible to quantify theeffects of changes in soil and air temperature, photoperiod, and leaf age on both the onset andprocess of leaf senescence. Moreover, the methods developed here could be applied to otherplant species and environmental conditions.I chose to study autumn foliar senescence in western larch (Larix occidentalis Nutt.)because it is a fast growing (Logan 1966), deciduous gymnosperm used in reforestation (Schmidtet al. 1976). The rapid growth rate reduces some of the methodological problems presented byother tree species. It is not known ifthe fast growth rate oflarch relative to sympatric evergreensis partly due to its deciduous nature. Water relations under stress, relatively efficient nutrientcycling, and needle morphology, are other characteristics that may contribute to the fast growth4rate (Gower and Richards 1990, Reynolds et al. 1989, Gower et al. 1987, Higgens et a!. 1987,Tyrrell and Boerner 1987, Matyssek 1986, Sandford and Jaivis 1986, Carlyle and Malcolm 1985,Tilton 1976). These studies provide data on responses by larch to favourable and unfavourablegrowth conditions. In addition, Larix occidentalis grows in latitude and altitude ranges that aresubject to different patterns of day length and temperature over the growing season. Onset ofleaf senescence differs between populations suggesting that senescence in larch is sensitive toboth photoperiod and temperature. Finally, if environmental effects on leaf senescence canextend the growing season by delaying the onset of senescence, it may be useful to test for effectsin early and late senescing hybrids. Larch species hybridize easily (Paques 1989) and additionalinformation on the model system provided by seedling physiology and sensitivity to senescencecues may be useful to improve reforestation through plant breeding programs (Rehfeldt 1982,Ledig and Botkin 1974).Chapters 3 and 4 of this dissertation contain a detailed description of the autumnsenescence process in western larch as well as the results of statistical methods designed todetermine the date of decline of different physiological parameters. I also examined the effectsofthe weather. The research focused on the chloroplast because changes in this organelle occurearly in the senescence process. I measured photosynthetic parameters in naturally senescingseedlings to characterize senescence and compared them with the process in seedlings receivingeither warmer soils or extended photoperiod. This experimental design allowed me to measurethe effect of soil temperature and photoperiod on senescence. Results showed that under mostconditions, pigments declined first, followed by carbon assimilation and then quantum yield.5However, the pattern of decline in these and other photosynthetic measures was sensitive to theenvironmental conditions.Chapter 5 contains results from experiments to induce foliar senescence in growthchambers set to different air temperature and photoperiod regimes. This experimental designallowed me to induce senescence at any time of the year, and with different air temperature andphotoperiod settings while holding constant all other environmental and leaf specific variables.I then used a regression model, with dichotomous variables identifying treatment groups, toquantify the effects of air temperature and photoperiod on the number of days until senescencebegan. Regression analysis is widely used in other fields, but its use in analyzing this plantphysiological data is both novel and appropriate. Results indicated that the onset of senescencedepends on environmental conditions and on the age of the leaf In addition, the successof thismethod in quantifying the effect of environmental conditions on components of the leaf suggeststhat future studies on the effect ofthe environment on leaf senescence may also benefit from thistype of analysis.Results in Chapter 6 show that autumn photosynthetic rates can be predicted. I developeda regression model that explained the decline in photosynthetic rates during autumn in terms ofvariables that accounted for past and current growing conditions of the leaf. The model wasestimated using three different data sets: data from two successive years of senescing autumnseedlings and data from one set of seedlings induced to senesce under cool growth chamberconditions. I found that the coefficients from separate regressions for the three different data setswere similar, showing 1) that the model was robust to substantial differences in autumn weather6and 2) that there was a predictable relationship between photosynthesis and chlorophyll duringleaf senescence when differences in weather conditions are accounted for.The principal achievements from my work (Chapter 7) are methodological advances thatgreatly improve our ability to study autumn foliar senescence. I show that the impact ofindividual environmental conditions, such as soil temperature and photoperiod, can be studiedusing outdoor growing plants in specially configured cold frames. I also demonstrate that larchtrees can be induced to senesce in growth chambers set to autumn weather conditions. Thus byvaiying growth chamber settings among experiments and comparing results, one can evaluate theimpact ofa wide variety of different environmental conditions on the senescence process. Finally,because leaf senescence typically is a function of a number of environmental variables that changeconcurrently over time (such as air and soil temperature, photoperiod, etc), two dimensionalgraphical analyses of plant characteristics against environmental attributes are often inadequate.To address that problem, in various chapters of this thesis I use regression analysis to resolve theeffects of different environmental attributes on various measures of senescence related plantdegradation.Drawing on these methods, this thesis provides the first detailed characterization ofchanges in autumn senescing leaves during autumn weather. My experiments indicate that smallchanges in soil temperature can affect photosynthetic rates during autumn, while photoperiodaffects chlorophyll levels but not assimilation rates. Finally, my work provides evidence thatleaves integrate temperature and photoperiod signals differently and that it is possible to quantif,rthe relative importance of these environmental conditions in triggering autumn chloroplastsenescence.7CHAPTER 2: A LITERATURE REVIEW: ENVIRONMENTAL CONDITIONS Th4DUCINGLEAF SENESCENCE2.1 INTRODUCTIONInterest in the influence of the environment on plant senescence can be traced back over100 years to the work ofRutland (1888) who noted that Salix babylonica near water yellowedearlier than in warmer, drier soils. An early definition of senescence was provided by Wiesnerwho discussed the effect oftemperature, frost and drought on leaf fall (Wiesner 1904, 1871). Inthe first part of the century, Molisch (1918) conducted a thorough study of the effects of light,temperature, oxygen, and age on leafyellowing in a number of different species. Since these earlyworks, numerous authors have examined the effects of environmental conditions on tissue’death.In addition, many recent studies have been conducted in a broader effort to understand the effectsofenvironmental stress on plant growth (Kozlowski 1991, Pell and Dann 1991, Larcher 1987).Nevertheless, despite considerable research, our understanding of the effects of environmentalconditions on senescence is still quite rudimentary. Partly for that reason it is difficult to makegeneralizations about the effect of environmental conditions on senescence.In this review I discuss the effects ofindividual climatic and edaphic variables on leaf andwhole plant senescence. I attempt to find common patterns among senescence triggers and topoint out differences in senescence pathways due to environmental acceleration of senescencecompared with natural ageing. I begin with the effect of light levels and photoperiod onsenescence and follow with discussion of supra- and sub-optimal water, temperature and mineral8nutrition on senescence. The human impact on plants is briefly examined by reviewing literatureon the effects of elevated carbon dioxide and other pollutants on plant senescence.2.2 ENVIRONMENTAL CONDITIONS2.2.1 Light Level and QualityThe effect of light on senescence has been extensively examined since research on thebiochemistiy of senescence began by placing detached leaves into darkness and characterizing thesubsequent senescence pathway. The process of senescence due to shading and excess light hasalso been frequently studied. Findings typically suggest that shading induces senescence primarilyby lowering carbon assimilation rates. On the other hand, the effect of high light levels onsenescence appears to be more complicated. Sever4l studies suggest that the stress ofphotoinhibition can lead to tissue death by photooxidation. In addition, it appears that in somecases high light levels accelerate the maturation process causing age-related senescence to occur.Specific examples of the effects of low, moderate, and high light levels on senescence areprovided below, along with examples of the effects of light quality on senescence. In reviewingthe effects of light levels on plants, I emphasize that low, moderate, and high light is a relativeterm that varies with plant species as well as growth and senescence conditions.Typically, low light levels trigger leaf chiorosis, although the rate and extent of the effectis sensitive to leaf age. For example, Molisch (1918) found that inner shaded leaves of somespecies senesce before sun leaves. However, he did not take into account the fact that innerleaves are often older than sun leaves. In contrast, Hopkinson (1966) grew Cucumis sativushorizontally to prevent excessive shading of inner leaves, and then subjected leaves of different9ages to the same level of shading. He found that when young sun leaves were shaded theysenesced earlier than older inner leaves, in contrast to growth under natural conditions. Thatfinding is consistent with results from a number of other studies in which younger sun leavessenesced before shaded inner leaves (e.g. Koike 1987).In contrast to low light, moderate light generally delays senescence by maintainingphotosynthetic rates and by affecting phytochrome mediated responses (Okada et al. 1992,Wilimer et al. 1988, Biswal and Biswal 1984). Burkey and Wells (1991) observed thatassimilation rates in Glycine max declined before chlorophyll when senescence was shade-induced. This suggests that the drop in photosynthesis induced by moderate light occurs earlyin the senescence process, and that the decline in assimilation rates rather than chlorophylldegradation began the senescence process. Similarly, Hi4ema et al. (1991) found that inOryzasativa a decrease in assimilation coincided with a decrease in variable fluorescence andcytochrome f levels. This suggests that decreased photosynthetic electron transport occurredwhen light levels declined. In addition to having a direct impact on photosynthesis, as in theabove examples, moderate light may indirectly delay senescence by maintaining cellular calciumlevels Huang and Kao 1992). Additionally, because calcium levels are affected by phytochrome(Shacklock et al. 1992), moderate light levels may affect phytochrome mediated senescenceprocesses.Although moderate light levels generally retard senescence, high light can accelerate it(Biswal and Biswal 1984). For example, sunlight (2000 i.tmol m2 s1) stimulated chlorophylldegradation in Oryza saliva relative to the loss in 400 J.Lmol m2 s’ (Hidema et al. 1991). InSecale cereale, grown under high light conditions, both chlorophyll and catalase decline were10accelerated relative to low light grown plants (Kar et al. 1993). These observations have alsobeen seen in perennials. Leaves of Larix seedlings grown in fbll sun began autumnal yellowingearlier than those grown in shade (Rosenthal, unpublished). In addition, chlorophyll breakdownin autumn Pinus silvestris was greater in sun-exposed than shade leaves (Oquist et al. 1978).High light also resulted in rapid senescence of shade tolerant Rhododendron maximum leaves(Nilsen and Bao 1987).In the cases where light appears to stimulate senescence, a different senescence pathwayis probably triggered compared to the senescence pathways induced by shading. High light-induced loss of chlorophyll is frequently due to photooxidation and may not be associated withother indicators of senescence such as protein loss (Biswal and Biswal 1984). One hypothesisfor light stimulated senescence is that tissues grown continually under high light have faster ratesof development and thus age more quickly than tissue grown under lower light (Biswal andBiswal 1984). In contrast, shade-induced senescence occurs because the leaf is no longer a netcarbon exporter.In addition to being able to trigger senescence, light levels can also affect the rate of leafyellowing once senescence has begun. Light exposure accelerated chlorophyll degradation inattached senescing Acer negundo leaves compared to attached senescing leaves kept in darkness.The absence ofany accompanying differences in protein or amino-nitrogen levels between light-exposed leaves and dark controls suggested that the role of light during chlorophyll degradationmay be primarily photochemical (Maunders and Brown 1983). Chlorophyll loss in senescingOryza sativa was also delayed under low irradiance due to slower degradation of the chlorophylla/b-bearing proteins ofLight-Harvesting Complex II (Okada et al. 1992). The effect of light on11the rate ofleafyellowing in senescing tissue has also been recorded for other species. However,in general, there are few studies comparing the rate of leafyellowing under different senescenceconditions.To separate the effects of light quality from light level it is necessary to examine the roleof phytochrome and other photoreceptors on senescence. Phytochrome is involved in bothshade-induced senescence and in light-mediated delay of senescence. In the former case, a redto far-red ratio of0.07, similar to that experienced by shaded leaves, accelerated chlorophyll andprotein loss in Glycine max leaves (Guiamet et al. 1989). In the latter case, senescence inmustard cotyledon was retarded by white and red, but not far-red light (Biswal and Biswal 1984).In Oryza sativa leaves chlorophyll degradation was also delayed by white and red light but notfar-red light (Okada et al. 1992). Further research will show which senescence triggers and’whichtissues are involved in the signal transduction pathway. Although different types of phytochromehave been described from etiolated and light-grown plants (Smith and Whitelam 1990), no onehas yet investigated if phytochrome present during leaf yellowing has unique characteristics orthe levels ofphytochrome during senescence.Because light affects the biochemistry of plants directly and indirectly, a clear picture ofthe role of light quality and levels during senescence will require more rigorous research,including more work on photoreceptors. A better understanding of the response of senescingtissue to light could have broad applications. For example, crop plants could be grown undermore dense conditions if strains were developed that could produce adequate grain yield despiteshaded conditions. Alternatively, high light levels in combination with drought or other stressconditions can severely retard both annual crop growth and seedlings planted for reforestation.12A more complete knowledge of the senescence pathway may help in developing strains that areless sensitive to such stress.2.2.2 PhotoperiodIn 1923 Garner and Allen suggested that autumn leaf senescence occurred as aconsequence of shorter photoperiod, a suggestion subsequently confirmed for many species.Extended photoperiod can delay leaf abscision in Acer, Aesculus, Ailanthus, Alnus, Caragana,Coiylus, Fraxinus, Juglans, Larix, Liriodendron, Pheiodendron, Pinus, Prunus, Pyrus, Rhus,Robinia, Salix, and Ulmus (Wareing 1956, Matzke 1936, Gevorkiantz and Roe 1935).Moreover, the delay in these plants was not caused by differences in timing of leaf emergence andoccurred at low light levels (Ivlatzke 1936). I should also note that declining temperatur&whichoften accompanies shorter photoperiod cannot fully explain the short photoperiod-inducedsenescence. A variety of authors for example have observed that leaf yellowing was delayed onbranches near streetlights where branches were exposed to natural temperatures (Osborne 1973,Olmsted 1951, Kramer 1936, Matzke 1936).In cases where long days delay senescence, the magnitude of the effect can be modifiedby other factors. In Acer saceharum, extended photoperiod can delay senescence for five monthsbut young leaves required longer to senesce than older ones (Olmsted 1951). There is also animportant interaction between photoperiod and temperature. Observations with Larixoccidentalis (Rosenthal, unpublished) show that long days can delay senescence for a few weeksiftemperature is declining. However, when warm temperatures are maintained, needles subjectedto extended photoperiod remain green for six to seven additional months (Rosenthal,13unpublished). Warm temperatures under short days have also been reported to delay leafsenescence of Acer, Betula, Liquidambar, Robinia, and Quercus (Vince-Prue 1975, Kramer1936).The fact that long days delay or inhibit senescence indicates that phytochrome may be alight receptor in this phenomenon. Leaf senescence of Cornus and Phaseolus, for example, wasdelayed under extended white light photoperiods but not under red light. This suggested that,in these species, the response to white light was due to increased irradiance rather thanphytochrome mediated changes (Tucker 1981). However, both red and white light delayedchlorophyll and protein loss in Cucumis but senescence occurred if far-red light was given(Tucker 1981). While these effects are of considerable interest, it is difficult to make conclusivestatements in this area because unfortunately the action of red and far-red light has not beentested in many species.2.2.3 WaterDrought can trigger rapid or gradual plant senescence. Rapid senescence is mediated byseveral factors including stomatal closure, leaf wilting and altered hormone levels. Additionaleffects ofwater stress include altered transpiration rates and protein transcription or translationchanges (Reid et al. 1991, Waring 1991). In contrast, gradual senescence occurs through a shiftin growth or maturation processes that shorten the life span of the plant. Changes during gradualsenescence include slowed root growth, altered source/sink relations and nutrient deficiçncies.Below I give examples of rapid and gradual senescence and also compare drought-inducedsenescence to age-related senescence. I also discuss the effects of excess water on senescence.14Rapid senescence in plants occurs under moderate to severe drought conditions. Manybiochemical pathways are affected and these are often similar to age-related senescence.Increased proteolysis, oxidative enzyme activities and lipid peroxidation are characteristic ofbothdrought and age-related senescence (Jones 1983). However, the drought-induced changes arenot necessarily the same as age-related senescence. For example, catalase activity increasedduring water stress inMedicago sativa (Irigoyen et al. 1992) although it decreased during age-related senescence in beet (Pistelli et al. 1993), hydrilla (Begam and Choudhuri 1992) andtobacco leaves (Dhindsa et al. 1981). Another difference between drought and ageing was notedby Majumdar et al. (1991) who found that changes in chlorophyll and chlorophyllase wereinversely related in Glycine max leaves during drought but not during ageing.The acceleration of senescence by water deprivation may be due to changing hormonelevels. For example, high levels of ABA can promote either drought or age-related senescence(Woolhouse 1983). This effect may be direct or indirect, since ABA-induced stomatal closurewould certainly limit carbon assimilation. As shown earlier, limited CO2 assimilation also canpromote senescence. Ethylene may also play a role; abscission of Gossypium hirsutumcotyledons was hastened by ethylene when drought stressed, although this effect was reversedby elevated CO2 levels which can inhibit ethylene action (Jordan et al. 1972). Ethylene levels alsoincreased in droughtedMedicago sativa, but decreased ethylene levels were observed under moresevere drought as plant water potential dropped below -1.6 MPa (Irigoyen et al. 1992). Increasesin ethylene are found in age-related senescing tissue (Brown et al. 1991) as well as in dro,ightedmaterial, though some ofthe reported increases may be a response to tissue excision (PhilosophHadas et al. 1991) rather than a response to senescence.15Under relatively mild drought, plant resistance to water shortage is increased (Larcher1987) and growth retarded, but senescence may not be triggered immediately. However, vigourand longevity ofthe plant are affected. For example, mild drought may accelerate reproductionrather than causing leaf or plant senescence directly (Aronson et al. 1992). The reproductivestate of the plant also affects its response to drought. In seedlings of Triticum aestivum andSorghum vulgare, leaf senescence occurred at the same rate in all varieties (Khanna-Chopra andSinha 1988). In contrast, exposure of mature plants to drought revealed that fertile varietiessenesced faster than sterile varieties. This suggests that the developmental shift to fruiting altersthe response of tissue to drought. For example, induction of leaf senescence by drought duringfruit maturation in Glycine max was hastened if the stress was imposed after mid-podfillcompared to before mid-podflll (Cortes and Sinclair 1986).Excess water can also cause plant senescence through several pathways. Soil floodingcausing anoxic conditions can lead to altered levels ofanaerobic shock proteins (Reid et al. 1991)and hormones. Examples ofhormone changes include increased leaf ethylene (Wang et al. 1990)and ABA levels (Larcher 1987) following excess water stress as well as decreased cytokinin andgibberellin levels (Morgan 1990). Related to the changes in hormones due to flooding arechanges in the nutrient status of the plant. These may be mostly caused by decreasedtranspiration. For example, tomatoes grown in high humidity formed yellow leaves due tocalcium and potassium deficiency (Adams 1991). More work is needed to establish if decreasedmineral nutrients and altered hormone levels can account for the senescence triggers. As ipterestin preserving and creating wetlands increases, it will be important to understand how species cansurvive in wet soils while other species are more sensitive to fluctuating water tables.162.2.4 TemperatureTemperature affects many metabolic pathways and usually occurs in conjunction withother environmental stress conditions. In addition, the rate and severity of temperature stress canresult in very different tissue reaction. While response varies according to whether a plant ischilling-sensitive or not, in general, low temperatures occurring over a period of time can leadto gradual senescence. In contrast, sudden cold exposure can bypass senescence completelyleading to organ abscision without de-greening. Finally, rather than inducing senescence, cooltemperatures can delay senescence. I discuss examples of these different responses totemperature and make comparisons to age-related senescence. I also point out that many of theresponses of plants to temperature stress are complex to isolate because of the difficulty ofholding constant other environmental conditions that are typically correlated with temperature.For example, changes in relative humidity that are correlated with temperature may altertemperature responses. In addition, in natural systems temperature stress often occurs inconjunction with excess light (Ottander et a!. 1993, Yordanov 1993, Oquist and Huner 1991,Oquist and Martin 1986) or invading pathogens (Hendry et al. 1987), both of which affectsenescence. Additionally, the influence of low temperature on physiological parameters alsodepends on the rate at which the temperature change occurs (Minorsky 1989), a feature that hasbeen overlooked in some studies.A gradual low temperature stress affects many metabolic pathways eventually leading tosenescence. In general, changes in thylakoid membrane function appear to be one of theearlierplant responses to low temperature stress (Yordanov 1993). In addition, low temperature candirectly affect enzyme kinetics rendering tissue more susceptible to degradation (Oquist and17Martin 1986). At the level ofthe whole plant, low temperature can affect transpiration rates andinduce senescence prematurely (Waring 1991). However, as with other environmental factors,the sensitivity of tissue to temperature is dependent on the developmental state of the plant. InEuonymus, low temperature induced chlorophyll loss and carbohydrate accumulation in matureplants but not in actively growing shoots (Creasy 1974).The ability oftemperature-stressed tissue of different ages to senesce suggests that theremay be differences between temperature-induced and age-related senescence. Indeed, chlorophyllloss in conifer needles during winter temperatures may be due to photooxidation. Additionally,membrane integrity and enzyme function may be more important in temperature-inducedsenescence than in age-related senescence. Although work has been done to examine chioroplastultrastructure during low temperature stress (Taylor and Craig 1971) as well as duringsenescence, the changes need to be analyzed in conjunction with other metabolic functions beforecomparisons can be made.Senescence appears to be slowed by low temperatures in both attached and detachedtissue. Molisch (1918) found that senescence associated yellowing was retarded in detachedleaves maintained at 5-13°C. The delay also occurred in attached leaves. For example, Pinuspumila needles have a longer life span when growing in low temperature environments (Kajimoto1993). In addition, leaf longevity in Larix lyallii is postulated to be lengthened in a coolenvironment (Worrall 1993). The slowed metabolic pathways during low temperatures whichdelay the onset of age-related senescence are in contrast to the faster maturation processassociated with excess light that accelerates senescence.18Rapid low temperature stress can prevent organ senescence altogether and lead to tissueabscision. Although plants can adapt to low temperatures, extreme or sudden temperature dropcan cause tissue death due to cellular freezing. Sudden changes also prevent adequate time forprotein synthesis or nutrient mobilization which could otherwise allow the plant to adapt to thelower temperatures. For example, canola seeds remain green if an early frost inhibits normal degreening (Johnson-Flanagen and McLachlan 1990). In autumn, ginkgo leaves subject to a coldnight undergo rapid green leaf abscision.The response oftissue to high temperature has been studied extensively (Yordanov 1993,Morgan 1990). Results typically suggest that within physiological range, high temperatureappears not to cause chlorophyll destruction in leaves (Hendry et al. 1987). Instead, membraneintegrity appears to be affected first (Yordanov 1993).2.2.5 Mineral nutritionMineral stress eventually leads to organ and plant senescence (Kelly and Davies 1988).However, the role ofmineral stress in plants is complicated by the degree of stress and the abilityof the tissue to respond to the stress. For example, Pinus monophylla needles can function aslong-term nutrient reserves because of their longevity (Everett and Thran 1992). Tn contrast,certain deciduous trees respond to mineral deficiency by increasing the amount of nutrientsreabsorbed before leaf abscission (Kozlowski 1991). Tn addition to species specific responses,there are also mineral specific responses. Once senescence begins, the pattern of chlorois andnecrosis varies depending on the mineral deficiency (Salisbury and Ross 1985). To elaborate on19these points, I first discuss the impact of soil nitrogen on plants and then other e!ementdeficiencies and their impact on senescence.Nitrogen levels can affect leafyellowing by influencing the rate and timing of senescence.Nitrogen deficiency usually leads to senescence though excess nitrogen does not necessarily delaythe process (Feller and Fischer 1994). For example, in the annuals Glycine mcix and Oryza sativaexcess nitrogen had no effect on senescence (Kelly and Davies 1988, Makino et al. 1984).However, in the perennialMalus domestica, autumn nitrogen fertilization delayed leaf senescence(Millard and Thomson 1989). The rate of senescence can also be affected by nitrogen levels.When Triticum aestivum was grown under nitrogen deficiency it did not senesce earlier thancontrols, but once senescence began it proceeded at a faster rate. Glutamate synthase waspresent during early and late stages of senescence in plants grown with adequate nitrogen.However, under limiting nitrogen the enzyme appeared to be present only in the early stages ofsenescence (Cincerova et al. 1991). This suggested that nitrogen recycling ended early in plantssubject to limited nitrogen. Regardless ofhow plants initially respond to lower nitrogen levels,such as by recycling Rubisco, the signal that a plant is mineral deficient is probably relayed toleaves by hormones. For example, in tobacco ammonium nitrate delayed leaf senescence in partby affecting cytokinin levels (Singh et al. 1992 and references therein).Deficiencies of phosphorous and potassium also trigger senescence, though not enoughresearch has been done on these elements to determine how they affect the onset and rate ofsenescence. Although phosphorus is essential for plant growth, its deficiency does not appearto be a cause for early age-related senescence. Neither excess nor deficient phosphorus levelsaffected the timing of leaf senescence in Glycine max (Crafts-Brandner 1992). Additionally,20electron transport in phosphorus deficient Helianthus annus and Zea mays was not limited (Jacoband Lawlor 1993) though under most senescence conditions electron transport is affected earlyin the process. Naturally occurring potassium deficiency in peatland soils causes conifer needleyellowing (Sarjala and Kaunisto 1993) but the process has not been sufficiently characterized todetermine how the senescence pathway may differ from other stress conditions leading to yellowneedles.Calcium is important in plants as a second messenger (Bowler and Chua 1994, Leshem1987) and it can both delay and promote leaf senescence. Although levels of calcium duringsenescence have not been examined frequently, calcium is undoubtedly involved in signaltransduction pathways which regulate certain aspects of senescence. For example, transientincreases in cytosolic calcium due to red light affects phytochrome (Shacklock et al. 1992) andmay in turn trigger senescence related responses. Calcium has also been shown to be involvedin light-mediated delay of senescence in detached corn leaves (Huang and Kao 1992). Calciumcan affect hormone levels such as methyl jasmonate (Chou and Kao 1992) and ethylene (Bowlerand Chua 1994), and these affect senescence. Studies examining the role of calcium duringsenescence have shown that senescence was delayed when calcium was added exogenously todetached Cucumis satEvus cotyledons (Ferguson et al. 1983), Avena saliva leaves (Dreier 1990)and Brassica oleracea leaf discs (Chéor et al. 1992). Although calcium may delay senescence,high calcium levels have been found in senescing autumn leaves, partly because calcium is notremobilized (Adams et al. 1990).Since magnesium is required for functional chlorophyll and as a cofactor for ATP, it is notunexpected that magnesium deficiency causes yellowing. Additionally, it is not surprising that21magnesium deficient tissue is sensitive to light. For example, Phaseolus vulgaris grown inmagnesium deficient soil showed chlorosis and this effect was delayed if plants were shaded(Cakmak and Marschner 1992). In addition, the response oftissue to magnesium deficiency doesnot appear to be sensitive to leaf age (Scott and Robson 1991).A few studies have examined deficiency of less common but equally essential mineralssuch as sulphur, iron and copper. Iron deficiency in leaves can cause an increase in chlorophylla/b and carotenoid to chlorophyll ratio (Young and Britton 1990) as well as decreasedphotosynthetic pigments. In addition, the xanthophyll cycle does not appear to function in irondeficient, senescing leaves (Abadia et a!. 1991). Sulphur starvation lead to Rubisco degradationin Lemna minor without leading to cell death (Ferreira and Teixeira 1992), providing aninteresting system in which to study senescence since the typical indicators do not declinesynchronously. Another such system is induction of a senescence pattern similar to the naturalone in Spirodela oligorrhiza and Triticum aestivum by high concentrations of cupric ions (Mehtaetal. 1992).Salinity stress in non-balophytes is becoming a greater concern as irrigation becomes morewide-spread. The effects of salinity on plants appears to involve changes in altered hormonelevels (Reid et al. 1991). Other physiological changes during salinity which lead to leafsenescence include lower assimilation rates and Rubisco activity, decreased transpiration, altereduptake ofother essential ions (Reid et a!. 1991) and appearance of distinct proteins (Reviron etal. 1992).All of the mineral deficiencies alter hormone levels, the effects of which have recentlybeen reviewed (Morgan 1990). Abscisic acid increased in nitrogen deficient plants; cytokinin22levels were affected by nitrogen (Singh et al. 1992) and phosphorus levels; and gibberellincontent was influenced by manganese deficiency (Morgan 1990). Hormone signals are alsoimportant under multiple nutrient deficiencies. For example, excess ammonium in conjunctionwith potassium deficiency induced ethylene production in Lycopersicon (Wang et al. 1990).There is some evidence that the promotion of senescence by nutrient deficiencies is due todecreased cytokinin production by the stressed roots. While it is true that cytokinin applicationscan often alleviate nutrient-induced symptoms of senescence, similar relief of symptoms of nitratestarvation followed direct supply of nitrate to leaves (Salama and Wareing 1979). This showsthat interference with production of root cytokinins is only one of the effects of nutrientdeficiency on foliar senescence (reviewed in Jackson 1993). The fact that nutrient deficiency cancause different hormone levels as well as distinct chiorosis patterns is a clear indication thatmultiple senescence pathways exist. Further research in this area could lead to potentiallyvaluable insight into how mineral levels at specific developmental points may delay or enhancecertain maturation processes leading to senescence.2.2.6 Carbon dioxideBoth elevated and low ambient carbon dioxide levels affect plant growth and the timingof senescence. The induction of senescence by CO2 levels below the compensation pointappeared to be due to photorespiration (Widholm and Ogren 1969). In contrast the results ofhigh CO2 levels are more complicated. I discuss the impact of elevated CQ on a number ofmetabolic processes below.23Research on the effect of increasing global CO2 levels on all phases of plant growth anddevelopment is very timely. Effects ofelevated CO2depend on the length of exposure and actualCO2 levels (Wullschleger et al. 1992, Kelly et al. 1991). The impact of elevated CO2 levels onphotosynthesis and growth also depends on the species and tissue age. Plant nutrient status caninfluence the response to elevated CO2 levels (Tissue et al. 1993) and the ability of the plants toobtain adequate nutrients may be in competition with soil microbes also influenced by rising CO2levels Diaz et al. 1993).The impact of elevated CO2 on senescence is likely to depend largely on its effect onphotosynthetic rates. Frequently, high CO2 levels stimulate growth. For example, elevated CO2can delay senescence by maintaining carbon assimilation rates in legumes although the effect isless dramatic in wheat (Lawrence and Mayne 1991). Higher assimilation rates were recorded inQuercus alba and Liriodendron tulipfera exposed for three years to elevated CO2 compared toambient CO2. In addition, the timing of leaf senescence was not affected b CO levels(Gunderson et al. 1993). However, in other experimental situations, the growth rate increase inresponse to high CO2 levels may actually level-off or even decline with long term exposure toelevated CO2 levels (Polle et al. 1993).Complicating the picture, elevated CO2 can also induce senescence by affecting pigmentlevels, protein function, and membrane integrity in a manner that differs from the onset of age-related senescence. Decreased chlorophyll and protein have been documented in CO2 exposedGossypium hirsutum leaf plastids (Chang 1975), Desmodium (Wuiff and Strain 1982), tropicaltrees (Oberbauer et al. 1985) and pine (Eloupis et al. 1988). However, the decline in chlorophylland protein do not necessarily occur concurrently (Satler and Thimann 1983) as they frequently24do in age-related senescence. For example, in Liriodendron tulzpfera and Quercus albachlorophyll declined after CO2 exposure while photosynthetic rates were maintained(Wullschleger et al. 1992).In addition to influencing photosynthetic rates, CO2 levels affect hormone levels thatinfluence senescence. For example, elevated CO2 levels may delay senescence in part byinhibiting ethylene synthesis (Lawrence and Mayne 1991) as well as other hormones. Thecomplexities of such interactions between elevated CO2 levels and other parameters has beenemphasized in a recent review (Bowes 1993). Yet despite a growing number of studies on boththe short and long-term elevated CU2, a clear picture of the manner in which CQ affectssenescence has not yet emerged. On the other hand, in a number of studies, CO2-inducedsenescence appears to affect the same metabolic pathways as are affected during age-relatedsenescence, which suggests that the two processes may be similar. However, this observationcould simply reflect the fact that comparisons may have been made of tissue well into senescencerather than shortly after the start of a CO2 or age-related response.2.2.7 PollutionAir- and water-borne pollutants are environmental facts of life for plants. The mostfrequently studied atmospheric pollutants include ozone, acid rain, sulphur dioxide, nitrogenoxide and hydrogen fluoride. In this section I compare atmospheric pollutant damage to agerelated senescence. All of these pollutants can induce senescence under appropriate con4itions.Decreased carbon assimilation and subsequent yellowing caused by pollutants dependson the species and age of tissue as well as the frequency of occurrence. Additionally, other25environmental factors such as light, humidity, and temperature can modify the response of plantsto pollutants (Darrall 1989). The effects of multiple factors on senescence are evident in novelforest decline in Europe, which seems to be caused by pollution in conjunction with soilacidification as well as climate change and pathogen stress (Guillemaut et al. 1992). Otherinteractions between environmental conditions have been documented. For example, high ozoneplus elevated CO2 levels Polle et al. 1993) or ozone plus water deficit (Chevone et al. 1990) areexamples ofadditive combinations of oxidative stress. But the effects of multiple stress are notalways additive. Droughted plants have lower stomatal conductance which has the benefit ofreducing the amount of ozone damage (Wieser and Havranek 1993).Despite the variety of air-borne chemicals, plants generally respond to pollutants with asimilar syndrome of altered metabolism. Decreased photosynthetic rates, lower Rubisco levels,stomatal closure, and increased respiration rates are common in pollutant stressed leaves (Bowleret al. 1992). Most pollutants can cause immediate damage to living cells through the formationof peroxides, hydroxyl radicals and singlet oxygen (Bowler et al. 1992).The senescence pathway due to pollution is distinct from age-related senescence in manycases. The pattern and the mechanism of pigment degradation is different between the twosenescence pathways. Pigment loss is probably an indirect effect of the pollutants since directeffects would require very high concentrations of the pollutants Rudiger and Schoch 1989).Chlorophyll loss due to pollution is slower and often incomplete compared with naturaldegradation (Brown et al. 1991). Carotenoids as well as chlorophyll are affected by pollutants.For example, Populus leaves exposed to combinations of pollutants did not show changes incarotenoids that were evident in naturally senescing leaves (Ballach et al. 1992). The pattern by26which pigment degradation occurs has in a few cases been reporeted to differ in pollutant-stressedtissue compared with age-related senescence. For example, in pine, ozone damage causedpatterns of leaf yellowing and abscission that were not seen in age-related senescence. Thesepatterns include chlorotic mottling of Pinus ponderosa needles (Addicott and Lyon 1973) andyellowing of the adaxial side of needles (Young and Britton 1990).Another difference between pigment degradation in pollutant-stressed tissue comparedto age-related senescence is the absence ofvisible pigment degradation even though senescenceis underway. Although there are examples of naturally senescing plants in which de-greeningdoes not occur (Bortlik et al. 1987), typically senescence is defined by chlorophyll loss.However, ozone has been found to cause leaf fall in the absence of visible foliar damage in oldMalus domestica leaves (Wiltshire et al. 1993).Additional differences between pollution-induced and age-related yellowing have beendocumented. Ozone-induced senescence occurred more rapidly and with less order than innaturally ageing leaves (Matyssek et al. 1991). Although carbon assimilation changes due toozone resembled autumn senescence in Betula pendula, nitrogen content decreased in autumnbut not in ozone-exposed leaves. Additionally, leaves senescing under different ozone regimesshowed decreased carbon assimilation, narrowed stomatal pores, increased air spaces inmesophyll cells and lowered water use efficiency. However, the latter two parameters were not27evident in autumn senescing leaves (Matyssek et al. 1991). Ozone may also act by affectingRubisco synthesis or Rubisco mRNA. For example, in potato, Rubisco mRNA decreased and thiswas accompanied by increased ethylene levels (Reddy et al. 1993).The response of hormones to pollutants may be a result of damage to normal growthpatterns and nutrient uptake. Yellowing ozone-damaged Picea abies needles contained higherlevels ofgibberellic acid than undamaged tissue. Those levels were similar to younger tissue andsuggested that damaged tissue had not matured normally (Schneider et al. 1991). Plant nutrientstatus is also affected by pollutants. In spruce, sulphur dioxide appeared to cause nutrientdeficiency in older leaves rather than direct damage (Kostner et al. 1990) though sulphur dioxidecan damage chioroplast ultrastructure (Gonzalez et al. 1993). Plant nutrient status can also beassessed through foliar leaching. Foliar leaching is known to occur in ozone and acid rainsubjected leaves, but the extent to which the pattern of mineral loss is similar to naturallysenescing leaves, in which elements are translocated into bark tissue, has not been documented.Other degradative pathways caused by atmospheric pollutants are similar to those innaturally senescing tissue. Exposure of Picea sitchensis to ozone for three years appeared tocause chiorosis by accelerating leafageing and perhaps altering source/sink relations, rather thanby soil nutrient deficiency, root damage, or increased foliar leaching (Lucas et a!. 1993). Anincrease in Dl, a rapidly turning-over protein important in repair mechanisms ofPhotosystem II,occurred in pollutant stressed spruce (Guilemaut et al. 1992) before changes in photosyntheticrates were evident LUtz et al. 1992). This response may be similar to the temporary increase inDl in naturally senescing tissue (Brown et al. 1991). Similarly, in Phaseolus vulgaris acid rain28appeared to induce changes in lipids, membranes, and chlorophyll that were comparable tochanges found in age-related senescing tissue (Chia et al. 1984).Perhaps more than for any single environmental condition discussed, the effects ofpollution on senescence arise from combinations ofnumerous immediate and long term impactson plants. However, whether examined individually or in combination, under appropriateconditions all ofthe different types ofenvironmental stress reviewed above can trigger prematureage-related senescence pathways or induce a senescence pathway that is distinct in metabolicprocesses from age-related senescence. Few studies however, have actually documented the rateofleafyellowing in senescing tissue under different environmental stress conditions or careflulycompared senescence pathways induced by different conditions. Similarities among mostconditions include dysfIrnction ofproteins leading to or caused by altered membrane integrity andpigment destruction. The pathways by which these occur, such as declining cytokinin levels dueto lower transpiration rates or decreased polyamine concentrations, may vary, though eventuallyproteolysis and free radical damage lead to irreversible tissue damage.2.3 CONCLUSIONSAlthough plants have a remarkable capacity to adapt to changing environments, we areonly just beginning to understand the extent to which environmental stress can be tolerated beforesenescence is induced. Senescence occurs at all levels of plant development, from cellularsenescence to organ and whole plant death. On the one hand, plants senescing under differentconditions display similar processes ofchlorophyll degradation, declining lipid and protein levels,and increased oxidative damage. Within this broad pattern, however, there appear to be29important differences in the process of senescence depending on the environmental signal, thetissue perceiving the signal, and the tissue that ultimately responds. Perhaps it is not surprisingthat there are differences in the senescence pathway caused by different environmental conditions.There is not even a consistent trend regarding two fhndamental pathways during age-relatedsenescence, that ofthe timing ofchlorophyll breakdown and Rubisco degradation. In some casesRubisco is reported to decline prior to chlorophyll Hensel et al. 1993, Mae et al. 1989) and inothers the decline occurs concomitantly (Crafts-Brandner et al. 1990, Martinoia et al. 1983).Moreover, although differences in both age and environmentally-induced senescence pathwaysexist, it is not always clear if the different pathways become similar at some point later in thesenescence process.These findings suggest that senescence in any cell or organ is the sum of an array ofsubprocesses occurring either simultaneously or sequentially. While it is possible that senescence-related genes are responsible for the initiation of senescence, it is also possible that multiplestress-induced genes, when expressed together, regulate the pathway of senescence. The formerhypothesis suggests that environmental conditions leading to tissue senescence induce a specificgene which then triggers a similar pathway in all tissue. It is speculated that this might controlthe organized, irreversible senescence process. In contrast, the latter hypothesis would suggestthat when different environmental conditions affect a plant simultaneously, several stress-inducedgenes are expressed and it is the combination of these stress-induced genes that lead to the onsetof senescence. Under such conditions one would expect to find numerous senescence pathwaysbecause of the many different combinations of genes induced by adverse conditions.30Progress has been made in identifying senescence related genes and in documentingdifferences in transcript levels as tissue senesces (Smart 1994). For example, distinct transcriptsappear in age-related senescence in Hordeum vulgare compared to stress-related senescence(Becker and Apel 1993). Additionally, it appears that mRNA levels during senescence can beaffected by cytokinin levels (Martin and Sabater 1989), ethylene levels (Davies and Grierson1989) as well as absence of light (Azumi and Watanabe 1991, Blank and McKeon 1991). Itwould be of interest to determine if the senescence-related genes appear under different stressconditions or ifthey are specific to the condition in which they were first reported. Additionally,it would be interesting to determine at which point such genes become expressed in thesenescence pathway -- in the onset of senescence or in organizing the mechanism of cellulardegradation once senescence has begun. Further work will additionally have to include detailedanalyses of the promoter regions of each gene and the factors influencing their expression.Species-specific responses will complicate analysis of the general roles of stress insenescence. For that reason, I suggest that stress-induced senescence be compared to age-relatedsenescence in the same tissue. It may also be usefbl to examine the effect of environmental stresson stay-green mutants (Nock et al. 1992, Ronning et al. 1991, Bortlik et al. 1987) since thesenescence pathway in these plants is somewhat simplified given that senescence processesassociated with chlorophyll degradation are absent. Mutants, such as tobacco plants transformedwith a gene for cytokinin (Smart et al. 1991), may also provide opportunities to test specifichypotheses. A fbrther research goal should be to develop a more precise definition of senescenceand suggest parameters that can serve as indicators of leaf senescence.31Understanding how environmental signals control senescence may be a key to increasingplant productivity. For example, late senescing Populus hybrids may have a longer growth seasonand hence increased productivity (Nelson et al. 1982). Increased demands on yield have meantthat productive species are planted in ever greater altitude and latitude ranges from their ource,as well as under more densely shaded conditions. The effects of environmental conditions on leaflongevity (Kikuzawa 1988, Begonia et al. 1987, Koike 1987) are also important in understandingthe response of tissue to senescence triggers. For these projects to be successful, we mustunderstand the impact ofthe environment on organ and whole plant senescence, especially in treespecies where pollutants can damage tissue before visible foliar yellowing. Additionally, a betterunderstanding ofthe senescence process will lead to a clearer picture of the stage in senescencewhen re-greening is still possible. Further studies on the effects of environmental conditions onplant and organ senescence will provide data to allow us to re-evaluate our current models ofage-related senescence as well as to understand the interaction of environmental stress conditionson plant senescence.32CHAPTER 3: ONSET OF FOLIAR SENESCENCE DURING AUTUMN: THE EFFECTSOF SOIL TEMPERATURE AND PHOTOPERIOD3.0 ABSTRACTPhotosynthetic and environmental measurements on one year old western larch (Larixoccidentalis Nutt.) seedlings during autumnal foliar senescence showed that chlorophyll andcarotenoid degradation preceded a decline in CO2assimilation. Chioroplast function was not lostuntil late in senescence. I also made measurements on outdoor seedlings senescing under warmersoil (average 3°C above naturally senescing seedlings) or extended 16 hour photoperiod. Theexperimental design allowed me to observe the effect of a single environmental variable onautumn senescence. The treatments delayed the onset of senescence and affected the sequenceof events during senescence. The differences between treatments in the order of decline ofdifferent photosynthetic measures is evidence for the existence ofdifferent environmental controlson the senescence process.3.1 INTRODUCTIONAutumn foliar colour change is triggered by both leaf-specific maturation processes andenvironmental conditions. The precise effects of individual environmental conditions on leaveshave proved difficult to identify because several of them, such as air temperature, photoperiodand rain fall, tend to change at the same time. In addition, some cellular components deteriorateduring senescence as a direct result of the environment while others respond to endogenousprocesses only indirectly affected by the environment. Such interactive effects make it even more. 33difficult to isolate the effect of the environment on events during senescence. Previous studieson autumn senescing tissue in deciduous trees examined pigment degradation (Sanger 1971,Goodwin 1958), chloroplast ultrastructure (Cunninghame et al. 1982), gas exchange (Matysseket al. 1991), chloroplast efficiency (Adams et al. 1990), respiratory enzymes (Dean et al. 1993),fluorescence-emission spectra (Lang and Lichtenthaler 1991), and plant hormone levels (Osborne1973). However, these studies did not compare their findings to changes in autumn weather.Other studies relating weather conditions to the timing of senescence were not accompanied bydetailed physiological measurements (Worrall 1993, Kozlowski et al. 1991, Addicott and Lyon1973).It is important to recognize that environmental conditions affect the onset of senescenceas well as the process of senescence, i.e. the sequence and rate of changes (Smart 1994). Onehypothesis is that autumn weather and cumulative stress imposed on leaves trigger senescence-related genes that coordinate a pre-programmed set of events. This leads to the prediction thatthe onset ofautumn foliar senescence will vary with environmental conditions but that senescenceprocesses will be relatively insensitive to autumn conditions. An alternate hypothesis is thatcomponents ofthe leaf are affected differently by autumn conditions, in which case both the onsetand process of senescence will vary in leaves that senesce under different autumn conditions. Forexample, in a year when air temperature declines before light levels, carbon reduction may declineprior to light harvesting capacity, whereas in a mild year, a series ofphytochrome mediated .eventscould begin senescence. The existence of different patterns of senescence has been documentedfor a number of systems (Pell and Dann 1991) but not previously for the interrelated eventsoccurring during autumn foliar senescence.34I collected data on the photosynthetic apparatus in one year old western larch (Larixoccidentalis Nutt.) to determine how the process of senescence was affected by different autumnconditions. Changes in pigments, photosynthesis and quantum yield were measured in treessenescing under outdoor autumn conditions and compared to the pattern of senescence inseedlings exposed to either warmer soil or longer photoperiod. This experimental design allowedme to observe the effect of a single environmental variable on autumn senescence. Resultsindicate that the onset ofdecline in chlorophyll was directly affected by the environment and thatthe decline in photosynthetic function was largely, but not entirely, the same in seedlingssenescing under different conditions.3.2 MATERIALS AND METHODS3.2.1 Experimental designOne year old western larch (Larix occidentalis Nutt.) seedlings, obtained from seed lot5266, Thompson Okanagan Dry Seed Planning Zone, British Columbia (elevation 1200m, 49 lat,119 long) were grown outdoors from March until August and watered regularly. In late August1992 (Julian date 240) potted trees were divided into treatment groups: naturally senescingcontrol seedlings (n=26), seedlings receiving warmer soil (n=34), and seedlings receivingextended photoperiod (n=3 1). Seedlings were placed in gravel-bottomed cold frames under atranslucent plastic roof. The frame for seedlings receiving warmer soil was heated with.cablesunder the gravel to raise soil temperature an average of 3°C above ambient during late autumn.Soil temperature probes were placed 3 cm under the soil in the 6 inch pots. The day length forseedlings receiving an extended photoperiod was increased to 16 hours with supplemental. 35fluorescent lights (100 j.tmol photons m2 s’). A CR10 datalogger (Campbell Scientific,Edmonton, Alberta) recorded air temperature, soil temperature and relative humidity. Lightintensity was measured using a Li-Cor quantum sensor (Li-Cor, Lincoln, NE).My choice ofa deciduous conifer allowed sequential harvesting of same age needles. On37 days over a 15 week period beginning Julian day 247 (September 4), needle samples (threeper treatment) each consisting of one needle from a side branch of each seedling were collectedrandomly from all treatments between 7:30 and 9:30 AM. This procedure yielded threeindependent and representative samples from each treatment group on each measurement day.The data presented in this chapter is based on samples collected during 1992. A similarexperiment was done during autumn 1991. However, the presence ofwintergreen needles in thesample of needles collected during late autumn made analysis of the 1991 data difficult.Wintergreen needles (needles which remain green and do not senesce) are often found on apicalportions ofbranches on young seedlings. During 1992, wintergreen needles were avoided by notsampling needles near the apical shoots where wintergreen needles occur. The data collectedduring 1991 were not used to characterize the onset of senescence because wintergreen needlesprevented me from estimating the onset ofdecline in pigments and photosynthesis. However, the1991 data were used to examine the rate of senescence (Chapter 6).3.2.2 Pigment and photosynthesis measurementsChlorophyll and carotenoids were determined after acetone extraction (Lichtenthaler andWellburn 1983). Carbon assimilation at 1000 .tmol photons m2 4 was measured at 20-25° Cusing a narrow leaf chamber of an ADC-LCA2 (Analytical Development Company, Hoddeston,36UK) infrared gas analyzer (IRGA). Leaf area of individual needles was determined from lengthand width measurements using a millimeter ruler, and adjusted for larch needle morphology(Vance and Running 1985). Stomatal conductance and internal CO2 concentrations wereobtained from the IRGA. Photosynthetic oxygen evolution was measured with a leaf disc oxygenelectrode (Hansatech, Kings Lynn, England) using a red light emitting diode (660nm). A dischaving a composite flat leaf surface area of 10 cm2 was made by placing thirty needles side byside, trimming the apex and base from the needles, and holding them together by four narrowpieces oftape. The tape was placed at the edge of the leaf disc to insure that gas exchange wasnot limited: stomata occurred mostly on the adaxial leaf surface. To maintain CO2 levels, the matin the leaf disc chamber contained 1.0 M NaHCO3(pH 9) and the chamber was flushed with 50mL U1 CO2from aNaHCO3solution (Walker 1987). Oxygen measurements were corrected forrespiration which was measured both after 10 mm in the dark and after 5-10 mm of light-saturating illumination. Respiration was assumed to remain constant as the actimc light increased.Quantum yield of oxygen evolution was calculated based on absorbed light. I used absorptionvalues of 0.84 for green and early senescing tissue and 0.65 for late senescing tissue (Adams etal. 1990) where early and late senescing tissue are defined in Section Chlorophyll a fluorescence measurementsFluorescence measurements were made on the adaxial side of needles using a pulsemodulated PAM 10 1/103 fluorometer equipped with a Schott lamp (H. Walz, Effeltrich,Germany and Schott, Mainz, Germany). The nomenclature suggested by Kooten and Snel (1990)was used. Needles were dark-adapted for 10 mm. The pulse intensity to achieve close to37saturation in a 20 mm dark adapted leafwas determined by measuring photochemical efficiency(Fv/Fm; Katajima and Butler 1975) for green October needles taken from trees grown in coldframes under 600 imol m2 s4 light (n=8 per light intensity): 1300 .imol n12 Si: 0.780 (0.023SE); 2000 j.imol: 0.781(0.022 SE); 4000 j.imol: 0.785 (0.020 SE); 7000 .tmo1: 0.782 (0.021 SE);13000 j.imol: 0.793 (0.013 SE). The method suggested by Markgraf and Berry (1990), whichestimates the “true” height of a saturating pulse, was used to correct for potential problems dueto sub-saturating pulses at higher actinic light levels. L waves (Larcher and Neuner 1989) werevisible at low light levels. Calculating photochemical quenching with or without L waves did notchange interpretation of the data.3.2.4 Determination of breakpointsI used piecewise linear regressions to determine the day on which photosyntheticparameters declined. For example, the level of chlorophyll was expressed as a function of Juliandate. The slope of the first line segment was set equal to zero, consistent with the assumptionthat chlorophyll had not yet begun to decline. The slope of the second line segment was allowedto differ from zero, consistent with the assumption that chlorophyll had begun to decline. Thus,the breakpoint date between the two line segments defined the start of the decline in chlorophyll.For each treatment group a series of piecewise linear regressions were estimated in which thebreakpoint varied from the first measurement day of the experiment to the second4o-lastmeasurement day. Repeating this procedure for all variables in each treatment group entailedestimating nearly 600 separate piecewise linear regressions (9 variables x 2 treatment groups x33 measurement days).38Results from the piecewise linear regressions were used to estimate the date of declineof chlorophyll. First, the sum-of-squared residuals from the piecewise linear regressions wereplotted against their corresponding breakpoints. The breakpoint associated with the lowest sumof squared residuals was chosen as the best estimate ofwhen chlorophyll began to decline. Thisprocedure was repeated for the other photosynthetic measures.3.3 RESULTS3.3.1 Environmental conditions and pigment lossThe difficulty in determining the environmental conditions that trigger autumn senescenceis that most often weather variables are correlated. During this experiment, temperature, lightlevels, day length, and relative humidity changed at approximately the same time during autumn.At the start ofthe experiment on Julian day 240 (27 August), the mean air temperature at needleharvest (Figure 3-la) was 23°C and it declined to an average of 12°C by day 285 (11 October).Soil temperature in control seedlings averaged 11.28±0.58°C (n=72, 95% confidence interval)from September to December. Light intensity, measured in the morning at needle harvest, fellfrom 600 to 200 j.imol m2 by day 285 (Figure 3-2a). Day length (AtmosphericEnvironmental Services, Vancouver BC) declined from 13.5 to 8.5 hours (Figure 3-2b) and theday length when chlorophyll began to decline was 11 hours. Relative humidity increased from60% in late summer to consistently above 90% after day 285 (Figure 3-2c). To determine theeffect of soil temperature and photoperiod on senescence, two treatment groups were comparedto control seedlings. In warm soil seedlings the temperature averaged 2.89±0.12°C (n=72, 95%confidence interval) above that of control seedlings (Figure 3-ib). Seedlings receiving an39Figure 3-1. Temperature at the experimental site during autumn, a) Air temperature (°C) atneedle harvest for seedlings senescing outdoors (Control, filled diamond), with heated soil (Warmsoil, shaded diamond), and with 16 hour photoperiod (Long day, open diamond) b) Minimumsoil temperature for warm soil seedlings (dashed line) averaged 3°C above naturally senescingseedlings and seedlings receiving extended photoperiod (solid line).4030 I I I I I(a)20•O•0 I I I I I I20 I IP (b)240 260 280 300 320 34027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec41Figure 3-2. Environmental site conditions during autumn, a) Photosynthetic photon flux density(jimol photons m2 s’) at needle harvest for seedlings senescing naturally (Control, filleddiamond), seedlings receiving heated soil (Warm soil, shaded diamond) and seedlings receivingsupplemental photoperiod (Long day, open diamond). b) Hours of day length during theexperiment c) Relative humidity at needle harvest for all seedlings.t) -CCRCRelativehumidity00CoaaaDaylength(hours)—————00OC—PFD(p.mol m2s)—I.JW.fl0%.00oaaaaaaaCC00000C0•øK%SKS46S.4,,43extended photoperiod received 16 hours at 100 j.tmol m2 s supplemental to ambient light levels.The lower light levels in the extended photoperiod group (Figure 3-2a) was due to the positionof the cold frames which gave the extended photoperiod seedlings more shade in the morning.Figure 3-3 shows the time-course of chlorophyll loss in the three treatment groups.Seedlings exposed to natural conditions and seedlings receiving warm soil treatment averaged276±26 mg chlorophyll m2 (95% confidence interval) and 256 ± 16, respectively, during earlyautumn while seedlings receiving an extended photoperiod averaged 294 ± 18 mg chlorophyll m2during the same period (Table 3-1). The date when chlorophyll started to decline (“sate ofdecline0)was determined using a piecewise linear regression assuming different dates of decline.The date at which the regression gave the lowest sum-of-squared residuals was chosen as thebreakpoint (Figure 3-4). The date ofchlorophyll decline determined in this way was day 285 forcontrol seedlings, day 295 for warm soil seedlings, and day 317 for extended photoperiodseedlings (Figure 3-3). Sampling was stopped when CO2 assimilation rates (see below) becamenegative. I ensured adequate material for late senescence samples by not collecting fromseedlings receiving extended photoperiod between days 320 and 338 (15 November and 3December).Dates of decline for total carotenoids (data not shown) were similar to those for totalchlorophyll: day 277 for control seedlings, day 291 for warm soil seedlings, and day 314 forextended photoperiod seedlings. Carotenoids declined from 300 to 100 mg m2 during thesampling period, and the chiorophyll/carotenoid ratio fell from 4 to 1.5 starting about 3 weeks44Figure 3-3. Chlorophyll and chlorophyll a/b ratio during autumn, a) Date of decline for seedlingssenescing under natural conditions (Control) was Julian day 285 (11 October) and day 306 forchlorophyll and chlorophyll a/b, respectively. b) Date of decline for seedlings senescing withheated soil (Warm soil) was day 295 (21 October) and day 314 for chlorophyll and chlorophylla/b, respectively. c) Date of decline for seedlings sénescing with supplemental 16 hourphotoperiod (Long day) was day 317 (12 November) for both chlorophyll and chlorophyll a/b.Each point is an average of3 measurements. Breakpoints were determined by a piecewise linearregression estimated for each measurement day (see text and Figure 3-4). I defined earlysenescence as the period when chlorophyll declined but the chlorophyll a/b ratio was maintainedwhile late senescence occurred when both chlorophyll and chlorophyll a/b declined.400 4 45100400300200C100Cclii)400300E 200100300200(a)OChla/b•mgChlm2 285 3060 I I I I I240Control32I250 260 270 280 290 300 310 320 330 340I çj I000• .(ThI I I2I.295 3i4o)I I240 250 260 270 280 290 300 310 320 330 340 350 -...- 4_’I I.. Warm(b)Soilo Chi a/b• mg Chi mOChla/b 0••rngChlni2 Long Day 3170240 250 260 2703210280 290 300 310 320 330 340 35027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec46Table 3-1. Chlorophyll and chlorophyll alb ratio in Larix occidentalis needles of differentsenescence stages (see Figure 3-3 for definition of green, early and late senescence). Values aremeans ± 95% confidence interval (sample size in parenthesis).StageGreen Early Senescence Late SenescenceChlorophyll (mg m2)Control 276±26 (43) 205 ±24 (24) 68 ± 17 (12)Warmsoil 256±16(58) 185±25(25) 56±28(4)Longday 294±18(86) 119±64(14)Chlorophyll a/b ratioControl 3.17±0.18(43) 3.12±0.19(29) 2.19±0.35(15)Warm soil 3.24 ± 0.11 (58) 3.29 ± 0.29 (27) 1.62 ± 0.78 (6)Long day 3.26 ± 0.09 (88) 2.06 ± 0.48 (15)47Figure 3-4. Determination ofdate of decline for chlorophyll, photosynthesis and chlorophyll a/bratio for naturally senescing seedlings during autumn. The sum-of-squared residuals from apiecewise linear regression was estimated for each measurement day. The date giving theminimum sum-of-squared residuals (arrows) was chosen as the breakpoint.48I I I0 Chlorophyll a/bmg Chlorophyll m2-. CO2 assimilationC,)C)C)I)240 250 260 270 280 290 300 310 320 33027Aug 16 Sep 6Oct 260ct l5Nov49after the date of decline in control and warm soil seedlings, and about 1 week later in extendedphotoperiod seedlings. The chlorophyll a/b ratio started to decrease 3 weeks after pigmentcontent in control and warm soil seedlings (Figure 3-3). It decreased from 3.2 to approximately2 at the end of sampling (Table 3-1). Dates of decline estimated by sum-of-squared residuals asin Figure 3-4 (control shown) were day 306 and day 314 for control and warm soil seedlingsrespectively. A date could not be determined for extended photoperiod seedlings becausesampling had to be suspended during the period when the chlorophyll a/b ratio would have startedto decline. Based on the different dates ofdecline for chlorophyll and chlorophyll a/b ratio, I havedefined early senescence as the period when chlorophyll is declining but chlorophyll a/b ratio isnot changing (Table 3-1). This stage occurred between days 285 and 306 (11 October - 1November) in control seedlings and between days 295 and 314 (21 October - 9 November) inseedlings receiving warm soil treatment. The most striking difference occurred in seedlingsreceiving extended photoperiod in which early senescence did not begin until after day 317 (12November). Late senescence was defined as the time after chlorophyll a/b ratio started to decline(Figure 3-3, Table 3-1).3.3.2 Gas ExchangeCarbon dioxide assimilation per unit area declined approximately a week after chlorophyllcontent. The decline began on day 296 in control seedlings (Figure 3-5a), on day 301 in seedlingsreceiving warm soil (Figure 3-6a), and after day 317 in seedlings receiving extended photoperiod.Stomatal conductance (Figure 3-7) in all seedlings decreased from 300 to less than 50 mmol m2s beginning on day 303 (29 October). This suggested that the decreased stomatal conductance50Figure 3-5. Photosynthetic parameters for seedlings senescing under outdoor conditions duringautumn. a) CO2 assimilation at light saturation (1000 j.tmol photons m2s1) declined on day 296(22 October) and oxygen quantum yield (‘)O) declined on day 310. b) Photochemical efficiency(FvfFm) of dark adapted needles (filled circles) declined on day 308. PSII quantum yield (cI,Genty et al. 1989) at actinic light levels of 25 (shaded circles), 100 (shaded squares) and 250i.tmol m2 s4 (open circles). Arrows show the date of decline in chlorophyll and chlorophyll a/b.1412.100.9u. 0.4-4.20.200‘I-I00. Ch1aj’bI I I I I: Q0 (a)0000:-•.•.. 00 Oxygen quantum yield• CO2 assimilation 296I I I I 0.00240 250 260 270 280 290 300 310 320 330I I I I I 0.9(b)- 0.6- 0.5- 0.4- 0.3- 0.2I I I I I I- 0.1240 250 260 270 280 290 300 310 320 33027Aug 16 Sep 6 Oct 26 Oct 15 Nov52Figure 3-6. Photosynthetic parameters for seedlings senescing under warm soil during autumn.a) CO2 assimilation at light saturation (1000 j.tmol photons m2 S’) declined on day 301 andoxygen quantum yield ((1)0,) declined on day 306. b) Photochemical efficiency (FvfFm) of darkadapted needles (filled circles) declined on day 306. PSII quantum yield (cD11) at actinic lightlevels of 25 (shaded circle), 100 (shaded square) and 250 pmol m2 s4 (open circles). Arrowsshow the date of decline of chlorophyll and chlorophyll a/b.53ChhChia9 a/b:(a).. .0 .-.0 Oxygen quantum yield 301 -• CO2 assimilation 306I I I I141210864200.>40.4-’—IE[14.(44‘4-U010240 250 260 270 280 290 300 310 320 330.e(bY• .F%•••Thcø °cPI I I I I3060. ‘.0.20.1;0.0240 250 260 270 280 290 300 310 320 33027 Aug 16 Sep 6 Oct 26 Oct 15 Nov54Figure 3-7. Stomatal conductance (mmol m2s1) during autumn declined in naturally senescingseedlings (filled diamond) on day 303 (29 October), and on day 305 in warm soil (shadeddiamond) and extended photoperiod seedlings (open diamond).55I I I I I I--. .•••• 14: .% *I I I I305:600500400300‘ 2000io:240 250 260 270 280 290 300 310 320 330 340 35027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec56was not affected by photoperiod or small increases in soil temperature. The 3-5% increase inratio of intercellular to ambient CO2concentration in late compared to early autumn tissue (Table3-2) was similar to other senescing leaves (Thomas et al. 1991).Photosynthetic oxygen exchange was measured with a leaf disc oxygen electrode. Darkrespiration was constant during autumn and averaged 0.5 j.tmol 02 m2 s when needles were preilluminated at 100 jimol m2 s4 and then dark-adapted for 10 mi When dark respiration wasmeasured immediately after illumination at 470 .tmol m2 s4, the rate averaged 1.5 jimol 02 rn2s’, which is consistent with reports (Walker 1992) that respiration rates are greater when the leafhas been previously exposed to high light intensity. Oxygen evolution per unit leaf area declinedfrom 18 to 3 .imol 02 m2 s’. Oxygen evolution per leaf area as a function of light levels showedlower maximum rates in senescing tissue (Figure 3-8). The date of decline in oxygen evolution(at 470 imol m2 s’), as determined from piecewise linear regressions, was day 298 for controlseedlings, day 295 for warm soil seedlings, and day 322 for extended photoperiod seedlings.These values are quite close to the dates of decline when photosynthesis was measured by carbonassimilation: days 296, 301 and 317, respectively, for control, warm soil and extendedphotoperiod seedlings.3.3.3 Chlorophyll a fluorescence and quantum yieldI measured quantum yield in green and senescing tissue to establish when the decline inphotosynthesis during autumn was due to a decline in chioroplast efficiency. Estimates ofquantum yield using oxygen evolution at light-limiting levels may be unreliable because darkrespiration rates at changing light intensities must be estimated (Walker 1992) and differences in57Table 3-2. Internal : ambient CO2 concentration in Larix occidentalis needles of differentsenescence stages (see Figure 3-3 for definition of green, early and late senescence). Ratios aremeans ±95% confidence interval (sample size in parenthesis).StageGreen Early Senescence Late senescenceControl 0.847±0.032 (42) 0.868±0.010 (26) 0.892±0.032(15)Warm soil 0.849±0.024 (55) 0.856±0.008 (27) 0.875±0.093 (6)Long day 0.846±0.014 (85) 0.883±0.03 9 (14)58Figure 3-8. Oxygen evolution per leaf area (20°C, 50 mL T} C02, 660nm light) at differentirradiances for naturally senescing seedlings in different phases of senescence: green (filled circle),early senescence (shaded circle) and late senescence (open circle). Senescence stages defined inFigure 3-3 and in text. Calculations for late senescing tissue were corrected for absorbance byassuming yellow leaves absorb 33% less light compared with green tissue (0.65 of incident lightcompared with 0.84 of incident light absorbed by green tissue; Adams et al. 1990). Data weresimilar for seedlings senescing under warmer soil or extended photoperiod. Error bars are 95%confidence interval and sample size per light intensity was 43 for green tissue, 29 for earlysenescing tissue and 15 for late senescing tissue.5915 I1o— oE ‘o e0E51I I0 100 200 300 400 500Absorbed PFD (tmo1 m2s1)60leaf absorbance between green and yellow tissue considered (Oquist and Chow 1992). Thus, Ialso determined quantum yield using chlorophyll a fluorescence data. Photochemical efficiencyofdark adapted tissue (Fv/Fm), as measured by a pulse modulated fluorometer, decreased from0.782±0.004 (95% confidence interval, n=139) in green needles to 0.5 in senescent needles(Figure 3-5b). This was due to a decline in both Fo and Fv (data not shown), and occutred byday 308 (3 November) and day 306 in seedlings senescing under outdoor conditions and warmsoil seedlings, respectively (Figure 3-5b, 3-6b). The changes occurred after day 317 in seedlingsreceiving extended photoperiod.Quantum yield of oxygen evolution per leaf area (4)02), determined from oxygenevolution rates at light-limiting levels (Figure 3-5, 3-6), declined in late senescing tissue at thesame time as the decline in quantum yield of PS11 (CI),Genty 1989). Quantum yield declinedearlier than the chlorophyll a/b ratio in warm soil seedlings but at the same time in controlseedlings. The earlier decline in quantum yield in warm soil seedlings may indicate that theformer were mildly drought stressed. Quantum yield ofPSII was approximately linearly relatedto 4)02 (Figure 3-9) and the slope of the relationship was similar when differences in leafabsorbance were accounted for by assuming that green leaves absorb 0.84 of incident light butyellow leaves absorb only 0.65 (Adams 1990).Photochemical quenching (qQ) was measured to determine if the proportion of oxidizedQa was rate limiting during photosynthesis and total non-photochemical quenching (qN) wascalculated to provide evidence of changes in energy dissipation mechanisms during senescence61Figure 3-9. Quantum yield ofPSII(1) calculated as in Genty et al. 1989, compared to oxygenquantum yield (GO,) calculated at light limiting levels (Walker 1992): green tissue (filled circle),early senescing tissue (shaded circle), late senescing tissue (open circle) and late senescing tissuecorrected for absorbance (open square). Calculations for late senescing tissue were corrected forabsorbance by assuming yellow leaves absorb 33% less light compared with green tissue (0.65of incident light compared with 0.84 of incident light absorbed by green tissue; Adams et al.1990). Sample size per data point for green tissue = 43, early senescing tissue = 29, and latesenescing tissue = 15. Error bars are 95% confidence interval. Senescence stages as defined inFigure 3-3.62I I I. 0.63 -‘ 0.56 T.O.49-T T0.42- ¶iE. .0.350.28-O.21- dE0.14-—— 0.07 -C,,—( 0.00 I I I0.00 0.02 0.04 0.06 0.08 0.10 0.12Oxygen quantum yield (F02)63(Schreiber et al. 1986). Data were grouped into green, early senescing and late senescing phasesas defined by chlorophyll loss (Figure 3-3) in order to plot qQ and qN as a function of lightintensity. Photochemical quenching was lower in late senescing tissue compared with greentissue for all seedlings at light levels greater than 25 i.tmol m2 s’ (Figure 3-lOa). A slower rateof Qa reoxidation during late senescence could cause the lower photochemical quenching. Nonphotochemical quenching changed only slightly between green and senescing tissue, for all lightintensities, in control (Figure 3-lob) and warm soil seedlings (data not shown). This suggeststhat in general, for a given amount of absorbed light, green and yellow leaves both dissipateexcess energy equally. In seedlings receiving extended photoperiod, mean values for qN werelower during late senescence compared with green tissue at all light levels. This suggests that atleast one component may be decreasing among the different energy dissipation mechanisms ofqN.3.4 DISCUSSIONI examined the onset ofdecline in photosynthetic components during naturally senescingautumn leaves and evaluated how this decline was affected by environmental conditions. Aprincipal finding was that CO2 assimilation declined after a decline in chlorophyll but before adecline in quantum yield. I then compared this process of senescence to that in seedlings exposedto different environmental conditions. Extended photoperiod and warmer soil temperaturedelayed the onset of senescence compared to naturally senescing autumn leaves and, to a smallextent, affected the sequence of decline in the photosynthetic parameters measured. The datawere easier to interpret if I defined early senescence as the period when chlorophyll levels64Figure 3-10. Chlorophyll a fluorescence quenching as a function of absorbed PFD (pmol m2s1).a) Photochemical quenching (qQ) in naturally senescing seedlings. Seedlings receiving warm soiland extended photoperiod showed a similar pattern. b) Non-photochemical quenching (qN) fornaturally senescing seedlings and Long day seedlings (insert). Seedlings receiving warm soilshowed a similar pattern to control seedlings. Green (filled circle), early (shaded circle) and latesenescing (open circle) tissue as defined in Figure 3-3. Error bars are the 95% confidenceinterval, sample size per data point for green tissue=43, early senescing tissue=29 and latesenescing tissue=1 5.651.0 I • IControl (a)0.80.6-++.0.4 -0.2 -001.0 I I IControl (b)0.8 I0.6-Z Long p’y.0.4 .0.2+0.0 I I0 100 200 300 400 500Absorbed PFD66declined but the chlorophyll a/b ratio was maintained and late senescence as the period when bothdecreased (Figure 3-3). Note that these definitions do not describe senescence in all species. Forexample, autumn yellowing does not occur in Alnus glutinosa (Bortlik et al. 1987) aid thechlorophyll a/b ratio appears to increase in Populus (Dean et al. 1993). I first describe thesenescence process in naturally senescing seedlings and then compare the process to seedlingssenescing in modified environments.CO2 assimilation per leaf area, measured at light-saturating levels, declined 11 days afterpigments started to decrease in naturally senescing larch seedlings (Figure 3_5)1• In contrast,quantum yield, measured from either 02 evolution at light-limiting photosynthetic flux density(660nm) or chlorophyll fluorescence, declined two weeks later (Figure 3-5, 3-6), 25 days afterthe decline in chlorophyll. Similar results were found in autumn Platanus leaves (Adams et al.1990) and suggests that although carbon fixation ability declined during early senescence, thenumber ofphotons required per electron transported remained constant until late in senescence.The presence offunctioning membranes late into senescence suggests that there was an organizeddegradation process (Adams et al. 1990). In addition, the concurrent decline in quantum yieldand chlorophyll a/b is evidence for final membrane deterioration and disassembly of the remainingpigment-protein complexes.The relationship between quantum yield ofPSll ((1)) and oxygen evolution ((1)02) has notpreviously been examined in senescing tissue, although c1), and (1)02 are well correlated in speciesgrowing under various conditions (Edwards and Baker 1993, Oquist and Chow 1992). My‘Similar results were obtained when photosynthesis was measured by oxygen evolutkrn. Inparticular, photosynthetic oxygen evolution declined 9 days after the decline in chlorophyll.67results show that ‘D,was a good predictor of (D02 in senescing tissue (Figure 3-9), provided thatcorrections were made for differences in leaf absorbance (Oquist and Chow 1992, Adams et al.1990). Other workers compared (D, to cDQ in tissue with low levels of light harvestingcomplexes. Oquist and Chow (1992) found a different ratio between cDO2 and cIJ in thechlorophyll b less barley mutant and the wild-type, in contrast to my observations where the slopebetween (DO2 and CD1 in senescing and green tissue was similar. I suggest that this may be dueto the absence of chlorophyll a/b light-harvesting complexes in the mutant.PSII quantum yield (calculated as the product of photochemical quenching [qQ] andphotochemical efficiency [Fv’/Fm]) declined during late senescence because of both lowerphotochemical quenching (Figure 3-lOa) and photochemical efficiency under different light levels(data not shown). The decline in qQ (Figure 3-lOa), a measure ofthe proportion of oxidized Qa,indicated that a greater portion of PS11 reaction centers were reduced in yellow than in greentissue. This is consistent with a slow rate of reoxidation of Qa in senescing leaves (Jenkins et al.1981), possibly caused by declining fbnction or mobility of cytochrome b6/f complex (Ben-Davidet al. 1983, Holloway et al 1983). Photochemical quenching was not lower in yellow comparedto green tissue at light levels below 25 jimol m2 s’, suggesting that the cytochrome complextransported electrons efficiently from PS11 to PSI when light was limiting carbon assimilation.In summary, photosynthetic membranes of early and late senescing larch needlesfunctioned differently. Leaves retained high photosynthetic rates, (DO2 and photochemicalefficiency (Fv/Fm) during initial loss ofchlorophyll and carotenoids. I infer from the unchangingchlorophyll a/b ratio that chlorophyll a and chlorophyll a/b containing-complexes degraded at thesame rate. Not until late senescence, when chlorophyll content had declined by more than 60%,68did chioroplasts show losses of function, as indicated by a decline in O2, (I), Fv/Fm andchlorophyll a/b ratio. The decline in the chlorophyll a/b ratio and the chlorophyll to carotenoidratio (chlorophyll a+b/xanthophyll+-carotene) suggested preferential loss of chlorophyll a-containing proteins closely associated with the reaction centers over loss of light harvestingproteins (Lichtenthaler 1987). Taken together, the data suggest that during early senescencesome photosynthetic membranes were completely degraded, while the remaining membranes werefully functional. This would occur if the initial decrease in photosynthesis of senescing tissue wasdue to a decline in the number of chloroplasts (Kura-Hotta et al. 1990, Camp et al. 1982).I can also draw inferences about the weather conditions that caused the photosyntheticdecline. Triggers for a particular pattern of leaf senescence are complex because of interactinginfluences of the environment. In addition, there are lags between the weather change and thephysiological responses. During this experiment, the decline in chlorophyll occurred in controlseedlings when average air temperature dropped from 23°C to 12°C (Figure 3-la), light levels haddropped to below 350 .tmol m2 s’ (Figure 3-2a), day length was less than 11 hours (Figure 3-2b),and relative humidity was above 85% (Figure 3-2c). To isolate the effect of a single weathervariable on senescence, I compared senescence in naturally senescing seedlings to seedlings whereone environmental condition was different compared to naturally senescing seedlings.Senescence began later in warm soil and extended photoperiod seedlings than seedlingssenescing under natural conditions. The 3°C increase in soil temperature (Figure 3-lb) and thelonger photoperiod was enough to delay loss of chlorophyll (Figure 3-3) by 10 and 32 days,respectively, compared with control seedlings. Because most of the changes in thephotosynthetic components were delayed in warm soil seedlings I conclude warm soils affected69the onset of senescence. I do not have a measure of significance for the delay because the datawere collected for one season only. Delayed senescence due to warmer soil is consistent with thefact that slightly colder soils cause lower photosynthetic rates in larch (Häsler 1985).An extended photoperiod during autumn has been reported to delay senescence indeciduous trees growing near street lights (Olmsted 1951, Matzke 1936). These observationsdid not include temperature measurement, so it is unclear whether the observed extendedgreening was the result of extended photoperiod, warmer air temperature, or a combination ofboth. In contrast, in my experiment, the delayed de-greening was not due to air temperaturewhich was the same for all seedling groups - control, warm soil and extended photoperiod (Figure3-la). Hence the longer green period in seedlings receiving extended day length reflected onlya longer photoperiod.There are several possible explanations for the delayed yellowing caused by extendedphotoperiod. Chlorophyll degraded faster when far-red light was the last light received by theplant (Okada 1992). This would occur naturally during dusk but not in the white light used forextending the photoperiod in this experiment. A second explanation for the delayed de-greeningis that the longer photoperiod at 100 j.tmol photons m2 s increased total irradiance hours forcarbon assimilation so that the onset of senescence was delayed because the leaf remained acarbohydrate source. In this case, yellowing would begin as other environmental conditionswould trigger senescence.In addition to delaying the onset of senescence, warm soil and extended photoperiod hada small impact on the extent of decline of senescence-related events. For example, in latesenescing tissue, non-photochemical quenching (qN) at high light was lower in seedlings reâeiving70longer photoperiods (Figure 3-lob, insert) than in control seedlings (Figure 3-lOb). The lowerqN during the delayed yellowing suggests that at higher light, energy was not transferred awayfrom PS11 (Horton and Bowyer 1990). Because Fv/Fm during late senescence was no kwer inseedlings receiving extended photoperiod compared to control seedlings, I infer that the lowerqN did not lead to photoinhibition.Another difference between senescence in the treatment groups was found in thestomates. Conductance decreased two weeks before the decline in photosynthesis in extendedphotoperiod seedlings, concurrently in warm soil seedlings, and a week after the decline in CO2assimilation in control seedlings. The decline in stomatal conductance may have been due to lightdeclining below 400 j.tmol m2 s1 (Jones 1983) or the increase in relative humidity from 70% toover 90%. These results suggest that conductance was not as sensitive to soil temperature andday length as it was to light intensity or relative humidity.I conclude that autumn foliar senescence in western larch needles is similar to that inangiosperms. it is characterized by an initial decline in some photosynthetic function. Thispattern exists regardless of the timing of senescence and the environmental conditions duringsenescence. There were differences in the extent of change in the photosynthetic apparatus ofseedlings grown in slightly warmer soils or under extended photoperiod. The differences providepreliminary evidence for different environmental controls of senescence in the photosyntheticapparatus.71CHAPTER 4: FURThER CHARACTERIZATION OF THE ONSET OF FOLIARSENESCENCE DURING AUTUMN: THE TIMING OF RTJBISCO DEGRADATION4.0 ABSTRACTRibulose- 1 ,5-bisphosphate carboxylase/oxygenase (Rubisco) levels may limit CO2assimilation during autumn foliar senescence. The amount of Rubisco large subunit (LSU) inautumn larch needles began to decline at the same time as carbon assimilation per leaf area,suggesting that Rubisco limited photosynthetic rates during senescence. Immunoblots showedfragments ofLSU as early as 10 days before carbon assimilation declined concurrent with theonset of decline in chlorophyll levels. These fragments were not apparent when leupeptin wasadded to the grinding buffer. These results showed that there was a developmental change in theleaf during senescence allowing Rubisco to become susceptible in vitro to a cysteine protease.I have not determined if this is the same protease responsible for degrading Rubisco in vivo.4.1 INTRODUCTIONAlthough protein degradation is sometimes correlated with increased protease levelsduring leaf senescence, it is more often due to increased activity of existing proteases (Huffaker1990). Proteolytic activity can be enhanced by changes in compartmentation of proteases fromsubstrates, identification of proteins for degradation by ubiquitin, or changes in pH and ATPlevels that can activate proteases (Dalling and Nettleton 1986). Proteolytic activity can also beenhanced by free radicals that alter protein conformation and make proteins more susceptible to72proteases. The difficulty in identil3,ing many ofthese pathways is that in vitro proteolytic activityis often not easily confirmed in vivo (Huffaker 1990).Degradation of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) has beenstudied by a number of researchers. Rubisco declines during senescence because of a decreasein synthesis and an increase in degradation. Rubisco synthesis is low in mature leaves (Laurière1983) and the decline in Rubisco large subunits (LSU) in older leaves occurred concurrently witha decrease in mRNA (Jiang et al. 1993).Increased degradation ofRubisco to amino acids that may be transported out of the leavescan occur by several pathways, including a mechanism in which Rubisco is altered by ubiquitinprior to degradation by proteases (Veierskov and Fergeson 1991). The redox state of cysteinemay also be involved in the sensitivity of Rubisco to oxidative inactivation (Garcia-Ferris andMoreno 1993, Mehta et al. 1992). On the other hand, the isoelectric point ofRubisco did notchange prior to degradation in soy-bean (Crafts-Brandner et a!. 1990), suggesting that there maynot always be conformational changes that lead to increased susceptibility to proteases.Regardless of how Rubisco is targeted for degradation, thiol proteases may be involved in thebreakdown (Hensel et a!. 1993, Mae et a!. 1989, Feller 1987).Prior to this study, no work had been done on the degradative events controlling Rubiscoduring autumn leaf senescence. I examined total leaf protein from autumn senescing westernlarch needles by SDS polyacrylamide electrophoresis and showed that the amount of Rubiscodeclined approximately at the same time as the decline in CO2 assimilation. Fragments ofLSU,identified on immunoblots, appeared at the same time as chlorophyll began to decline but prior73to the decline in carbon assimilation. These LSU fragments were not observed when leupeptinwas in the isolation buffer but were seen when N-ethylmaleimide (NEM) was present.4.2 MATERIALS AN]) METHODSWestern larch seedlings were grown in outdoor cold frames under natural conditions aspreviously described (Chapter 3). A sample of needles were harvested every few days, quick-frozen in liquid N2, and stored at -20°C until analysis. Before analyzing these needles, preliminaryexperiments were conducted on both green and yellowing larch needles to establish the preferredbuffer and protease inhibitor mixture. Needles were ground on ice in lOx volume ofbuffer perfresh weight and spun at 5000 g for 5 miii to pellet debris. A modified Laemmli buffer (Laemmli1970) was used containing 0.0625 M Tris pH 6.8, 10% glycerol and 2% SDS. In addition, thebuffer contained several protease inhibitors. These inhibitors included 10 mMethylenediaminetetraacetic acid (EDTA) and 20 mlvi diethyithiocarbamate (DIECA) as cholators;lOniM €-aminocaproic acid as a serine protease inhibitor; 0.5% soluble polyvinylpolypyrrolidone(PVP) as an anti-phenol (Gray 1989); and 20 M leupeptin and 2.5 mlvi N-ethylmaleimide(NEM) as cysteine protease inhibitors (Mae et al. 1989). Phenylmethylsulfonyl fluoride did notappear to affect Rubisco degradation and was not used after initial trials.SDS PAGE gels were run following the procedure of Laemmli (1970). Betamercaptoethanol was added before the sample was heated and loaded onto the gels. The amountofLSU was estimated from densitometric scans of SDS PAGE gels stained in Commassie BlueR. The dates of decline of CO2 assimilation were determined as previously described (Figure 3-4).74For western blots, mini gels (12% acrylamide) were transferred for 1 hour at 1 OOV in 25mM Tris and 192 mM glycine according to the Bio-Rad (Richmond, CA) manual. Gels wereblotted for 30 mm in a blocking buffer (3% fish skin gelatin in PBS with 0.01% Na Azide), andprobed with primary antibody for 1 hour, followed by three 5 mm washes in PBS with 0.05%Tween and a final wash ofPBS without Tween. Detection was achieved after 1 hour reactionwith goat anti rabbit alicaline phosphate (BRL 1: 1500 in 3% fish skin gelatin) followed by three5 mm washes in PBS containing 0.05% Tween and a final wash of 50 mM Tris pH 8.0. Colorreagent (0.02g Fast Red TR Salt and 0.OlgNapthol AS IvlX phosphoric acid per 10 ml of 50mMTris pH 8.0, Sigma, St. Louis, MO) was then added until intensity was achieved (20 mm).Antibody to spinach Rubisco was a gift from Dr. Terry Crawford at the University of BritishColumbia.4.3 RESULTS AND DISCUSSIONThe amount of Rubisco in autumn senescing larch declined at approximately the sametime as the decline in CO2 assimilation per unit leaf area (Figure 4-1), suggesting that Rubiscolevels were limiting carbon assimilation. The amount ofLSU, as determined from densitometricscans of SDS-PAGE gels stained with Commassie blue, declined by day 296 (October 22) inseedlings senescing naturally and by day 301 in seedlings senescing in soils that were 3°C warmer(Figure 4-1).Trnmunoblots were made to ensure that the decline in Rubisco was not an artifact of thegrinding buffer. An initial immunoblot detected low molecular weight degradation fragments ofRubisco. Fragments of about 24 kDa in size were first detected on day 284 (October 10) in75Figure 4-1. Carbon assimilation and Rubisco amounts during autumn in naturally senescing Larixneedles, a) Control and b) warm soil seedlings. Carbon assimilation rates (PN) and large subunitof Rubisco LSU) are open and filled circles, respectively. Amount of LSU determined bydensiotemetric scan of SDS PAGE gels. Arrows for CHL (mg chlorophyll m2), carbonassimilation rates (PN) and chlorophyll a/b ratio (AB) indicate the date of decline of thesemeasures (as described in Section 3.2.4).AmountofRubiscolargesubunitAmountofRubiscolargesubunit____________________________0II0Cl0.&IJ10ot000Cl••0—N•00Qo000•0pp0_______________________________________000 ClIfl0ISZ:iuzODTowliTWODjotuil77naturally senescing larch needles (Figure 4-2). This occurred at the same time as the decline inchlorophyll, and 10 days prior to the decline in CO2 assimilation. Additional analysis withprotease inhibitors showed that including leupeptin in the grinding buffer prevented Rubiscodegradation (Figure 4-3) in senescing tissue. In contrast, NEM, another cysteine proteaseinhibitor, was not able to prevent LSU from degradation (Figure 4-4).Two conclusions can be drawn from these results. First, the degradation of Rubisco(Figure 4-2) into lower molecular weight fragments was an artifact of the grinding buffer sincedegradation was prevented by the presence of leupeptin. Because Rubisco was not degraded inneedles from early October but was degraded in needles from late October, it appears thatchanges in the leaf during October made Rubisco susceptible to proteolytic activity in vitro. Thischange occurred at the same time as the onset of decline in chlorophyll. Clearly LSU wasdeclining during autumn, since the amount of LSU was less in senescing tissue than in greentissue. However, I had not expected to detect these fragments since they would be rapidlydegraded into amino acids for export. This suggests that the fragments in Figure 4-2 wereartifacts of the isolation procedure.A second conclusion is that leupeptin could prevent Rubisco degradation in vitro. Thisis consistent with Mae et al. (1989) who found that leupeptin inhibited LSU degradation in lysedwheat chioroplasts induced to senesce at 37°C at several different pHs and SDS concentrations(Mae et al. 1989). Mae et al. also found that Rubisco degradation was prevented by NEMalthough I did not find this effect (Figure 4-4). It is not clear why leupeptin prevented Rubiscodegradation while NEM did not. Leupeptin is a competitive inhibitor of the cysteine residues and78Figure 4-2. Immunoblot of large subunit ofRubisco in autumn senescing Larix needles, materialprepared with several protease inhibitors but not leupeptin. Gels run with total leaf proteinsamples of needles harvested from October through November. Lanes loaded per gram freshweight. Leaf grinds prepared in buffer containing different protease inhibitors (10mM EDTA,20mM DIECA, 10mM c-aminocaproic acid, 0.5% soluble PVP) but not leupeptin.79Rubisco LSUAutumn senescing larch needles‘ f\ o% %,c00 0014“—I-- ——-——-‘.—i—--& 0&0&.C,$04080Figure 4-3. Immunoblot of large subunit ofRubisco in autumn senescing Larix needles, materialprepared with several protease inhibitors including leupeptin. Gels run with total leaf proteinsamples of needles harvested from October through November. Lanes loaded per gram freshweight. Leaf grinds prepared in buffer containing different protease inhibitors (10mM EDTA,20mM D]ECA, 10mM c-aminocaproic acid, 0.5% soluble PVP) plus the additional proteaseinhibitor leupeptin (20.tM).Rubisco LSUAutumn senescing larch needlesf’o& O o o o o____4)8182Figure 4-4. Large subunit of Rubisco and proteolytic fragments in autumn senescing larchneedles prepared in buffers containing different protease inhibitors. Detection ofRubisco largesubunit by Western blots. Gels run with total leaf protein samples from either August orNovember needles. Leaves ground in different buffers: three needles samples were ground inthe stock buffer contained O.0625M Tris pH 6.8, 10mM EDTA, 20mM DIECA, 10mM caninocaproic acid, 0.5% soluble PVP. In addition, three sets ofneedles were ground in the stockbuffer plus the additional protease inhibitor 201.tM leupeptin. Finally, three sets ofneedles wereground in the stock buffer plus 2.5mM NEM.83Western larch needlesLarge subunit of Rubisco+leu +NEM Grinding Buffer—-- -,.T——AugustNovember+leu +NEM‘IGrinding BufferIL— —6645 — — — —36 -24!1484NEM covalently reacts with the active cysteine (Gray 1989). Most likely, NEM preventedRubisco degradation in isolated wheat chioroplasts (Mae et al. 1989) because NEM was able toprevent chioroplastic proteases from degrading Rubisco. In contrast, I examined total larchneedle proteins and one possibility would be that the proteases degrading Rubisco were nonchioroplastic ones that NEM was unable to prevent from degrading LSU. Perhaps higherconcentrations ofNEM could have prevented Rubisco degradation in autumn larch needles.Another interesting observation is that NEM does not prevent, and may even increase,proteolysis (Figure 4-4). One possibility is that NEM affected Rubisco conformation byinterfering with disuiphide bridge formation, and allowing Rubisco to be more susceptible toproteolytic activity. Why NEM appeared to increase Rubisco degradation in larch while theopposite effect is reported in other systems, is not known.The evidence for proteolytic activity on Rubisco during autumnal senescence ispreliminary. Further research is needed to confirm that Rubisco amount is correlated to activityin senescing tissue. In addition, it would be interesting to repeat these experiments using tissuesenescing in environmentally controlled growth chambers set to autumn conditions. Such studieswould allow for a more precise measure ofwhen Rubisco begins to degrade and whether Rubiscodeclines at the same time as chlorophyll. If fl.irther studies confirm that Rubisco and chlorophyllbegin to decline at the same time, then one can speculate about the control of these events. Forexample, changing environmental conditions by mid October could result in increased oxidativestress conditions and decreased antioxidant enzymes (Casano et al. 1994). This would jn turnlead to both an increase in Rubisco degradation (Garcia-Ferris and Moreno 1993, Mehta et al.851992) and a decline in chlorophyll due to oxidative cleavage of the porphyrin ring (Langmeier etal. 1993).86CHAPTER 5: THE ONSET OF FOLIAR SENESCENCE IN ENVIRONMENTALLYCONTROLLED CHAMBERS: THE EFFECTS OF AIR TEMPERATURE, PHOTOPERTOD,AND LEAF AGE5.0 ABSTRACTThe response of the photosynthetic apparatus to autumn is difficult to study because avariety of environmental conditions change simultaneously and may affect different parts of theleaf. I measured the impact ofmdividual environmental conditions on leaf senescence by placingwestern larch (Larix occidentalis Nutt.) in chambers set to simulate different autumn conditions.I used regression analysis to quantifr the effects of air temperature and photoperiod on the dateof decline in chlorophyll, net photosynthesis, and chlorophyll aJb ratio. The regression alsocontrolled for leaf specific variables (age and provenance). Results indicated that increased leafage significantly accelerated the onset of decline in chlorophyll, photosynthesis, and chloiophylla/b ratio. In contrast, increased air temperature delayed the onset of decline in both chlorophylland photosynthesis without affecting chlorophyll a/b ratio. Extended day length delayed only theonset of chlorophyll decline without affecting photosynthesis or chlorophyll a/b ratio. Thesefindings show that the timing and process of senescence varies with leaf age and environmentalconditions.5.1 INTRODUCTIONPerennial plants integrate autumn signals to regulate the onset and sequence of eventsduring leafdegradation. The controlled senescence process ensures that plants reabsorb t least87some nutrients before leaf abscission. Despite the growing literature on senescence, littleattention has been given to the leaf components that respond to environmental changes duringautumn. Most evidence on the effect of temperature and photoperiod on leaf senescence isanecdotal (Kozlowski et al. 1991, Addicott and Lyon 1973). Rigorous analyses of the effect ofenvironmental factors on the leaf are less common, in part because naturally growing plants areexposed to concurrent changes in a variety of environmental conditions.I studied changes in the photosynthetic apparatus of outdoor-growing western larchseedlings during one autumn season and showed that both the onset and the sequence of eventsduring senescence were affected by changes in photoperiod and soil temperature (Chapter 3).Because the data were collected only for one year, I could not determine how soil temperatureand photoperiod effects varied between years. Moreover, the difficulty remains of how to controlfor differences in weather conditions between years so that one could quantift the effect ofextended photoperiod or warmer soil temperatures on senescence.As an alternative to isolating the effect of an environmental condition on the leaf Imeasured net photosynthetic rates and pigment content of western larch leaves from seedlingssenescing in growth chambers set to different temperature and photoperiod regimes. I examinedchanges in chlorophyll, net photosynthesis and chlorophyll a/b ratio as indicators of senescence.I also measured the chlorophyll a/b ratio because in larch it is an approximate indicator of the finalstage of senescence when quantum yield has declined (Chapter 3). I analyzed the data using aregression model in which I estimated the number of days to the start of senescence as a functionof photoperiod, air temperature and leaf-specific variables such as leaf age and seed lot.88Regression models have been used previously to quantify the effect of ecophysiologicaldata on the photosynthetic apparatus (Gratiani et al. 1987, Nelson et al. 1982). The advantageofthe regression model is that it quantifies the effect of a single weather or leaf specific variableon a photosynthetic measure while other variables in the model are kept constant. Thedisadvantage of statistical models is that results do not directly explain the underlying mechanismsthat account for the effect of a given environmental condition or leaf specific attribute on thechioroplast (Sharpe and Rykiel 1991). I found that a regression model could quantify the effectof air temperature and day length (in addition to leaf age and provenance) on the photosyntheticapparatus during senescence. Results indicated that changes in individual environmentalconditions and leaf specific attributes affected both the timing and sequence of events duringsenescence.5.2 MAThRTALS AIlD METHODS5.2.1 Experimental designWestern larch (Larix occidentalis Nutt.) seedlings (West Kootenay Seed Planning Zone,British Columbia, 900m, 49 lat, 117 long; Thompson Okanagan Dry Seed Planning Zone, BritishColumbia, 1200 m, 49 lat, 119 long; and West Kootenay Seed Planning Zone, British Columbia,1500m, 49 lat, 115 long) were removed from cold storage in March and grown either outdoorsunder natural conditions orin the greenhouse. Green seedlings were placed into environmentallyregulated growth chambers (Conviron, Winnipeg, Manitoba Canada) set to 16 hours daylight,8 hours daylight, or 8 hours daylight with a 15 mm white-light night-break (30 watt incandescentlight bulb). Air temperature was either 8°C day and night or 15°C day: 12°C night, and relative89humidity was set to a minimum of 75%. Average light intensity at mid-seedling height was 150j.tmol m2s4which is approximately 75% of light intensity during autumn (Figure 3-2). Seedlingswere placed in the growth chambers at various times from June (jrior to the onset of natura1lyoccurring senescence) through April of the following year (after the onset of naturally occurringleaf senescence). Varying the timing of experiments allowed me to measure the effects of leafage (100-400 days after bud burst) on senescence.Thirty-two growth chamber experiments were conducted over a three year period from1991 to 1994, using 3-12 seedlings and an average of thirteen measurement days per experiment.Measurement days were spread roughly uniformly over the duration of individual experiments(35 to 80 days). Each measurement ofneedle conditions was from a combined sample containinga single needle from each side branch of all seedlings in the chamber. The sampling procedureattempted to ensure that my data reflected the average condition of seedlings in the chamber atthe time ofmeasurement. All measurements were taken between one and three hours after lightswere turned on in the morning to control for time of day effects.5.2.2 Pigment and photosynthesis measurementsChiorophylls were measured after acetone extraction (Lichtenthaler and Wellburn 1983).In six experiments, photosynthetic rates were measured as CO2 assimilation using an infra red gasanalyzer (ADC-LCA2, Analytical Development Company, Hoddeston, UK). In all otherexperiments, photosynthetic rates were measured as oxygen evolution (Hansatech, Kings Lynn,England) as previously described (Chapter 3).905.2.3 Determination of breakpoints and the onset of senescenceI used piecewise linear regressions to determine the day on which photosyntheticparameters declined, as previously described (Chapter 3). For example, the level of chlotophyllwas expressed as a function ofthe number of days in the growth chamber. The slope of the firstline segment was set equal to zero, consistent with the assumption that chlorophyll had not yetbegun to decline. The slope of the second line segment was allowed to differ from zero,consistent with the assumption that chlorophyll had begun to decline. Thus, the breakpoint datebetween the two line segments defined the start of the decline in chlorophyll. For eachexperiment, a series ofpiecewise linear regressions were estimated in which the breakpoint variedfrom the first measurement day of the experiment to the second-to-last measurement day.Repeating this procedure for chlorophyll in each experiment entailed estimating 400 separatepiecewise linear regressions (13 measurement days x 32 experiments).Results from the piecewise linear regressions were used to estimate the date of declineof chlorophyll as follows. First, the sum-of-squared residuals from the piecewise linearregressions were plotted against their corresponding breakpoints. For most experiments, thebreakpoint associated with the lowest sum of squared residuals was then chosen as the bestestimate ofwhen chlorophyll began to decline for that experiment. This procedure was repeatedfor photosynthetic rates and chlorophyll a/b ratio, and involved estimating approximately 800additional regressions.In a few cases, the minimum sum-of-squared residuals for a given parameter andexperiment occurred shortly after seedlings were placed in the chamber and was followed by adistinct local minimum several weeks later. The initial minimum may reflect stress caused by91moving the seedlings into the chamber rather than the onset of senescence caused by temperatureor photoperiod. In these cases, the best estimate of when the parameter began to decline waschosen by examination of both the scatter in the raw data used to estimate the piecewisç linearregressions and the sum-of-squared residuals from those regressions.5.2.4 Regression modelHaving determined the breakpoint for each experiment, I estimated the number of daysuntil the onset of senescence as a function of leaf age (LEAF Age), photoperiod (LONG Day,NIGHT Break), air temperature (WARM Temp) and seed source elevation (ELEV), using theequationDAYS= + (LEAF Age) + (2 (LONG Day) + 33 (NIGHT Break) [1]+ 34 (WARM Temp) + 135 (ELEV) + ewhere the regression was estimated separately for chlorophyll, photosynthesis and chlorophylla/b ratio. Variables in equation [1] were defined as follows. DAYS is the number of days in thegrowth chamber before the parameter in question began to decline; LEAF Age is the age of theleaf since bud burst and varied from 100-400 days; LONG Day equals 1 if photoperiod equals16 hours of daylight, and 0 otherwise; NIGHT Break equals 1 if photoperiod equals 8 hours ofdaylight with a 15 minute white light night break, and 0 otherwise; WARM Temp equals 1 if airtemperature equals 15°C day and 12 C night, and 0 otherwise; and ELEV equals 1 for highelevation seed lot, and 0 otherwise.Given these definitions, the coefficients in the regression are interpreted as follows. First,the coefficient on LEAF Age reflects the impact of a 1 day change in leaf age on the time until92senescence begins, holding constant all other variables in the model. The constant (30) reflectsthe estimated number ofdays until senescence began when environmental conditions were set to8 hours ofdaylight (LONG Day = NIGHT Break = 0), 8°C day and 8°C night (WARM temp0), and low elevation provenances (ELEV = 0). The coefficients on LONG Day, NIGHT Break,WARM Temp, and ELEV, all reflect the estimated impact of a specific environmental conditionrelative to the constant (see Zar 1984 for further discussion of dummy variables in regressionsand the interpretation of the coefficients on those variables).A problem arose in estimating equation [1] because for a number of experiments one ormore parameters (chlorophyll, chlorophyll a/b ratio, or photosynthesis) did not decline. For thesecases I did not directly observe the true value for DAY. Instead, all I know of the breakpoint inthese cases is that it would have occurred at some date after the last measurement day in anexperiment (i.e., my observation of the breakpoint is censored). For example, for mostexperiments in which chambers were set to 16 hours of daylight and warm air temperature(LONG = WARM = 1), none of the parameters discussed above declined even though theexperiments under these settings lasted up to 80 days. In addition, although for most experimentsin which chambers were set to 8 hours of daylight (LONG = NIGHT = 0) chlorophyll a/b ratiodeclined late in the experiment, chlorophyll a/b ratio did not decline in some of these experiments.Omitting censored observations would generally bias the results (Judge et al. 1988).Similarly, setting the breakpoint to the last measurement day would also bias the results since Iknow that the true breakpoint was later than the last day of measurement. On the other hand, ifI assume that the error term, e, in equation [1] is normally distributed, one can estimate equation[1] using a censored normal regression model (STATA 1993) instead of ordinary least squares.93The censored regression model includes both censored and uncensored observations and isestimated by maximum likelihood whereby the coefficients in equation [1J are chosen to maximizethe joint probability of observing the various breakpoints in the sample. The contributiomto thelikelihood fbnction from the uncensored observations is the probability of having observed a givenbreakpoint conditional on the right hand side variables. The contribution to the likelihoodfunction from the censored observations is the probability that the true breakpoint occurs afterthe last measurement day conditional on the right hand side variables (see Schmid et al. 1994 fora recent example of the use of censored regressions when analyzing biological variables).5.3 RESULTSAs an example ofthe data obtained, Figure 5-1 shows results from one experiment whereseedlings senesced under either 8 or 16 hour photoperiod and 8°C or 13 C daytime airtemperature. Under short days, chlorophyll declined about 40 days after seedlings were placedinto chambers and CO2 assimilation declined at about 45 days. The chlorophyll alb ratio did notdecline in this experiment, though it did decline in others.I used a regression model to analyze the data from all 32 growth chamber experiments,rather than two-dimensional graphical analysis, because the timing of when senescence beginsdepends not only on treatment groups but also on leaf age and provenance. The date of declinein chlorophyll, photosynthesis, and chlorophyll a/b ratio for the experiments is given in Table 5-1.Note that 6 of the 32 experiments measured photosynthesis as carbon assimilation rather thanoxygen evolution. Recall, however, that results from the outdoor autunm senescing seedlings(Chapter 3) showed that the dates of decline between carbon assimilation and oxygen evolution94Figure 5-1. Chlorophyll mg m2(a), chlorophyll alb ratio (b) and CO2 assimilation m2 (c) fromwestern larch (Larix occidentalis) seedlings placed into environmentally controlled chambers setto either 16 hour days and 15°C day, 1 2 C night air temperatures (open circles) or 8 hourphotoperiod and 8°C day, 8°C night air temperatures (filled circles). In this experiment, needleswere 107 days old. Light levels were 150 i.imol m2 at mid-seedling height and relativehumidity was a minimum of 75%.95500 I I I I(a)400 -0 OD00Q.300- • o0oo ••‘.0êø. •..200 -, 100 - 0 Long Day. Warm..• Short Day. Cool0 I I I I0 10 20 30 40 50 60 705 I I•(b).0 0•’-’ Q0 • . .1 - 0 Long Day. Warm• Short Day. Cool0 I I I I0 10 20 30 40 50 60 70C15O I I I I I00 (c)• 10 0 • 00E..0 Long Day. Warm• Short Day. Cool0 I Ir 00 10 20 30 40 50 60 70Days in Chamber96Table 5-1. Number of days until decline of chlorophyll (C), chlorophyll a/b ratio (AB), orphotosynthetic rates (P; measured either as carbon assimilation or oxygen evolution) for seedlingssenescing under different photoperiod and temperature regimes. Each line represents anindividual experiment using between 3-12 seedlings. Chlorophyll (cchl), chlorophyll a/b ratio(cab), or photosynthetic rates (cp) which did not decline during the duration of the experiment(censored) are indicated by a 1, 0 otherwise. Environmental conditions in the chambers were asfollows. Photoperiod was either 8 hour short day (SIT), 16 hour long day (LO) or short day witha 15 nun night break (NB). Temperature was either warm: 15°C day, 12°C night (W) or cold:8°C day, 8°C night (CO). Source of seedlot elevation was either 900m, 1200m, or 1500m. AGErefers to the age of the needles at the start of the experiment (number of days since bud burst).Note that data was collected until there was no net photosynthetic rate. Hence, the duration ofeach experiments was different. Several experiments were done with leaves> 1 year old.AGE C AB P echi cab ep SH LO NB W CO 900m 1200m 1500m9494949410710710710710710710710710711411713013020320320320320320332332336436438538541541542546 77 7764 64 6417 60 3473 73 7322 52 4240 30 5247 61 3146 74 6119 61 6319 34 2463 63 6363 63 6363 63 6367 51 6740 51 3539 44 3944 31 287 3323 44 277 3311 2727 44 2730 3329 52 2952 52 5223 50 2350 30 5010 27 146 19 1950 27 5030 27 3621 35 250 1 00 1 00 1 00 1 01 0 00 1 00 0 11 0 00 1 00 0 10 1 00 1 00 1 00 1 00 1 00 1 00 1 01 0 00 1 00 0 11 0 00 1 00 0 10 1 00 1 00 1 00 1 00 1 00 1 00 1 00 1 00 1 00 1 1 1 0 0 101 1 1 1 0 0010 1 0 00 1011 1 1 0 1 0100 0 0 1 0 00100 0 1 0 0010 1 0 1 00010 1 0 0 1 0010 0 0 0 1 00100 0 0 1 0011 1 1 0 1 0101 1 1 0 1 0101 1 1 0 1 0101 0 1 0 1 01000 0 0 1 001000 0 1 0011 0 0 0 1 0100 1 0 1 0 0010 1 0 1 0 0010 1 0 1 0 0010 0 0 0 0 1010 1 0 00 1011 1 0 0 0 1010 1 0 1 0 0 1•00 1 1 0 1 0100 1 0 1 0 0011 0 1 0 1 01001 0 1 000100 0 0 0 1011 0 1 0 1 010000 1 00100 1 0 0 1 00197were quite similar. For that reason, I pooled measures from experiments using carbonassimilation and oxygen evolution when estimating the regressions in order to increase samplesize and, thereby, decrease the confidence intervals around the estimated effects ofvariables onthe onset of senescence.Results from the regression analysis showed that the degradation pattern for seedlingsexposed to short photoperiod and cool air temperature was similar to that of autumn senescingleaves (Chapter 3). Needles that were 100 days old began to lose chlorophyll in an average of30 days (Table 5-2) holding other attributes constant: observe that the constant 13 equals 38while the coefficient on leafage was -0.077 per day. Hence, for a 100 day old needle, chlorophyllloss would begin in 38 minus 8 (-0.077 times 100) days. In contrast, photosynthesis did notdecline until 47 days after the start of the experiment (p0=57 and the coefficient on leaf age = -0.095), and the chlorophyll a/b ratio declined in 73 days (p0=83 and the coefficient on leaf age= -0.096 days). Thus the decline in chlorophyll was followed by a decline in photosynthetic ratesand, finally, a decline in chlorophyll a/b ratio. The probability that chlorophyll levels declinedbefore photosynthetic rates declined, and that photosynthetic rates declined before chlorophylla/b declined, was significant (p=0.0025).Additional growth chamber experiments provided estimates on the effect of warmer airtemperatures and longer photoperiod on the dates of decline in chlorophyll, photosynthesis andchlorophyll a/b ratio for seedlings senescing under short cold days. Photoperiod affected the timeto chlorophyll decline but did not affect the days to decline of photosynthesis orchlorophyll a/b ratio. The short day treatment accelerated the onset of chlorophyll decline by anaverage of 12 days (p0.05) compared with seedlings that senesced under long days. As a result,98Table 5-2. The effect of Larix occidentalis needle age and environmental conditions on thenumber of days in the growth chamber until chlorophyll m2, photosynthesis (measured as CO2assimilation or 02 evolution) and chlorophyll a/b ratio declined.Chlorophyll Photosynthesis Chlorophyll a/bCONSTANT 38.12 ± 14.12 57.08 ± 15.93 82.45 ± 24.85(0.0001) (0.0001) (0.0001)LEAF Age -0.077 ± 0.049 -0.095 ± 0.05 1 -0.096 ± 0.077(0.003) (0.001) (0.017)LONGDay 12.61 ± 12.60 5.721 ± 13.99 -12.02± 21.49(0.050) (0.858) (0.263)NIGHT Break -1.35 ± 13.28 -8.78 ± 20.26 -12.66 ± 28.16(0.864) (0.385) (0.367)WARM Temp 32.48 ± 14.08 24.35 ± 15.00 8.47 ± 21.38(0.0001) (0.003) (0.426)ELEV -1.26 ± 14.17 -22.26 ± 22.32 -2.49 ± 31.20(0.878) (0.063) (0.872)N 32 28 32Censored 10 10 19Coefficient ± 95% confidence interval with 2-Tailed P value in parentheses. Coefficients wereobtained based on a censored normal regression model because in some cases, the physioJogicalparameter did not decline during the duration of the experiment. Photosynthetic rates in six ofthe 32 experiments were measured as carbon assimilation, the remaining were measured asoxygen evolution.99chlorophyll declined before photosynthesis in seedlings senescing under short days, andconcurrently with photosynthesis when seedlings senesced under long days. I tested thepossibility that phytochrome influenced the delayed yellowing due to the extended photoperiodby inducing seedlings to senesce under 8 hour days with 15 mm ofwhite-light during the night.Seedlings exposed to the interrupted night senesced at the same time as seedlings receiving anuninterrupted night (p:O.864).Air temperature affected both chlorophyll content and photosynthesis levels. Seedlingsexposed to an 8°C air temperature experienced an earlier onset of decline in both photosynthesis(p<0.05) and chlorophyll (J)<O.0001 compared to seedlings senescing in 15°C (Table 5-2; I.e. forchlorophyll, the coefficient on WARM Temp for a 100 day old leafwould be [3 8-81+32=62 days.Thus seedlings with 100 day old leaves senescing under warm air temperature lost chlorophyllin 62 days while those senescing under cold temperature lost chlorophyll in [3 8-8]=3 0 days). Airtemperature did not appear to affect the order of decline of chlorophyll and photosynthesis. Airtemperature also did not affect the onset of decline in the chlorophyll a/b ratio (p=O.426). Thislatter result reflects that chlorophyll a/b did not decline until late during senescence.Leaves responded differently to environmental cues depending on their age, which rangedbetween 100-400 days in these experiments. Increased leaf age significantly (p<0.000l)accelerated the decline of all photosynthetic components measured. There was an 0.08 dayearlier decline in chlorophyll for a one day older leaf (Table 5-2). For example, chlorophyll lossbegan in an average of30 days for needles 100 days old, but in 22 days in seedlings with needles200 days old, holding other variables constant.Provenance did not significantly affect the onset of decline in chlorophyll in these100experiments, though seed lot may have been important in the onset of decline in photosynthesis.For example, for a typical “autumn” seedling (200 day old leaves senescing under short days andcool temperatures), seedlings from the higher elevation (1500m) provenance lost photosyntheticability earlier (p=0.063) than seedlings from the lower provenances. As a result, among seedlingsfrom the high elevation provenance, chlorophyll declined after photosynthesis while amongseedlings from the lower provenances, chlorophyll declined before photosynthesis.5.4 DISCUSSIONI found the same pattern of senescence in seedlings in the growth chamber as iii thosesenescing under autumn conditions. All seedlings senescing under short photoperiod and cooltemperature experienced a decline in chlorophyll, followed by a decline in photosynthetic rates,and then, a decline in chlorophyll a/b ratio (Table 5-2). These similarities gave me confidence thatmy interpretations from the growth chamber experiments are good approximations of senescencein outdoor seedlings during autumn. I identified environmental conditions that affect differentcomponents of the chioroplast by inducing plants to senesce under a variety of conditions inenvironmentally controlled growth chambers and I showed how increased leaf age affected theseedling’s response to environmental changes.Extended photoperiod delayed the onset of chlorophyll decline but did not affectphotosynthesis. I previously showed that extended photoperiod in larch delayed the onset ofdecline in chlorophyll (Chapter 3). Olmsted et al. (1951) found similar results in other speciessenescing outdoors near streetlights. My results are novel because they show that the delay in101senescence was due to a delay in the onset of chlorophyll loss and not to a delay in the declineof photosynthetic rates.I induced seedlings to senesce under short photoperiods but with a white light night breakto test the hypothesis that phytochrome controlled the delay in yellowing under extendedphotoperiods. The absence of a night-break effect suggested lack of phytochrome involvement.Although a response to a night-break might have occurred with a different wavelength of light,Tucker (1981) found no effect of red light on chlorophyll levels in Cornus.Phytochrome may not affect leaf senescence directly in deciduous trees. A response toa night break has not yet been shown and altered hormone levels caused by dormancy (Wareing1956) and frost hardiness may indirectly affect chlorophyll levels. For example, shoot extensionis continuous in larch unless the plant receives a short photoperiod (Gowin et al. 1980). In Larixleptolepis, bud set occurred three weeks after exposure to a week of 10 hour days (Meijer andvan der Veen 1957). Thus, leaf senescence could be delayed by long days which can cause butburst and shoot elongation.It is not known ifgrowth rates are affected in seedlings that remain green longer becauseof exposure to an extended photoperiod. Data from non native species of Populus (Tschaplinskiand Blake 1989, Nelson et al. 1982) and Lonicera (Schierenbeck and Marshall 1993) show thatthere may be a growth advantage to cohorts that retain leaves later into the season. The potentialgain from the longer growing season may be lost if an early frost causes leaf abscission of greenleaves.Low air temperature accelerated initiation of senescence in larch seedlings. Seedlingsbegan the decline of chlorophyll and photosynthesis earlier (j<0.01) when exposed to 8°C air102temperature than to 15°C. I note that while I did not measure soil temperature, and thus onlydiscuss the effect ofthe measured difference in air temperature, most likely the effect of the lowerair temperature is also in part due to lower soil temperature. The temperature effect onchlorophyll could be due to decreased chlorophyll synthesis as well as an increased chlorophylldegradation. Gowin et al. (1980) showed that chlorophyll levels were lower in larch grown atlow air temperature and that air temperature did not affect the chlorophyll a/b ratio, consistentwith my results. Since the process ofchlorophyll degradation is still under investigation (Hendryet al. 1987), it is difficult to explain how low temperatures initiate or accelerate chlorophylldegradation. Some studies suggest that low temperatures may affect chlorophyll loss indirectlyby exposing tissue to photoinhibitory conditions (Hendry et al. 1987). However, assimilationrates did not cause lower chlorophyll levels in larch because chlorophyll declined before thedecline in photosynthesis. The temperature-sensitive process of chlorophyll degradation mayinvolve oxidative damage caused by temperature-induced changes in the membrane (Cuello andLahora 1993).In addition to affecting onset of chlorophyll decline, lower air temperature significantly(p<0.01) hastened the onset of decline in photosynthesis. Plants often show decreasedphotosynthetic rates when exposed to cooler than average temperatures (Oquist and Martin1986). I recognize that my estimate of the effect of temperature on the date of decline inchlorophyll and photosynthesis may be biased because measurements were done at roomtemperature (20-24°C) rather than at the growing conditions ofthe seedlings (Gowin et al. 1980).However, the bias was small since CO2 assimilation rates in green needles growing under warmor cool temperatures were the same (Figure 5-ic). Another potential overestimation of the103coefficients on air temperature may have occurred because I began experiments with non-hardened plants (Minorsky 1989, Tranquillini et al. 1986), though they may have becomehardened during the experiment.The regression model allowed me to estimate the effect of leaf-specific variables (leaf ageand provenance) on the timing of senescence. I did not find significant differences in the onsetof senescence among the lower and higher elevation provenances. This suggested that differencesin the timing of senescence among provenances were largely due to climatic factors and notbecause ofgenetic differences across seed lots. One implication of this result is that differencesin the onset ofsenescence among provenances did not persist when seedlings were growii in thesame location.Older larch needles responded more quickly to senescence cues than younger leaves, butthe cause ofthe delay in senescence ofyounger leaves is not known. Resting Eounymus leaveslost chlorophyll under cool temperatures while actively growing plants could not undergo thecolor change (Creasy 1974). This suggests that it may not be leaf age but their developmentalstate that is important in initiating senescence. Fischer and Feller (1994) induced senescence inyoung and old detached wheat leaves. Younger leaves were delayed in the onset of the artificiallyinduced senescence compared to older leaves. Fischer and Feller (1994) concluded that factorsother than source-sink relations between the leaf and the plant were important during leafsenescence because the leaves were detached and thus not receiving senescence cues from therest ofthe plant. Finally, if post-transcriptional regulation of senescence genes is required for theonset of senescence, then it is possible that younger leaves may not have the apparatus necessary104for initiating the senescence process (Smart 1994). Younger larch needles can senesce, theysimply require a longer time to respond to senescence-inducing conditions.I found that the order of decline in chlorophyll, photosynthesis and the chlorophyll a/bratio was the same as previously recorded in outdoor senescing seedlings (Chapter 3). Thissuggests that if differences in leaf age and the environment are accounted for, the processes ofchlorophyll loss and photosynthetic decline were similar in seedlings growing under differentenvironmental conditions. My results also showed that the decline in photosynthesis andchlorophyll were controlled by different environmental triggers, and that these responsesdepended on leaf age. This suggests that leaves alter their senescence program depending on thesenescence-inducing conditions. Most likely, senescence in deciduous trees proceeds in anorderly fashion in which assimilation rates are maximized during nutrient translocation. Byhaving a variety of pathways available for degradation, trees can integrate signals from theconstantly changing environment and initiate senescence at the optimal time.105CHAPTER 6: THE RATE OF PHOTOSYNTHETIC DECLINE DURING AUTUMN LEAFSENESCENCE: THE EFFECTS OF SOIL TEMPERATURE AND PHOTOPERIOD6.0 ABSTRACTI developed a regression model to predict photosynthetic rates in western larch (LarixoccidentalisNutt.) during senescence. Treatment groups were used to identify soil temperatureand photoperiod effects, and a regression model was used to account for all other environmentalvariables that could affect photosynthetic rates during autumn. The variables in the regressionmodel were chlorophyll content, air temperature, and Julian date. These variables were chosento control for past and current growing conditions of the needles. Coefficients from theregression for 1991 and 1992 were similar, suggesting that the model was robust despitesubstantial weather differences during the two years. Results for naturally senescing leaves wereas expected. CO2assimilation rates were lower in leaves with less chlorophyll, regardless of leafage or air temperature. Furthermore, assimilation rates decreased in older leaves for a given levelof chlorophyll and air temperature. Treatment groups were different. An extended photoperiodduring autumn did not affect CO2 assimilation rates, which is expected given that photosynthesisat any given moment is governed by light levels and time-of-day rather than total hours ofdaylight. Seedlings senescing under above ambient soil temperatures had higher photosyntheticrates for a given level of chlorophyll in older leaves compared to naturally senescing seedlings.This was evident in 1991 when the difference in soil temperature between the treatment groupswas 8°C but not in 1992 when the difference was only 3°C. These results show that the process106of senescence can be affected by differences in autumn weather and that photosynthetic ratesduring autumn can be predicted.6.1 INTRODUCTIONIn previous chapters of this thesis I examined the effects of environmental and plantspecific conditions on the timing of senescence in larch seedlings. In this chapter I addressdifferent questions, namely, what is the rate of photosynthetic decline during autumn and how isthat rate affected by soil temperature and photoperiod? I also examine the impact of soiltemperature and photoperiod on the relationship between photosynthesis and chlorophyll insenescing leaves.The methods to examine photosynthetic rates were based on those described for studyingthe onset of senescence. In Chapter 5, I induced seedlings to senesce in environmentallycontrolled chambers that held all but a few environmental variables constant. I then used aregression model to quanti[’ the effect ofenvironmental and plant specific conditions on the onsetof senescence. In the present chapter, I use an analogous regression model to determine theeffect of environmental conditions on the rate of photosynthetic decline using data from outdoorsenescing seedlings. Data for the outdoor senescing plants were obtained as described in Chapter3 and include assimilation rates, chlorophyll content, and weather related variables for seedlingssenescing under natural conditions, warm soil conditions, and extended photoperiod.The regression model is designed to determine the effect of soil temperature andphotoperiod on the rate of photosynthetic decline during autumn. To highlight those effects,variables identifying the control (naturally senescing), warm soil, and extended photoperiod107treatment groups were included as regressors. Additional variables were included to control forother past and present environmental conditions which affect photosynthetic rates and which maybe correlated with soil temperature and photoperiod. In this way, the regression modeldetermines the effect of soil temperature and photoperiod while holding constant other weatherrelated variables that affect photosynthesis and which may be correlated with soil temperature andphotoperiod.6.2 MAThRJALS AND METHODS6.2.1 Experimental designOne-year old western larch seedlings (Larix occidentalis Nutt.) were grown in outdoorcold frames under either natural, warm soil, or extended photoperiod conditions. Controlseedlings were exposed to natural outdoor temperatures and day length. Warm soil seedlings hadheating cables under the soil but were exposed to naturally declining day length. Extendedphotoperiod seedlings received a sixteen hour photoperiod but were subjected to decliningambient temperatures. For each treatment group, chlorophyll content and CO2 assimilation rateswere measured every two to three days during the autumn period as previously described for1992 data (Chapter 3). During 1991, the same methods were used as during 1992 except thatmeasurements were not made in triplicate each day. Second, I used a different samplingprocedure: rather than harvesting a single needle from each tree in a treatment group, andpooling all needles, as in 1992, I harvested several needles from several trees in the treatmentgroup. The result was that in 1991, there was more variation in the data (Figures 6-1 and 6-2).108Figure 6-1. Carbon assimilation rates for control (filled circles) and warm soil (open circles)seedlings in (a) 1992 and (b) 1991.10914 p I I I I I- 12- (a)E 10-08--a’’0 I I I I •I I I240 250 260 270 280 290 300 310 320 330 340 35014 I I I I I I I I I- 12 0 Co. (b)ioj00E 6-oo.a’a’2-0 I I240 250 260 270 280 290 300 310 320 330 340 35027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec110Figure 6-2. Chlorophyll content in control (filled circles) and warm soil (open circles) seedlingsin (a) 1992 and (b) 1991.111400I Io (a)E 300200Q..o1000 I I I I I240 250 260 270 280 290 300 310 320 330 340 350400•1• • ‘300•,(b)200- 0..100-0 I I I I I240 250 260 270 280 290 300 310 320 330 340 35027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec1126.2.2 Development of modelPhotosynthetic rates on a given day are determined by many factors, including time ofday, current weather condition, history of the plant (i.e. cumulative stress received), age of theplant, and age of the needle. Therefore, to determine the effect of soil temperature andphotoperiod on photosynthetic rates (and related effects on the relationship between chlorophylland photosynthesis), it is necessary to control for other environmental conditions that may becorrelated with soil temperature and photoperiod and which also affect photosynthesis. Forexample, as a preliminary step, I plotted autumn CO2 assimilation rates against both soiltemperature and photoperiod for control seedlings. Although these plots showed clear evidenceofa positive correlation between photosynthetic rates and both soil temperature and photoperiod,the graphs do not necessarily imply that photosynthetic rates decline with either soil temperatureor day length. Instead, it is possible that photosynthetic rates are lower later in autumn becauseofdeclining air temperatures that happen to be correlated with day length and soil temperature.Hence, two-dimensional graphs are limited for the questions addressed in this chapter becausephotosynthetic rates depend on a variety of environmental conditions in addition to soiltemperature and photoperiod.Some of the other conditions affecting photosynthetic rates during autumn werecontrolled in the sampling procedure. Measurements were made at approximately the same timeeach day, thus controffing for time-of-day effects. Similarly, only one-year old seedlings wereused in both 1991 and 1992, ensuring that plant and needle age were similar when autumnsenescence began.113Although the sampling procedure does control for time-of-day and both plant and needleage, I still needed to control for numerous other past and present environmental conditions inorder to isolate the effect of soil temperature and photoperiod on photosynthetic rates. For thatreason, I developed a regression model analogous to the one used in Chapter 5. Specifically, Iexpressed photosynthetic rates as a linear function of dummy variables pertaining to the control,soil temperature, and photoperiod treatment groups, in addition to several variables that accountfor other past and present environmental conditions to which the seedlings were subjected.In selecting variables for the regression model two potential problems had to be a’oided:multicollinearity and correlation between regressors and the error term (i.e., the inclusion ofendogenous regressors). Including highly correlated variables in the model (such as relativehumidity and air temperature) can lead to multicollinearity if there is not enough variation in thedata to separate the individual contributions from each variable. Although multicollinearity doesnot bias estimates from a linear regression model, standard errors become large making it difficultto evaluate the impact of individual right hand side variables. The only way to control formulticoffinearity is to increase sample size (thus increasing variation in the data) or to include adifferent set of regressors that are not too correlated with each other.In contrast, inclusion of endogenous right hand side variables leads to biased estimatesbecause such variables would be correlated with the error term in the model. For example,although stomatal conductance can limit photosynthetic rates, photosynthetic rates can alsodirectly affect stomatal conductance. Thus including stomatal conductance as a regressor without114controlling for the simultaneous relationship between stomata! conductance and photosyntheticrates would lead to overestimates ofthe impact of stomata! conductance on photosynthesis. Forthat reason I chose to use only exogenous regressors in the model.Considering these points, I selected three variables to represent past and presentenvironmental conditions influencing photosynthetic rates: air temperature, Julian date, andchlorophyll content. These three variables were all correlated with photosynthesis but were notlikely to be closely correlated with each other. Also, all three regressors are exogenous: a hangein photosynthesis at a given moment has no effect on air temperature, date, or even chlorophyllcontent since chlorophyll changes only slowly over time compared to photosynthetic rates.In selecting the three regressors for the model, air temperature was chosen to proxycurrent environmental conditions since air temperature fluctuates on a daily basis. I also triedusing light intensity or relative humidity to control for current environmental conditions sinceboth of those variables also fluctuate daily. However, air temperature explained more of thevariation in photosynthesis and was chosen for that reason. Including all three (or even two)variables in the model led to collinearity problems.Julian date was included to control for environmental conditions that changed graduallyduring autumn. Julian date also captures changes in plant and needle age that occur as autumnprogresses. Together, air temperature and Julian date were assumed to capture most of thecurrent environmental conditions affecting the seedlings on a measurement day.Chlorophyll was used to proxy past environmental conditions to which the plant has beenexposed. Chlorophyll content reflects the accumulated history of environmental conditions underwhich a plant has grown. As examples, leaves growing in nutrient deficient conditions typically115have less chlorophyll than leaves growing in nutrient rich environments while leaves growing inthe shade will have different levels of chlorophyll from those growing in the sun (Salisbury andRoss 1985). These cases support use of chlorophyll content as a broad measure of the degreeto which the plant has previously been subjected to favourable or unfavourable environmentalconditions.Given the above arguments, I initially specified the following model,AMAX=b0+b1CHL+2J+b3IRT e. [1]where AMAX is the level of photosynthesis, CHL is mg chlorophyll m2, 3 is Julian date, andAIRT is air temperature (°C). In addition, b0 is a constant, b1, b2, and b3 are the coefficients onCHL, 3, and AIRT, respectively, and e is the error term. I assume that e is normally distributedwith mean zero.The model described by equation [1] assumes that warm soil and extended photoperiodhave no effect on photosynthetic rates since equation [1] does not control for differences intreatment groups. For this reason I created two dummy variables (Zar 1984), T and D, todetermine the impact of the treatment groups on photosynthetic rates. T was set to 1 if theobservation was from the warm soil group and zero otherwise, while D was set to 1 if theobservation was from the extended photoperiod group and zero otherwise. I then set thesevariables to interact with the other regressors in the model. Including all of the interactive termsin the model resulted in the following equation which was estimated by ordinary least squares(Neter et al. 1990),AMAX = b0 + bICHL+b2CHL*T+b3CHL*D + [2]b4J + b5J *T+b6J*D+b7AJRT +b8AIRT *T+b9AIRT*D + e.116Interpretation of equation [2] would be as follows: on average, a 1.0 mg decrease inchlorophyll for control seedlings would give a b1 decrease in photosynthesis, for any given date(J) and air temperature (AIRT). But a 1.0 mg decrease in chlorophyll for the warm soil seedlingswould give ab1+b2 decrease in photosynthesis, and for the extended photoperiod seedlings, ab1+b3decrease. Similar interpretation would hold for date and air temperature.When equation [2] was estimated with assimilation rates transformed to log form, thecoefficients remained significant and of the same sign as when the model was run with ‘no logtransformation (Table 6-1). This suggests that the results were not sensitive to reasonablechanges in functional form: although the true form may not be linear, there is no evidence tosuggest that a linear model is inappropriate as a first approximation towards analyzing the data.6.3 RESULTSFigures 6-1 and 6-2 show that carbon assimilation rates and chlorophyll content declinedapproximately at the same time. However, the effect ofwarm soil and extended photoperiod onphotosynthetic rates was not easily identified because assimilation rates are a function of severalenvironmental conditions. In addition, the effect of warm soils on both the onset and rate ofsenescence was relatively small during 1991 and 1992, and graphs did not provide informationon the degree to which the observed impact ofwarm soil was statistically significant. For thesereasons, the regression model described above was used to estimate the effect of soil temperatureand photoperiod on the rate of photosynthetic decline.The regression quantified the effect of Julian date, chlorophyll content, and airtemperature on photosynthetic rates during autumn and provided a level of significance for these117Table 6-1. Regression analysis ofphotosynthetic rates inLarix occidentali seedlings, for autumn1991 and 19921991 1992DEP VAR AMAX LOG AMAX AMAX LOG AMAX(j.imol m’s x 10) (.tmo1 m2s’ x lOjCONSTANT 16.065 3.610 18.137 3.773(3 955)1 (4.1 16’ (4•959)1 (5.29O13 -462.747 -93.560 -478.879 -80.647(-3.852’ (-3.6O7’ (-4.3O2’ (-3.702’3*T 76.828 19.594 21.777 0.150(2.2232 (2.625’ (0.601) (0.021)J*D 14.808 6.558 26.286 -6.105(0.420) (0.861) (0.799) (-0.866)CHL 94.674 19.506 91.905 21.469(3 .728 (3 .557’ (3 .242’ (3 .832’CHL*T-52.124 -12.354 2.248 0.802(-1.487) (-1.633) (0.053) (0.097)CBL*D-32.948 -7.945 2.934 0.661(-0.950) (-1.060) (0.083) (0.093)AIRT 1399.640 290.184 675.752 2.358(1.718) (1.650) (1.389) (0.025)AIRT*T-385.343 -106.063 -14.334 39.980(-0.400) (-0.510) (-0.022) (0.317)AIRT*D-88.618 -29.132 -438.950 72.718(-0.087) (-0.133) (-0.761) (0.642)Adj r’ 0.306 0.265 0.417 0.374n 270 270 252 247Res SS 1699.669 79.240 1184.743 44.028Fratio 14.196’ 11.795’ 20.9621 17.307’Model as explained in text Coefficients are given with the t-ratio in parenthesis. Jjulian date, J*T and 3*D interactivevariables for warm soil and extended photoperiod treatment groups; CHL mg chlorophyll m’; CHL*T and CHL*Dinteractive variables for treatment groups; AIRT air temperature (°C), AIRT*T and AIRT*D interactive variables fortreatment groups. 1p <0.001 for 2 tail ‘p <0.01 for 2 tail test.118estimates (Table 6-1). The model was robust to two years of data from outdoor senescing larchseedlings. In both years, a decrease in CM resulted in significantly lower photosynthetic rates(p<0.001) for any given date and air temperature. Specifically, a leafwith 1 mg less chlorophyllm2would have 0.009 j.imol m2 4 less carbon assimilation. In both years, an older leaf had lowerphotosynthetic rates (p<0.01) than a younger one, for any given level of chlorophyll and airtemperature. Specifically, a one day older autumn leaf fixed 0.05 j.tmol m2 s’ less carbon for agiven amount of chlorophyll and air temperature. Finally, in both years, air temperature did notsignificantly affect photosynthetic rates. This was not unexpected given that air temperatureswere mild (Figure 6-3), after mid October (day 290) averaging 7.5°C ± 0.28 SE (n=68) in 1991and 13°C± 0.17 SE(n=140)in 1992.In addition to providing estimates on the rate of photosynthetic decline among naturallysenescing seedlings, the model also determined the effect of the treatment groups onphotosynthetic rates. Seedlings receiving warm soils showed significantly (p<0.05) higherphotosynthetic rates in 1991 than control seedlings in late autumn for a similar amount ofchlorophyll (Table 6-1). Thus, Wa 1991 leaffixes 0.0463 j.tmol CO2m2 s’ less carbon than a oneday younger leaf a one day older warm-soil leaffixes only 0.03 86 (-0.0463+0.00768) J.LmQl CO2m2 less carbon. In contrast, warm soil and control seedlings were not significantly differentin 1992. This likely reflects the fact that in 1991, the difference in soil temperature betweencontrol and warm soil seedlings (Figure 6-4) averaged 8.09 ± 0.43°C SE (n=67) while in 1992the difference was only 2.89±0.12°C SE (n=72).119Figure 6-3. Air temperature (°C) at needle harvest for a) seedlings in 1992 growing under naturaloutdoor conditions (filled circles), with warmer soils (shaded circles), or with extendedphotoperiod (open circles). b) Air temperature for control seedlings in 1991.120Air temperature °C30I I.(a).4)20•154) %‘ oE104)50 I I I I I240 250 260 270 280 290 300 310 320 330 340 35030 i I I(bYOb 25l—i20154)C4t 44)5.0 I I I I240 250 260 270 280 290 300 310 320 330 340 35027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec121Figure 6-4. Minimum soil temperatures (°C) for control (lower line) and warm soil (upper line)seedlings in a) 1992 and b) 1991.122Minimum Soil Temperature15020(a)Eio.‘-05rI0 I .1 I I I I I• I I240 250 260 270 280 290 300 310 320 330 340 350—425(b)20C)015E•+- 1000 I I I240 250 260 270 280 290 300 310 320 330 340 35027 Aug 16 Sep 6 Oct 26 Oct 15 Nov 5 Dec123I previously showed that seedlings receiving an extended photoperiod remained greensignificantly longer into autumn compared to control and warm soil seedlings (Chapter 3).However, Table 6-1 shows that the interactive term between photoperiod and the otherregressors in the model was insignificant. This suggests that the extended photoperiod did notaffect the rate of senescence after controlling for Julian date, chlorophyll content, and airtemperature. Thus extended photoperiod delayed the onset of senescence but did not affect theability of the leaf to photosynthesize for a given level of chlorophyll.To further examine the relationship between carbon assimilation and chlorophyll content,I ran a related regression to the one used on autumn data. I measured chlorophyll content onsame age needles from partially senescent seedlings induced to yellow in growth chambers set to16 hour days and 8°C air temperature. Because all environmental conditions were constant in thegrowth chamber, photosynthetic rates could be estimated as a function of chlorophyll, with noadditional variables in the model.The coefficient on chlorophyll from the regression of photosynthesis as a function ofchlorophyll (n30) was 0.0140±0.002 (95% confidence interval). The value was the same asmeasured in the autumn data (0.00947±0.00498 in 1991 and 0.00919±0.00556 in 1992). Thesimilarity between the three regression coefficients on chlorophyll from growth chamber seedlingsand from outdoor seedlings provides a robustness check of the regression model. Furthermore,the similarity across three different data sets suggests that the relationship between chlorophylland photosynthesis is stable in western larch when environmental conditions are accommodated.1246.4 DISCUSSIONPhotosynthetic rates in deciduous leaves vary during autumn because of changes innumerous environmental conditions. Since the environmental conditions often changeconcurrently, it is difficult to resolve the effect of a specific variable. Furthermore, althoughanecdotal evidence suggests that temperature and photoperiod are important in regulatingsenescence, rigorous analyses have been difficult because autumn conditions are not identical,making it difficult to compare results of repeated experiments from different years.The data for 1991 and 1992 presented in Figures 6-1 and 6-2 demonstrate some of theseproblems. First, both soil and air temperature differed in the two years (Figures 6-3 and 6-4).Second, differences in the sampling procedure resulted in smaller scatter in the 1992 data set,making it possible to determine the date of decline in chlorophyll and carbon assimilation in 1992but not in 1991. Chlorophyll levels in the 1991 sample did not decline below 100 mg chlorophyllm2 (Figure 6-2b). The presence of green needles late into autumn was a result of wintergreenneedles in the sample of needles harvested for analysis. Larch may retain green needlesthroughout winter (Richards and Bliss 1986), and senescence did not occur in these needles untilthe following autumn. The apparent rise in chlorophyll oflate 1991 autumn needles (Figure 6-2b)also reflected the presence ofwintergreen needles. Because a significant portion ofwintergreenneedles were sampled in 1991, it was not possible to identify the onset of senescence as done in1992 where wintergreen needles were not sampled.The data from 1991 could not be analyzed in a similar manner to that collected from 1992because of the wintergreen needles and the different sampling procedure. However, there wasvaluable information in the variation that was recorded in photosynthetic rates. I developed a125regression model to analyze the variation and determine ifwarm soil and extended photoperiodhad a significant effect on photosynthetic rates during autumn senescence.I used a regression model to predict changes in assimilation rates during autumn and toisolate the effect of two environmental conditions, soil temperature and photoperiod, onphotosynthetic rates. Senescence under natural conditions was compared between two treatmentgroups where either soil temperature or day length were different from naturally senescingseedlings. All other autumn conditions were held constant in the regression model. This wasdone by selecting three variables to approximate the environmental conditions affecting the leavesapart from soil temperature and photoperiod. In this way it was possible to isolate the effect ofsoil temperature and day length on senescence, controlling for environmental changes. Oneadvantage of the regression model is that it was possible to compare results from two years ofdata because autumn conditions during the two years were held constant in the regression model.Results from the regression model (Table 6-1) showed that despite the variation recordedin 1991, trends were similar to 1992. First, several inferences from the model may appearobvious but are mentioned because they serve to increasc confidence in the model. A d&reasein chlorophyll resulted in significantly (p<O.OO1) lower carbon assimilation for any given date andair temperature. The decline was expected since chlorophyll levels below a certain threshold canresult in less light harvesting capacity and energy for photosynthesis.It was also not unexpected to find that a one day older leaf had significantly (p<O.OO1)lower carbon assimilation than a younger one, for a given level of chlorophyll and airtemperature. Previous studies have also shown that older leaves have reduced ability tophotosynthesize (Schaper and Chacko 1993, Constable and Rawson 1980). The results from the126regression analysis provide additional information, because they show that the reduced level ofphotosynthesis in older leaves was independent of a decline in chlorophyll and air temperature,since these variables were held constant in the regression model. In other words, the decline inphotosynthesis in older leaves cannot be completely explained by declining chlorophyll levels orair temperatures.In contrast, some inferences from the model were less obvious. For example, airtemperature did not significantly influence carbon assimilation during autumn. Although decliningair temperatures often reduce assimilation rates, the rate of temperature change as well as themagnitude of the temperature change can be important factors in the response of the leaves totemperature changes (Minorsky 1989). Tn fact, air temperatures during both autumn 1991 and1992 were mild for hardened tree seedlings and did not average below 7°C after mid October.Thus air temperature was not as important in these outdoor experiments as it may be in others.In contrast to air temperature, soil temperature did affect photosynthetic rates. Seedlingsreceiving warmer soil temperatures showed significantly higher carbon assimilation rates thancontrol seedlings in late autumn 1991 but not 1992. The 3°C difference in soil temperature during1992 was apparently not enough to cause a significant physiological change in the seedlings butthe 8°C difference in 1991 did alter photosynthetic rates. These results show that small variationsin weather can have important effects on the ability of seedlings to obtain photosynthate lateduring the season.In addition to temperature, photoperiod is the most often cited environmental variableaffecting the autumn senescence process. Seedlings receiving an extended photoperiod remainedgreen two weeks later than did control seedlings (Chapter 3), yet the ability of the leaves to127photosynthesize was the same: the coefficients from extended photoperiod seedlings were notsignificantly different from control seedlings. This suggests that although leaf yellowing wasdelayed, photosynthetic decline was the same for both control and extended photoperiodseedlings when changes in date, chlorophyll content and air temperature were accounted for. Itwould be interesting to examine further the effect of photoperiod on photosynthetic rates duringautumn by incorporating total irradiance hours available for photosynthesis as an additionalmeasure ofphotosynthetic rates at a given moment in the day. The increased hours available tophotosynthesize is not accounted for by this model but may be important in evaluating the effectof photoperiod on assimilation rates during autumn.Finally, it was also possible to determine the effect of soil temperature and photoperiodon the relationship between chlorophyll and photosynthesis because chlorophyll was one of thevariables used in the regression model. To compare the values estimated for the effect of a 1.0mg loss in chlorophyll on photosynthetic rates during autumn, I conducted an additionalexperiment in which I obtained an estimate of the effect of chlorophyll loss on photosyntheticrates from seedlings senescing in environmentally controlled chambers. The values in bothautumn and growth chamber regressions were the same. These results suggest that, oncesenescence has begun, the rate of change in photosynthesis per unit chlorophyll is constant.128CHAPTER 7: CONCLUSIONSFoliar senescence is an intricate process: its onset can begin at almost any developmentalstage of the leaf, and it can be triggered by exogenous and endogenous stimuli. Furthermore,once senescence has been initiated, the sequence of events leading to necrosis can vary. Theprocess occurring during autumn reflects the complexities found during senescence because manyenvironmental and leaf-specific conditions interact to determine the timing and sequence of eventsduring autumn color change. It will be increasingly important to have a wide range of techniquesto study senescence as global demands on soils increase and plants are cultivated under additionalstresses.While researchers have described causative conditions and biochemical processes ofenvironmentally induced senescence, few studies have identified the environmentally sensitivebiochemical pathways in the leaf, or the magnitude of changes needed to elicit the onset ofsenescence. This dissertation applies statistical methods to document the effects of theenvironment on the leaf With these tools I quantified the relative importance of environmentalconditions during autumn senescence as they affected both the onset and process of senescence.Although I focused on the effects of temperature, photoperiod, and leaf age, the methodsdeveloped here can also be used to examine the effects of other conditions on the leaf such asdrought, light levels, ozone, or any combination of stress conditions.I have contributed several important insights about the orderly decline in function duringleaf senescence. Although the sequence of degradative events appears to be similar for differentaged leaves, older leaves respond more quickly to senescence cues. One implication of theseconclusions is that the effect of autumn stresses on deciduous foliage should be predictable.129Indeed, I developed and executed a methodology that enabled me to predict the timing of foliarsenescence as induced by environmental changes. In addition, I showed that it is possible topredict the sequence of events during senescence in western larch.A second broad conclusion from my work is that environmentally-induced senescencemay be best characterized as a series of responses to individual environmental factors rather thana pathway induced by senescence genes. In other words, there may be a set of genes that respondto temperature changes while a different set respond to photoperiod changes. The result is aseries of interacting signals that yield a specific pattern of senescence. This idea is different fromone in which changes in temperature and photoperiod trigger senescence genes, which then inturn triggers a set of predetermined responses. While I did not examine the induction of genesduring autumn senescence, I did record that different processes in the leaf are sensitive todifferent autumn conditions, and that the senescence process was different depending on theenvironmental conditions during autumn.Finally, a third major conclusion is that the effect of the environment on senescence issensitive to the magnitude of the environmental change. For example, my data showed that inwestern larch, the rate of photosynthetic decline during autumn was not influenced by a 3°Cabove-ambient soil temperature but an 8°C above-ambient soil temperature did increasephotosynthetic rates. Thus conclusions about the effect of an environmental condition onsenescence depend on the degree and rate of environmental change relative to the growingcondition ofthe plant. The sensitivity of senescence to small environmental changes may partlyexplain the previously recorded differences in the effects of an environmental condition on theprocesses of senescence for different species and different growing conditions.130The results presented in this dissertation fill several important gaps in the literature on theeffect of the environment on senescence, though there are remaining challenges. A study onsenescent larch mesophyll cells has been done (Cunninghame et al. 1982), but it would be usefulto correlate the ultrastructure changes to physiological data. I used one year old seedlings whichwere grown together under non-stressed conditions to avoid variation caused by growingconditions prior to senescence. If saplings or mature trees are examined over a period of severalyears, the cumulative stresses on the tree may alter the response of the leaves in a given year toautumn weather. An example of this complexity has been noted in pine where the combinationof ozone and drought may have an immediate impact on the tree as a whole, and several seasonsmay pass before premature leaf senescence occurs (Chen et al. 1994, Goldstein and Ferson 1994).This delay may in part be due to the fact that pine holds its needles for several years. Thetechniques described in this thesis can be used to follow these types of interactions.The interaction between the leafand the tree during autumn is important in understandingthe response of the leaf to autumn weather. For example, changes in the seedling due to frosthardiness may influence the response of leaves to environmental changes. In larch the degree offrost hardiness is difficult to quantif’ because common measures of frost hardiness in evergreensanalyze needles during autumn, which are of course senescing in larch. Larch seedlings also havean interesting growth habit because the apical shoot may begin to elongate (lammas growth)during a warm autumn. The effect of this growth on foliar senescence is not known.The value of studying leaf senescence has been previously described (Smart 1994, Hendryet al. 1987) and can be summarized by noting the importance of plants in our economy assuppliers of materials, food, energy, and medicines. While plants continue to show remarkable131capacities for tolerance to non-ideal conditions, the limits to their tolerance are becoming moreevident as soil and air-borne pollutant levels increase and as species are planted in greatergeographic ranges. The research in this thesis provides new methods to quantifj the effect of theenvironment on plants in order to make adequate predictions about the effects of changingenvironmental conditions on the health of plant communities.132CHAPTER 8: REFERENCESAbadia A., Poe A., Abadia J. (1991) Could iron nutrition status be evaluated throughphotosynthetic pigment changes? Journal of Plant Nutrition 14: 987-999.Adams P. (1991) Effect of diurnal fluctuations in humidity on the accumulation of nutrients inthe leaves of tomato (Lycopersicon esculentum). Journal ofHorticultural Science 66:545- 550.Adams W.W. ifi, Winter K., Schreiber U., Schramel P. (1990) Photosynthesis and chlorophyllfluorescence characteristics in relationship to changes in pigment and elementcomposition of leaves of Platanus occidentalis L. during autumnal senescence. PlantPhysiology 93: 1184-1190.Addicott F.T. & Lyon J.L. (1973) Physiological ecology of abscission. In Shedding of PlantParts (ed. T.T. Kozlowski), pp. 85-117. Academic Press, San Diego, CA.Aronson J., Kigel J., Shmida A., Klein 3. (1992) Adaptive phenology of desert andMediterranean populations of annual plants grown with and without water stress.Oecologia 89: 17-26.Azumi Y. & Watanabe A. (1991) Evidence for a senescence-associated gene induced bydarkness. Plant Physiology 95: 577-583.Ballach H., Mooi J., Wittig R. (1992) Premature ageing in Populus nigra L. after exposure toair pollutants. Angewandte Botanik 66: 14-20.Becker W. & Apel K. (1993) Differences in gene expression between natural and artificiallyinduced leaf senescence. Planta 189: 74-79.BegamH.H. & ChoudhuriM.A. (1992) H20metabolism during senescence of two submergedangiosperms Hydrilla and Ottelia: changes in enzyme activities in light and darkness.Biochemische Physiologie Pflanzen 188: 105-115.Begonia G.B., Hesketh J.D., Frederick J.R., Finke Ri., Pettigrew W.R. (1987) Factors affectingleaf duration in soybean and maize. Photosynthetica 21: 285-295.Ben-David H., Nelson N., Gepstein S. (1983) Differential changes in the amount of proteincomplexes in the chloroplast membrane during senescence of oat and bean leaves. PlantPhysiology 73: 501-5 10.Biswal U.C. & Biswal B. (1984) Photocontrol ofleaf senescence. Photochemistry Photobiology39: 875-879.133Blank A. & McKeon T. (1991) Expression of three RNase activities during natural and dark-induced senescence ofwheat leaves. Plant Physiology 97: 1409-1413.Bortlik K., Gut H., Matile P. (1987) Yellowing and non-yellowing trees: a comparison ofprotein- and chlorophyll-loss in senescent leaves. Botanica Helvetica 97: 323-328.Bowes G. (1993) Facing the inevitable: Plants and increasing atmospheric CO2. Annual ReviewofPlantPhysiology PlantMolecular Biology 44: 309-332.Bowler C. & Chua N.H. (1994) Emergent themes of plant signal transduction. The Plant Cell6: 1528-1541.Bowler C., Montagu M.V., Jnzé D. (1992) Superoxide dismutase and stress tolerance. AnnualReview Plant Physiology PlantMolecular Biology 43: 83-116.Brown J.H., Paliyath G., Thompson J.E. (1991) Physiological mechanisms of plant senescence.In Plant Physiology: A Treatise. Vol X: Growth and Development (ed. F.C. Steward),pp. 227-275. Academic Press, NY.Burkey K.O. & Wells R. (1991) Response of soybean photosynthesis and chloroplast membranefunction to canopy development and mutual shading. Plant Physiology 97: 245-252.Cakmak I. & Marschner H. (1992) Magnesium deficiency and high light intensity enhanceactivities of superoxide dismutase, ascorbate peroxidase, and glutathione reductase inbean leaves. Plant Physiology 98: 1222-1227.Camp P.J., Huber S.C., Burke J.J., Moreland D.E. (1982) Biochemical changes that occur duringsenescence of wheat leaves. Plant Physiology 70: 164 1-1646.Carlyle J.C. & Malcom D.C. (1985) Larch litter and nitrogen availability in mixed larch-sprucestands. Canadian Journal ofForestry Research 16: 327-3 29.Casano L.M., Martin M., Sabater B. (1994) Sensitivity of superoxide dismutase transcript levelsand activities to oxidative stress is lower in mature-senescent than in young barley leaves.Plant Physiology 106: 1033-1039.Chang C.W. (1975) Carbon dioxide and senescence in cotton plants. Plant Physiology 55: 515-519.Chen C.T. & Kao C.H. (1992) Senescence of rice leaves. XXXII. Effects of abscisic acid andbenzyladenine on polyamines and ethylene production during senescence. Journal ofPlant Physiology 139: 6 17-620.134Chen C.W., Tsai W.T., Gomez L.E. (1994) Modeling responses of ponderosa pine to interactingstresses of ozone and drought. Forest Science 40: 267-288.Chéor F., Arul J., MakhloufJ., Willemot C. (1992) Delay of membrane lipid degradation bycalcium treatment during cabbage leaf senescence. Plant Physiology 100: 1656-1660.Chevone B.I., Seiler J.R., Melkonian J., Amundson R.G. (1990) Ozone-water stress interactions.In Stress Responses in Plants: Adaptation and acclimation mechanisms. (eds. R.G.Alscher & J.R. Cumming), pp. 311-328. Wiley-Liss Inc., NY.ChiaL.S., Mayfield C.I., Thompson J.E. (1984) Simulated acid rain induces lipid peroxidationand membrane damage in foliage. Plant, Cell andEnvironment 7: 333-33 8.Chou C.H. & Kao C.H. (1992) Methyl jasmonate, calcium, and leaf senescence in rice. PlantPhysiology 99: 1693-1694.Cincerova A., Novotna D., Dvorak M. (1991) NADH- and ferredoxin-dependent glutamatesynthase in the life span ofthe second leaf ofwheat plant under conditions of senescenceinduced by nitrogen deficiency and natural senescence. Biologia Plantarum (Prague) 33:3 17-324.Constable G.A. & Rawson H.M. (1980) Effect of leaf position, expansion and age onphotosynthesis, transpiration and water use efficiency of cotton. Australian Journal ofPlant Physiology 7: 89-100.Cortes P.M. & Sinclair T.R. (1986) Gas exchange of field-grown soybean under drought.Agronomy Journal 78: 454-458.Crafts-Brandner S.J. (1992) Phosphorus nutrition influence on leaf senescence in soybean. PlantPhysiology 98: 1128-1132.Crafts-Brandner S.J., Salvucci M.E., Egli D.B. (1990) Changes in ribulose bisphosphatecarboxylase/oxygenase and ribulose 5-phosphate kinase abundances and photosyntheticcapacity during leaf senescence. Photosynthesis Research 23: 223-230.Creasy L.L. (1974) Sequence of development of autumn coloration in Euonymus.Phytochemistry 13: 1391-1394.Cuello J. & Lahora A. (1993) Chlorophyll degradation by free radicals derived from linolenicacid in incubated barley thylakoids. Acta Botanica Neerlandica 42: 481-490.135Cunninghame M.E., ITlilman J.R., Bowes B.G. (1982) Ultrastructural changes in mesophyll cellsof Larix decidua x kaempferi during leafmaturation and senescence. Flora 172: 161-172.Dafling M.J. & Nettleton kM. (1986) Chloroplast senescence and proteolytic enzymes. In Plantproteolytic enzymes, Vol II (ed. M.J. Dalling), pp. 125-153. CRC Press, Boca Raton,FL.Danall N.M. (1989) The effect ofair pollutants on physiological processes in plants. Plant, Celland Environment 12: 1-30.Davies K.M. & Grierson D. (1989) Identification of cDNA clones for tomato (Lycopersiconesculentum IVIill.) mRNAs that accumulate during fruit ripening and leaf senescence inresponse to ethylene. Planta 179: 73-80.DeanM.A., Letner C.A., Eley J.H. (1993) Effect of autumn foliar senescence on chlorophyll a:bratio and respiratory enzymes of Populus tremuloides. Bulletin of the Torrey BotanicalClub 120: 269-274.Dhindsa R.S., Plumb-Dhindsa P., Thorpe T.A. (1981) Leaf senescence: correlated withincreased levels of membrane permeability and lipid peroxidation, and decreased levelsof superoxide dismutase and catalase. Journal ofExperimental Botany 32: 93-101.Diaz S., Grime J.P., Harris J., McPhersonE. (1993) Evidence of a feedback mechanism limitingplant response to elevated carbon dioxide. Nature 364: 616-6 17.Dreier W. (1990) Senescence inhibition and promotion in isolated primary leaves from oat bythe calcium-calmodulin complex. Biochemie Physiologie Pflanzen 186: 31-36.Edwards G.E. & Baker N.R. (1993) Can CO2 assimilation in maize leaves be predictedaccurately from chlorophyll fluorescence analysis? Photosynthesis Research 37: 89-102.Everett R.L. & Thran D.F. (1992) Nutrient dynamics in singleleafpinyon (Pinus monophyllaTorn and Frem.) needles. Tree Physiology 10: 59-68.Feller U. 1987. Proteolytic enzymes in relation to leaf senescence. In Plant proteolytic enzymes,Vol II (ed. M.J. Dalling), pp. 49-68. CRC Press, Boca Raton, FL.Feller U. & Fischer A. (1994) Nitrogen metabolism in senescing leaves. Critical Reviews inPlant Sciences 13: 241-273.Ferguson I.B., Watkins C.B., Harman J.E. 1983. Inhibition by calcium of senescence ofdetached cucumber cotyledons. Plant Physiology 71: 182-186.136Ferreira R.M. & Teixeira A.R. (1992) Sulphur starvation in Lemna leads to degradation ofribulose-bisphosphate carboxylase without plant death. Journal ofBiological Chemistry267: 7253-7257.Fischer A. & Feller U. (1994) Senescence and protein degradation in leaf segments of youngwinter wheat: influence of leaf age. Journal ofExperimental Botany 45: 103-109.Garcia-Ferris C. & Moreno J. (1993) Redox regulation of enzymatic activity and proteolyticsusceptibility ofribulose-1,5-bisphosphate carboxylase/oxygenase from Euglena gracilis.Photosynthesis Research 34: 5 5-66.Garner W.W. & Allen H.A. (1923) Further studies in photoperiodism, the response of the plantto relative length of day and night. Journal ofAgriculturalResearch 23: 87 1-920.Genty B., Briantais J., Baker N.R. (1989) The relationship between the quantum yield ofphotosynthetic electron transport and quenching ofchlorophyll fluorescence. BiochimicaetBiophysicaActa 990: 87-92.Gevorkiantz S.R. & Roe El. (1935) Photoperiodism in forestry. Journal ofForestry 33: 599-602.Goldstein R. & Ferson S. (1994) Responses of plants to interacting stresses (ROPIS): Programrationale, design, and implications. Journal ofEnvironmental Quality 23: 407-411.Gonzalez R., Segura A., Gonzalez M. (1993) Ultrastructure of chioroplasts of pine needlesexposed to an industrial environment. Biologia Plantarum (Prague) 35: 407-416.Goodwin T.W. (1958) Studies in carotenogenesis. The changes in carotenoid and chlorophyllpigments in the leaves of deciduous trees during autumn necrosis. BiochemicalJournal68: 503-511.Gower S.T. & Richards J.H. (1990) Larches: deciduous conifers in an evergreen world.BioScience 40: 818-826.Gower S.T., Grier C.C., VogtD.J., Vogt K.A. (1987) Allometric relations of deciduous (Larixoccidentalis) and evergreen conifers (Pius contorta and Pseudotsuga menziesii> of theCascade mountains in central Washington. Canadian Journal ofForestry Research 17:630-634.Gowin T., Lourtioux A., Mousseau M. (1980) Influence of constant growth temperature uponthe productivity and gas exchange of seedlings of Scots pine and European larch. ForestScience 26: 30 1-309.137Gratiani L., Fiorentino E., Camiz S., Fida C. (1987) Relationships among chlorophyll contentsand some ecophysiological parameters with relation to their distribution within the beechcrown. Photosynthetica 21: 333-340.Gray J.C. (1982) Use of proteolytic inhibitors during isolation of plastid proteins. In Methodsin chioroplast molecular biology (eds. M. Edelman, RB. Haflick, N.-H. Chua), pp. 1093-1102. Elsevier Biomedical Press, Amsterdam.Guiamet J.J., Willemoes J.G., Montaldi E.R. (1989) Modulation of progressive leaf senescenceby the red: far-red ratio of incident light. Botanical Gazette 150: 148-151.Guillemaut P., Weber-Lofti F., Blache D., Prost M., Rether B., Dietrich A. (1992) Coniferdecline in the north-east of France: Characteristic changes in chloroplast protein patternand absence of anti-oxidative defense capability point to an involvement of ozone.Physiologia Plantarum 85: 215-222.Gunderson C.A., Norby R.J., Wullschleger S.D. (1993) Foliar gas exchange responses of twodeciduous hardwoods during 3 years of growth in elevated C02: no loss ofphotosynthetic enhancement. Plant, Cell andEnvironment 16: 797-807.Hasler R. (1985) CO2 uptake in european larch (Larix decidua Mill.) near treeline inSwitzerland. EidgenOssische Anstattfür dasforstliche Versuchswesen, Berichte 270:113-122.Hendry GA., Houghton J.D., Brown S.B. (1987) The degradation of chlorophyll- a biologicalenigma. New Phytologist 107: 255-3 02.HenselL.L., Grbië V., BaumgartenD.A., Bleecker A.B. (1993) Developmental and age-relatedprocesses that influence the longevity and senescence of photosynthetic tissues inArabidopsis. Plant Cell5: 553-564.Hidema 3., Makino A., Mae T., Ojima K. (1991) Photosynthetic characteristics of rice leavesaged under different irradiances from full expansion through senescence. PlantPhysiology 97: 1287-1293.Higgens S.S., BlackR.A., Radamaker G.K., Bidlake W.R. (1987) Gas exchange characteristicsand water relations of Larix occidentalis. Canadian Journal ofForestry Research 17:1364-1370.Holloway P.J., Maclean D.J., Scoff, K.J. (1983) Rate-limiting steps of electron transport inchloroplasts during ontogeny and senescence ofbarley. Plant Physiology 72: 795-80 1.138Hopkinson J.M. (1966) Studies on the expansion of the leaf surface. VI. Senescence and theusefulness of old leaves. Journal ofExperimentalBotany 17: 762-770.Horton P. & Bowyer J.R. (1990) Chlorophyll fluorescence transients. In Methods in PlantBiochemistry (eds. P.M. Dey & J.B. Harborne), pp. 259-296. Academic Press, NY.Houpis J.L., Surano K.A., Cowles S., Shinn J.H. (1988) Chlorophyll and carotenoidconcentrations in two varieties of Pinus ponderosa seedlings subjected to long-termelevated carbon dioxide. Tree Physiology 4: 187-193.Huang Y. & Kao C.H. (1992) Calcium in the regulation of corn leaf senescence by light.Botanical Bulletin Academie Sinicia 33: 161-165.Huffaker RC. (1990) Tansley Review No. 25. Proteolytic activity during senescence of plants.New Phytologist 116: 199-231.Irigoyen J.J., Emerich D.W., Sanchez-DIaz M. (1992) Alfalfa leaf senescence induced bydrought stress: photosynthesis, hydrogen peroxide metabolism, lipid peroxidation andethylene evolution. Physiologia Plantarum 84: 67-72.Jackson M.B. (1993) Are plant hormones involved in root to shoot communication. Advancesin Botanical Research 19: 104-187.Jacob J. & Lawlor D.W. (1993) In vivo photosynthetic electron transport does not limitphotosynthetic capacity in phosphate-deficient sunflower and maize leaves. Plant, Celland Environment 16: 785-795.Jenkins G.I., Baker N.R., Bradbury M., Woolhouse H.W. (1981) Photosynthetic electrontransport during senescence of the plant primary leaves of Phaseolus vulgaris. JournalofExperimental Botany 32: 999-1008.Jiang C., Rodermel S.R., Shibles R.M. (1993) Photosynthesis, rubisco activity and amount, andtheir regulation by transcription in senescing soybean leaves. Plant Physiology 101: 105-112.Johnson-Flanagan A.M., McLachlan G. (1990) Peroxidase-mediated chlorophyll bleaching indegreening canola (Brassica napus cv Westar) seeds and its inhibition by sublethalfreezing. Physiologia Plantarum 80: 1-7.Jones H.G. (1983) Plants and microclimate. pp. 212-237. Cambridge Univ Press, NY.Jordan W.R., Morgan P.W., Davenport T.L. (1972) Water stress enhances ethylene-mediatedleaf abscission in cotton. Plant Physiology 50: 756-758.139Judge 0.0., Hill R.C., Griffiths W.E., Lutkepohl H., Lee T. (1988) Introduction to the theoryand practice of econometrics. pp. 786-801. John Wiley & Sons, NY.Kajimoto T. (1993) Shoot dynamics of Pinuspumila in relation to altitudinal and wind exposuregradients on the Kiso mountain range, central Japan. Tree Physiology 13: 4 1-53.Kar M., Streb P., Hertwig B., Feierabend J. (1993) Sensitivity to photodamage increases duringsenescence in excised leaves. Journal ofPlant Physiology 141: 53 8-544.Katajima M. & Butler W.L. (1975) Quenching of chlorophyll fluorescence and primaryphotochemistry in chioroplasts by dibromothymoquinone. Biochimica et Biophysica Acta376: 105-115.Kelly D.W., Hicklenton P.R., Reekie E.G. (1991) Photosynthetic response of geranium toelevated CO2 as affected by leaf age and time of CO2 exposure. Canadian Journal ofBotany 69: 2482-2488.Kelly M.O. & Davies P.J. (1988) The control of whole plant senescence. Critical Review ofPlant Science (Cornell) 7: 139-73.Khanna-Chopra R. & Sinha S.K. (1988) Enhancement of drought-induced senescence by thereproductive sink in fertile lines of wheat and sorghum. Annals ofBotany 61: 649-653.Kikuzawa K. (1988) Leaf survivals of tree species in deciduous broad-leaved forests. PlantSpecies andBiology 3: 67-76.Koike T. (1987) Photosynthesis and expansion in leaves of early, mid, and late successional treespecies, birch, ash, and maple. Photo.synthetica 21: 503-508.Kooten 0. & Snel J.F. (1990) The use of chlorophyll fluorescence nomenclature in plant stressphysiology. Photosynthesis Research 25: 147-150.KOstner B., Czygan F-C., Lange O.L. (1990) An analysis of needle yellowing in healthy andchiorotic Norway spruce (Picea abies) in a forest decline area of the Fichtelgebirge (NEBavaria): 1. Annual time-course changes in chioroplast pigments for five different needleage classes. Trees 4: 55-67.Kozlowski T.T. (1991) Effects of environmental stresses on deciduous trees. In Response ofplants to multiple stresses (eds. H.A. Mooney, W.E. Winner, E.J. Pell), pp. 391-411.Academic Press, San Diego, CA.140Kozlowski T.T., KramerP.J., Pallardy S.G. (1991) Extent and significance of shedding of plantparts. In The physiological ecology of woody plants (ed. T.T. Kozlowski), pp. 1-34.Academic Press, NY.Kramer P.3. (1936) Effect ofvariation in length of day on growth and dormancy of trees. PlantPhysiology 11: 127-137.Kura-Hotta M., Hashimoto H., Satoh K., Katoh S. (1990) Quantitative determination of changesin the number and size of ehioroplasts in naturally senescing leaves of rice seedlings.Plant and Cell Physiology 31: 33-3 8.Laemmli U.K. (1970) Cleavage of structural proteins during the assembly of the head ofbacteriophage T4. Nature 227: 680-685.Landry L.G. & Pell E.J. (1993) Modification of rubisco and altered proteolytic activity in 03-stressed hybrid poplar (Populus maximowizii x trichocarpa). Plant Physiology 101:1355-1362.Lang M. & Lichtenthaler H.K. (1991) Changes in the blue-green and red fluorescence-emissionspectra ofbeech leaves during the autumnal chlorophyll breakdown. Journal ofPlantPhysiology 138: 550-553.Langmeier M., Ginsburg S., Matile P. (1993) Chlorophyll breakdown in senescent leaves:demonstration ofMg-dechelatase activity. Physiologia Plantarum 89: 347-353.Larcher W. (1987) Stress bei Pflanzen. ZeitschrftNaturwissenschaft 74: 15 8-167.Larcher W. & Neuner G. (1989) Cold induced sudden reversible lowering of in vivo chlorophyllfluorescence after saturating light pulses. PlantPhysiology 89: 740-742.Laurière C. (1983) Enzymes and leaf senescence. Physiologie Vegetales 21: 1159-1177.Lawrence D.K. & Mayne R.G. (1991) Plant growth regulators and photosynthesis. InHerbicides (eds. N.R. Baker & M.P. Percival), pp. 300-33 5. Elsevier Science Publishers,NY.Ledig F.T. & Botkin D.B. (1974) Photosynthetic CO2 -uptake and the distribution ofphotosynthate as related to growth oflarch and sycamore progenies. Silvae Genetica 23:188-192.Leshem Y.Y. (1987) Membrane phospholipid catabolism and Ca2 activity in control ofsenescence. Physiologia Plantarum 69: 551-559.141Lichtenthaler H.K. (1987) Chlorophyll fluorescence signatures of leaves during the autumnalchlorophyll breakdown. Journal ofPlant Physiology 131: 101-110.Lichtenthaler H.K. & Welibum A.R. (1983) Determination of total carotenoids and chiorophyllsa and b ofleafextracts in different solvents. Biochemical Society Transactions 11: 591-592.Logan K.T. (1966) Growth oftree seedlings as affected by light intensity. Canadian DepartmentofForestry Publication No. 1160. Ottawa, Canada.Lucas P.W., Rantanen L., Mehihorn H. (1993) Needle chiorosis in Sitka spruce following athree-year exposure to low concentrations of ozone: changes in mineral content,pigmentation and ascorbic acid. New Phytologist 124: 265-275.LUtz C., Steiger A., Godde D. (1992) Influence of air pollutants and nutrient deficiency on D-1protein content and photosynthesis in young spruce trees. Physiologia Plantarum 85:611-617.Mae T., Kamei C., Funaki K., Miyadai K., Makino A., Ohira K., Ojima K. (1989) Degradationof ribulose-1,5-bisphosphate carboxylase/oxygenase in the lysates of the chioroplastsisolated mechanically from wheat (Triticum aestivum L.) leaves. Plant Cell Physiology30: 193-200.Majumdar S., Ghosh S., Glick B.R., Dumbroff E.B. (1991) Activities of chiorophyllase,phosphoenolpyruvate carboxylase and ribulose-1,5 bisphosphate carboxylase in theprimary leaves of soybean during senescence and drought. Physiologia Plantarum 81:473-480.Makino A., Mae T., Ohira K. (1984) Changes in photosynthetic capacity in rice leaves fromemergence through senescence. Analysis from ribulose 1,5 bisphosphate carboxylase andleaf conductance. Plant Cell Physiology 25: 511-521.Markgraf T. & Berry J.A. (1990) Measurement of photochemical and non-photochemicalquenching: correction for turnover ofPS2 during steady-state photosynthesis. CurrentResearch in Photosynthesis 4: 279-282.Martin M. & Sabater B. (1989) Translational control of chioroplast protein synthesis duringsenescence ofbarley leaves. Physiologia Plantarum 75: 374-381.Martinoia E., Heck U., Dalling M.J., Matile P. (1983) Changes in chloroplast number andchioroplast constituents in senescing barley leaves. Biochemische Physiologie Pflanzen78: 147-155.142Matyssek R. (1986) Carbon, water and nitrogen relations in evergreen and deciduous conifers.Tree Physiology 2: 177-187.Matyssek R., Gunthardt-Goerg M. S., Keller T., Scheidegger C. (1991) Impairment of gasexchange and structure in birch leaves (Betula pendula) caused by low ozoneconcentrations. Trees 5: 5-13.Matzke E.B. (1936) The effect of street lights in delaying leaf-fall in certain trees. AmericanJournal ofBotany 23: 446-452.Maunders M.J. & Brown S.B. (1983) The effect of light on chlorophyll loss in senescing leavesof sycamore (Acerpseudoplatanus L.). Planta 158: 309-311.Mehta R.A., Fawcett T.W., Porath D., Mattoo A.K. (1992) Oxidative stress causes rapidmembrane translocation and in vivo degradation of ribulose-1,5 bisphosphatecarboxylase/oxygenase. Journal ofBiological Chemistiy 267: 2810-2816.Meijer G. & van der Veen R. (1957) Wavelength dependence on photoperiodic responses. ActaBotanica Neerlandica 6: 429-433.Millard P. & Thomson C.M. (1989) The effect of the autumn senescence of leaves on theinternal cycling ofnitrogen for the spring growth of apple trees. Journal ofExperimentalBotany 40: 1285-1289.Minorsky P.V. (1989) Temperature sensing by plants: a review and hypothesis. Plant, Cell andEnvironment 12: 119-135.Molisch H. (1918) Uber die vergilbung der Blätter. Sitzungsbericht Akademische Wissenshaft,Wien, Naturwissenschafthiche Klasse 127: 3-34.Morgan P.W. (1990) Effects ofabiotic stresses on plant hormone systems. In Stress Responsesin Plants: Adaptation and acclimation mechanisms (eds. R.G. Alscher & J.R. Cumming),pp. 113-146. Wiley Liss Inc., NY.Nelson N.D., Dickmann D.I., Gottschalk K.W. (1982) Autumnal photosynthesis in short-rotation intensively cultured Populus clones. Photosynthetica 16: 321-333.Neter J., Wasserman W., Kutner M.H. (1990) Applied linear statistical models. pp. 225-314.3rd ed. Irwin, Homewood, IL.Nilsen E.T. & Bao Y. (1987) The influence of leaf age, season, and microclimate on thephotosynthetic apparatus of Rhododendron maximum 1. Chlorophylls. Photosynthetica21: 535-542.143NockL.P., Rogers L.J., Thomas H. (1992) Metabolism ofprotein and chlorophyll in leaf tissueofFestucapratensis during chioroplast assembly and senescence. Phytochemistry 31:1465-1470.Oberbauer S.F., StrainB.R., FetcherN. (1985) Effect ofC02-enrichment on seedling physiologyand growth of two tropical tree species. Physiologia Plantarum 65: 352-356.Okada K., Inoue Y., Satoh K., Katoh 5. (1992) Effects of light on degradation of chlorophylland proteins during senescence ofdetached rice leaves. Plant Cell Physiology 33: 1183-1191.Olmsted C.E. (195 1) Experiments on photoperiodism, dormancy and leaf age and abscission insugar maple. Botanical Gazette 112: 365-393.Oquist G. & Chow W.S. (1992) On the relationship between quantum yield ofPhotosystem IIelectron transport, as determined by chlorophyll fluorescence and the quantum yield ofC02-dependent 02 evolution. Photosynthesis Re.earch 33: 51-62.Oquist G. & Huner N.P. (1991) Effects of cold acclimation on the susceptibility ofphotosynthesis to photoinhibition in Scots pine and in winter and spring cereals: afluorescence analysis. Functional Ecology 5: 91-100.Oquist G. & Martin B. (1986) Cold Climates. In Photosynthesis in contrasting environments(eds. N.R. Baker & S.P. Long), pp. 238-293. Elsevier Science Publishers B.V.Oquist G., Martensson 0., Martin B., Malmberg 0. (1978) Seasonal effects on chlorophyllprotein complexes isolated from Pinus silvestris. Physiologia Plantarum 44: 187-192.Osborne D.J. (1973) Internal factors regulating abscission. In Shedding ofplant parts (ed. T.T.Kozlowski), pp. 125-147. Academic Press, NY.Ottander C., Hundal T., Anderson B., Honer N.C., Oquist G. (1993) Low-temperature effectson photosynthesis and correlation with freezing tolerance in spring and winter cultivarsofwheat and rye. Plant Physiology 101: 245-250.Paques L.E. (1989) A critical review of larch hybridization and its incidence on breedingstrategies. Annals Science (Paris) 46: 14 1-154.Pell E.J. & Dann M. S. (1991) Multiple stress-induced foliar senescence and implications forwhole plant longevity. In Response of plants to multiple stresses (eds. H.A. Mooney,W.E. Winner, E.J. PcI!), pp. 189-204. Academic Press, San Diego, CA.144Philosoph-Hadas S., Meir S., Aharoni N. (1991) Effect ofwounding on ethylene biosynthesisand senescence of detached spinach leaves. Physiologia Plantarum 83: 241-246.Pistelli L., Perata P., Alpi A. (1993) Effect of leaf senescence on glyoxylate cycle enzymeactivities. Australian Journal ofPlant Physiology 19: 723-729.Polle A., Pfirrmann T., Chakrabarti S., Rennenberg H. (1993) The effects of enhanced ozoneand enhanced carbon dioxide concentrations on biomass, pigments and antioxidativeenzymes in spruce needles (Picea abies L.). Plant, Cell andEnvironment 16: 311-316.Reddy G.N., Arteca R.N., Dai Y-R., Flores H.E., Negm F.B., Pell E.J. (1993) Changes inethylene and polyamines in relation to mRNA levels of the large and small subunits ofribulose bisphosphate carboxylase/oxygenase in ozone-stressed potato foliage. Plant,Cell andEnvironment 16: 8 19-826.Rehfeldt G.E. (1982) Differentiation of Larix occidentalis populations from the northern RockyMountains. Silvae Genetica 31: 13-19.Reid D.M., BeaU F.D., Pharis R.P. (1991) Environmental cues in plant growth and development.In Plant Physiology: A Treatise. Vol X: Growth and Development (ed. F.C. Steward),pp. 65-181. Academic Press, NY.Reviron M., Vartanian N., SalLantin M., Heut J., Pernollet J., Vienne D. (1992) Characterizationofa novel protein induced by progressive or rapid drought and salinity in Brassica napusleaves. Plant Physiology 100: 1486-1493.Reynolds B., Cape J.N., Paterson I.S. (1989) A comparison of element fluxes in throughfallbeneath larch and sitka spruce at two contrasting sites in the United Kingdom. Forestry62: 29-39.Richards J.H. & Bliss L.C. (1986) Winter water relations of a deciduous timberline conifer,Larix lyallii Pan. Oecologia 69: 16-24.Ronning C.M., Bouwkamp J.C., Solomos T. (1991) Observations on the senescence of a mutantnon-yellowing genotype of Phaseolus vulgaris L. Journal ofExperimental Botany 42:235-24 1.Rudiger W. & Schoch S. (1989) Abbau von Chlorophyll. Naturwissenschaften 76: 453-457.Rutland J. (1888) The fall of the leaf. Royal Society ofNew Zealand, Transactions 21: 110-120.145Salama A.M. & Wareing P.F. (1979) Effects of mineral nutrition on endogenous cytokinins inplants of sunflower (Helianthus annuus L.). Journal ofExperimental Botany 30: 971-981.Salisbury F.B. & Ross C.W. (1985) Plant Physiology. Chapters 5, 11 and 23. WadsworthPublishing Company Inc., Belmont, CA.Sandford A.P. & Jarvis P.G. (1986) Stomatal responses to humidity in selected conifers. TreePhysiology 2: 89-103.Sanger J.E. (1971) Quantitative investigation of leaf pigments from their inception in budsthrough autumn coloration to decomposition in falling leaves. Ecology 52: 1075-1089.Sarjala T. & Kaunisto S. (1993) Needle polyamine concentrations and potassium nutrition inScots pine. Tree Physiology 13: 87-96.Satler S.O. & Thimann K.V. (1983) Metabolism of oat leaves during senescence. VII. Theinteraction of carbon dioxide and other atmospheric gases with light in controllingchlorophyll loss and senescence. Plant Physiology 71: 67-70.Schaper H. & Chacko E.K. (1993) Effect ofirradiance, leaf age, chlorophyll content and branch-girdling on gas exchange of cashew (Anacardium occidentale L.) leaves. Journal ofHorticultural Science 68: 541-550.SchierenbeckK.A. & Marshall J.D. (1993) Seasonal and diurnal patterns of photosynthetic gasexchange for Lonicera sempervirens and L. japonica (Caprifoliaceae). American JournalofBotany 80: 1292-1299.Schmid B., Polasek W., Weiner J., Krause A., Stoll P. (1994) Modeling of discontinuousrelationships in biology with censored regression. American Naturalist 143: 494-5 07.Schmidt W.C, Shearer R.C., Roe A.L. (1976) Ecology and silviculture ofwestern larch forests.pp. 1-5. Forest Service, U. S. Department of Agriculture, Technical Bulletin No. 1520.Schneider P., Horn K., Lauterbach R., Hock B. (1991) Quantitative determination of thegiberellins GA3 and GA1 in spruce (Picea abies [L.]) needles by a highly specific enzymeimmunoassay. Journal ofPlant Physiology 139: 229-234.Schreiber U., Schliwa U., Bilger W. (1986) Continuous recording of photochemical and nonphotochemical chlorophyll fluorescence quenching with a new type of modulationfluorometer. Photosynthesis Research 10: 51-62.146Scott B.J. & Robson A.D. (1991) Distribution of magnesium in wheat (Triticum aestivum L.)in relation to supply. Plant Soil 136: 183-194.Shacklock P.S., Read N.D., Trewavas A.J. (1992) Cytosolic free calcium mediates red light-induced photomorphogenesis. Nature 358: 753-754.Sharpe P.J. & Rykiel E.J. (1991) Modelling integrated response of plants to multiple stresses.In Response of plants to multiple stresses (eds. H.A. Mooney, W.E. Winner, E.J. Pell),pp. 205-224. Academic Press, San Diego CA.Singh S., Letham D.S., Zhang X., Palni L.M.S. (1992) Cytokinin biochemistry in relation to leafsenescence. VI. Effect of nitrogenous nutrients on cytokinin levels and senescence oftobacco leaves. Physiologia Plantarum 84: 262-268.Smart C.M. (1994) Tansley Review No. 64. Gene expression during leaf senescence. NewPhytologist 126: 419-448.Smart C.M., Scofield S.R., Bevan M.V., Dyer T.A. (1991) Delayed leaf senescence in tobaccoplants transformed with tmr, a gene for cytokinin production inAgrobacterium. PlantCell3: 647-656.Smith H. & Whitelam G.C. (1990) Phytochrome, a family of photoreceptors with multiplephysiological roles. Plant, Cell andEnvironment 13: 695-707.STATA (1993) Statistical Data Analysis, Version 3.1. Stata Corporation, College Station, TX.Stoddart J.L & Thomas H. (1982) Leaf senescence. Encyclopedia ofPlant Physiology. NewSeries 14A: 592-636.Taylor A.O. & Craig A.S, (1971) Plants under climatic stress. II. Low temperature, high lighteffects on chioroplast ultrastructure. Plant Physiology 47: 7 19-725.Thomas C., Davis S.D., Tallman G. (1991) Responses of stomata of senescing and nonsenescingleaves of Nicotiana glauca to changes in intercellular concentrations of leaf carbondioxide. Plant, Cell andEnvironment 14: 971-978.Tilton D.L. (1976) Seasonal growth and foliar nutrients of Larix laricina in three wetlandecosystems. Canadian Journal ofBotany 55: 129 1-1298.Tissue D.T., Thomas RB., Strain B.R. (1993) Long-term effects of elevated CO2 and nutrientson photosynthesis and rubisco in loblolly pine seedlings. Plant, Cell andEnvironment16: 859-865.147Tranquihini W., Havranek W.M., Ecker P. (1986) Effects of atmospheric humidity andacclimation temperature on the temperature response of photosynthesis in young Larixdecidua Mill. Tree Physiology 1: 37-45.Tschaplinski T.J. & Blake T.J. (1989) Water relations, photosynthetic capacity, and root/shootpartitioning ofphotosynthate as determinants of productivity in hybrid poplar. CanadianJournal ofBotany 67: 1689-1697.Tucker D.J. (1981) Phytochrome regulation of leaf senescence in cucumber and tomato. PlantScience Letters 23: 103-108.Tyrrell L.E. & Boerner RE. (1987) Larix laricina and Picea mariana: relationships among leafage-span, foliar nutrient patterns, nutrient conservation, and growth efficiency. CanadianJournal ofBotany 65: 1570-1577.Vance N.C. & Running S.W. (1985) Light reduction and moisture stress: effects on growth andwater relations ofwestern larch seedlings. Canadian Journal ofForest Research f5: 72-77.Veirskov B. & Fergeson I.B. (1991) Ubiquitin conjugating activity in leaves and isolatedchloroplasts fromAvena sativa L. during senescence. Journal ofPlant Physiology 138:608-613.Vince-Prue D. (1975) Photoperiodism in plants. p. 339. McGraw Hill, NY.Walker D.A. (1992) Excited leaves. New Phytologist 121: 325-345.Walker D.A. (1987) The use of the oxygen electrode and fluorescence probes in simplemeasurements of photosynthesis. pp. 1-16. Oxygraphics Ltd., University of Sheffield,Kings Lynn, UK.Wang S.Y., Wang C.Y., Weilburn A.R. (1990) Role of ethylene under stress conditions. InStress Responses in Plants: Adaptation and acclimation mechanisms (eds. R.G. Alscher& J.R. Cumming), pp. 147-173. Wiley-Liss Inc., NY.Wareing P.F. (1956) Photoperiodism in woody plants. Annual Review Plant Physiology 7: 191-214.Waring R.H. (1991) Responses of evergreen trees to multiple stresses. In Response of plantsto multiple stresses (eds. H.A. Mooney, W.E. Wmner, E.J. Pehl), pp. 371-390. AcademicPress, San Diego, CA.148Widhoim J.M. & Ogren W.L. (1969) Photorespiratory-induced senescence of plants underconditions of low carbon dioxide. Proceedings NationalAcademy Science USA 63: 668-675.Wieser G. & Havranek W.M. (1993) Ozone uptake in the sun and shade crown of spruce:quantifying the physiological effects of ozone exposure. Trees 7: 227-232.Wiesner J. (1904) Uber den Hitzelaubfall. Berlin Deutsche Botanische Gesellschaft 22: 501-505.Wiesner J. (1871) Untersuchungen Uber die herbstliche Entlaubung der Holzgewachse.Sitzungsbericht Akademische Wissenschaft, Wien, Mathematik NaturwissenschaftlicheKlasse 64: 465-509.Wilimer M., Wilson A.B., Jones H.G. (1988) Changing responses of stomata to abscisic acid andCO2 as leaves and plants age. Journal ofExperimental Botany 39: 401-410.Wiltshire J.J., Wright C.J., UnsworthM.H., Craigon J. (1993) The effects of ozone episodes onautumn leaf fall in apple. New Phytologist 124: 433-43 7.Woolhouse H.W. (1983) Leaf senescence. In The molecular biology of plant development.(eds. H. Smith & D. Grierson), pp. 256-281. Blackwell Science Publishers, Oxford.Worrall J. (1993) Temperature effects on bud-burst and leaf-fall in subalpine larch. Journal ofSustainable Forestry 1: 1-18.WulffR.D. & Strain B.R. (1982) Effects of CO2 enrichment on growth and photosynthesis inDesmodiumpaniculatum. Canadian Journal ofBotany 60: 1084-109 1.Wullschleger S.D., Norby R.J., Hendrix D.L. (1992) Carbon exchange rates, chlorophyllcontent, and carbohydrate status of two forest tree species exposed to carbon dioxideenrichment. Tree Physiology 10: 21-3 1.Yordanov I. (1993) Response of photosynthetic apparatus to temperature stress and molecularmechanisms of its adaptations. Photosynthetica 26: 517-531.Young A. & Britton 0. (1990) Carotenoids and Stress. In Stress Responses in Plants:Adaptation and acclimation mechanisms (eds. R.G. Alscher & J.R. Cumming), pp. 87-112. Wiley Liss Inc., NY.Zar J.H. (1984) Biostatistical analysis. pp. 346-347. Prentice Hall, Englewood Cliffs, NJ.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items